From fitzsim at redhat.com Sun May 1 00:25:04 2005 From: fitzsim at redhat.com (Thomas Fitzsimmons) Date: Sat, 30 Apr 2005 20:25:04 -0400 Subject: rawhide report: 20050430 changes In-Reply-To: <1114877909.24014.84.camel@localhost.localdomain> References: <200504301544.j3UFiMTg030812@porkchop.devel.redhat.com> <1114877909.24014.84.camel@localhost.localdomain> Message-ID: <1114907104.21901.31.camel@tortoise.toronto.redhat.com> On Sat, 2005-04-30 at 17:18 +0100, David Woodhouse wrote: > On Sat, 2005-04-30 at 11:44 -0400, Build System wrote: > > gcc-4.0.0-2 > > ----------- > > - https Handler.java from GNU classpath (Thomas Fitzsimmons, #155466) > > Does this mean gcjwebplugin can do https now? > Yes, gcjwebplugin should be able to access https sites now. Tom From antonio at apache.org Sun May 1 00:50:27 2005 From: antonio at apache.org (Antonio Gallardo) Date: Sat, 30 Apr 2005 19:50:27 -0500 (CDT) Subject: Plase update tzdata to 2005i In-Reply-To: <1114828507.7966.0.camel@ignacio.ignacio.lan> References: <42078.165.98.153.184.1114827118.squirrel@www.agssa.net> <1114828507.7966.0.camel@ignacio.ignacio.lan> Message-ID: <34471.165.98.153.184.1114908627.squirrel@www.agssa.net> On Vie, 29 de Abril de 2005, 21:35, Ignacio Vazquez-Abrams dijo: > On Fri, 2005-04-29 at 21:11 -0500, Antonio Gallardo wrote: >> Is posible to see a patch for FC3 and include it in the next FC4 >> release? > > Request it in Bugzilla. > > http://bugzilla.redhat.com/ Thanks for the hint. ;-) I just sent the report. Sorry for my missunderstanding. I was too tired yesterday. Best Regards, Antonio Gallardo. From ivazquez at ivazquez.net Sun May 1 03:13:05 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Sat, 30 Apr 2005 23:13:05 -0400 Subject: Plase update tzdata to 2005i In-Reply-To: <34471.165.98.153.184.1114908627.squirrel@www.agssa.net> References: <42078.165.98.153.184.1114827118.squirrel@www.agssa.net> <1114828507.7966.0.camel@ignacio.ignacio.lan> <34471.165.98.153.184.1114908627.squirrel@www.agssa.net> Message-ID: <1114917185.704.2.camel@ignacio.ignacio.lan> On Sat, 2005-04-30 at 19:50 -0500, Antonio Gallardo wrote: > On Vie, 29 de Abril de 2005, 21:35, Ignacio Vazquez-Abrams dijo: > > On Fri, 2005-04-29 at 21:11 -0500, Antonio Gallardo wrote: > >> Is posible to see a patch for FC3 and include it in the next FC4 > >> release? > > > > Request it in Bugzilla. > > > > http://bugzilla.redhat.com/ > > Thanks for the hint. ;-) > > I just sent the report. Sorry for my missunderstanding. I was too tired > yesterday. No worries. According to the CVS commits list it will definitely be in Rawhide. I suppose we shall see if it makes it into FC3. -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From antonio at apache.org Sun May 1 06:39:36 2005 From: antonio at apache.org (Antonio Gallardo) Date: Sun, 1 May 2005 01:39:36 -0500 (CDT) Subject: Plase update tzdata to 2005i Message-ID: <36576.165.98.153.184.1114929576.squirrel@www.agssa.net> On Sab, 30 de Abril de 2005, 22:13, Ignacio Vazquez-Abrams dijo: > On Sat, 2005-04-30 at 19:50 -0500, Antonio Gallardo wrote: >> On Vie, 29 de Abril de 2005, 21:35, Ignacio Vazquez-Abrams dijo: >> > On Fri, 2005-04-29 at 21:11 -0500, Antonio Gallardo wrote: >> >> Is posible to see a patch for FC3 and include it in the next FC4 release? >> > >> > Request it in Bugzilla. >> > >> > http://bugzilla.redhat.com/ >> >> Thanks for the hint. ;-) >> >> I just sent the report. Sorry for my missunderstanding. I was too tired yesterday. > > No worries. > > According to the CVS commits list it will definitely be in Rawhide. Yep! I see it (committed 22 hours ago): http://cvs.fedora.redhat.com/viewcvs/devel/tzdata/tzdata.spec?rev=1.14&view=auto > I suppose we shall see if it makes it into FC3. This is the idea. Actually as a workaround, I setted the timezone to North Dakota/Center. Best Regards, Antonio Gallardo. > -- > Ignacio Vazquez-Abrams > http://fedora.ivazquez.net/ > > gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list From dwmw2 at infradead.org Sun May 1 07:55:28 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Sun, 01 May 2005 08:55:28 +0100 Subject: rawhide report: 20050430 changes In-Reply-To: <1114907104.21901.31.camel@tortoise.toronto.redhat.com> References: <200504301544.j3UFiMTg030812@porkchop.devel.redhat.com> <1114877909.24014.84.camel@localhost.localdomain> <1114907104.21901.31.camel@tortoise.toronto.redhat.com> Message-ID: <1114934129.24014.96.camel@localhost.localdomain> On Sat, 2005-04-30 at 20:25 -0400, Thomas Fitzsimmons wrote: > Yes, gcjwebplugin should be able to access https sites now. Seems it can do https but apparently not SSLv3. Do I need to install OpenSSL Java bindings (or a complete reimplementation of SSL in Java) for it? ... $ gcjappletviewer "https://service.oneaccount.com/Service/Status/int001p.config_check?brand=2" raw arguments: https://service.oneaccount.com/Service/Status/int001p.config_check?brand=2 java.lang.RuntimeException: error instantiating default socket factory: java.security.NoSuchAlgorithmException: SSLv3 at javax.net.ssl.SSLSocketFactory.getDefault() (/usr/lib/libgcj.so.6.0.0) -- dwmw2 From toshio at tiki-lounge.com Sun May 1 12:26:38 2005 From: toshio at tiki-lounge.com (Toshio Kuratomi) Date: Sun, 1 May 2005 05:26:38 -0700 Subject: ARTSD In-Reply-To: <200504301854.44620.rjune@bravegnuworld.com>; from rjune@bravegnuworld.com on Sat, Apr 30, 2005 at 06:54:41PM -0500 References: <042920052009.431.427294910005280B000001AF2200748184020A04049A030C03A103079B@comcast.net> <200504292214.06093.ml-fedora@fathomssen.de> <000a01c54d0e$aebefc00$0201a8c0@Greeney> <200504301854.44620.rjune@bravegnuworld.com> Message-ID: <20050501052638.A18884@tiki-lounge.com> On Sat, Apr 30, 2005 at 06:54:41PM -0500, Richard June wrote: > Arts was most definately not designed to work transparently over the network. > rather than send the file for processing, arts will send a filename over the > network, so unless you tell arts to use NAS or ESD, it's useless for network > audio. > I agree it would be great to have alsa do that though AFAIK ALSA is Linux specific whereas gnome and kde are cross-platform. So the important thing isn't to have alsa do network audio. It's to have a cross platform API that addresses the present shortcomings (Alan posts about those nearly every time this comes up.) Then it can be implemented on Linux and other platforms in whatever form makes sense. -Toshio From enrico.scholz at informatik.tu-chemnitz.de Sun May 1 13:00:00 2005 From: enrico.scholz at informatik.tu-chemnitz.de (Enrico Scholz) Date: Sun, 01 May 2005 15:00:00 +0200 Subject: gcc-4 inline assembler question (no register in class 'GENERAL_REGS' error) Message-ID: <87pswbcd9r.fsf@kosh.bigo.ensc.de> Hello, while trying to build qemu, I see errors in the inline assembler. Investigating it shows that the problem is caused by code like: ----- typedef unsigned int size_t; #define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER) #define Xoffsetof(TYPE, MEMBER) (*(unsigned char*)offsetof(TYPE, MEMBER)) typedef struct { unsigned long regs[8]; } CPUState; int foo() { asm(" fs movl %0, %%eax\n" " fs movl %1, %%ebx\n" " fs movl %2, %%ecx\n" " fs movl %3, %%edx\n" " fs movl %4, %%esp\n" " fs movl %5, %%ebp\n" " fs movl %6, %%esi\n" // ** : : "m" (Xoffsetof(CPUState, regs[0])), "m" (Xoffsetof(CPUState, regs[1])), "m" (Xoffsetof(CPUState, regs[2])), "m" (Xoffsetof(CPUState, regs[3])), "m" (Xoffsetof(CPUState, regs[4])), "m" (Xoffsetof(CPUState, regs[5])), // ** "m" (Xoffsetof(CPUState, regs[6])) ); } ------- | $ gcc -c foo.c | foo.c: In function 'foo': | foo.c:11: error: can't find a register in class 'GENERAL_REGS' while reloading 'asm' gcc32 does not have problems with it and generates code like: | $ gcc32 -c foo.c | $ objdump -d foo.o | 00000000 : | 0: 55 push %ebp | 1: 89 e5 mov %esp,%ebp | 3: 64 a1 00 00 00 00 mov %fs:0x0,%eax | 9: 64 8b 1d 04 00 00 00 mov %fs:0x4,%ebx | 10: 64 8b 0d 08 00 00 00 mov %fs:0x8,%ecx | 17: 64 8b 15 0c 00 00 00 mov %fs:0xc,%edx | 1e: 64 8b 25 10 00 00 00 mov %fs:0x10,%esp | 25: 64 8b 2d 14 00 00 00 mov %fs:0x14,%ebp | 2c: 64 8b 35 18 00 00 00 mov %fs:0x18,%esi | 33: c9 leave | 34: c3 ret Further investigations show that the removal of the lines marked with '**' lets it succeed with gcc-4 also, and generates correct but ineffizient code: | $ gcc -c foo.c | $ objdump -d foo.o | 00000000 : | 0: 55 push %ebp | 1: 89 e5 mov %esp,%ebp | 3: 57 push %edi | 4: 56 push %esi | 5: 53 push %ebx | 6: bf 00 00 00 00 mov $0x0,%edi | b: bb 04 00 00 00 mov $0x4,%ebx | 10: be 08 00 00 00 mov $0x8,%esi | 15: b9 0c 00 00 00 mov $0xc,%ecx | 1a: ba 10 00 00 00 mov $0x10,%edx | 1f: b8 18 00 00 00 mov $0x18,%eax | 24: 64 8b 07 mov %fs:(%edi),%eax | 27: 64 8b 1b mov %fs:(%ebx),%ebx | 2a: 64 8b 0e mov %fs:(%esi),%ecx | 2d: 64 8b 11 mov %fs:(%ecx),%edx | 30: 64 8b 22 mov %fs:(%edx),%esp | 33: 64 8b 28 mov %fs:(%eax),%ebp | 36: 5b pop %ebx | 37: 5e pop %esi | 38: 5f pop %edi | 39: c9 leave | 3a: c3 ret This explains why gcc-4 complains about the missing regs: it tries to load the immediate offset values into general regs first and accesses the memory based on them then. With more than 6 variables there are not enough regs available. Now my question: what is the correct way to achieve the first object code with inline assembler? Using the specific registers as machine constraints does not work as the posted assembly block happens within more complex code in the real qemu code. I tried things like | asm("fs movl (%0),%%ebx" : : "i"(offsetof(CPUState, regs[1]))) already but it generates the wrong code ('mov %fs:0x0,%ebx'). Enrico -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 480 bytes Desc: not available URL: From buildsys at redhat.com Sun May 1 15:37:54 2005 From: buildsys at redhat.com (Build System) Date: Sun, 1 May 2005 11:37:54 -0400 Subject: rawhide report: 20050501 changes Message-ID: <200505011537.j41FbsIm014483@porkchop.devel.redhat.com> Updated Packages: From mrsam at courier-mta.com Sun May 1 16:01:32 2005 From: mrsam at courier-mta.com (Sam Varshavchik) Date: Sun, 01 May 2005 12:01:32 -0400 Subject: Who is maintaining libwww? Message-ID: According to http://www.w3.org/Library/, W3C is not maintaining libwww any more. I have a couple of bugs I'd like to see squashed in libwww, but I'm still running FC2. If someone can confirm that they're still present in FC3/FC4 I suppose I can put them into Bugzilla, but they should really go upstream. The bugs are: 1) Segfault if HTProfile_delete() is invoked before HTLibTerminate(). 2) The order of libraries from "libwww-config --libs" does not match their internal dependency order. This does not matter when building dynamically-linked www application. But a static link will fail due to unresolved symbols. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From fitzsim at redhat.com Sun May 1 16:55:57 2005 From: fitzsim at redhat.com (Thomas Fitzsimmons) Date: Sun, 01 May 2005 12:55:57 -0400 Subject: rawhide report: 20050430 changes In-Reply-To: <1114934129.24014.96.camel@localhost.localdomain> References: <200504301544.j3UFiMTg030812@porkchop.devel.redhat.com> <1114877909.24014.84.camel@localhost.localdomain> <1114907104.21901.31.camel@tortoise.toronto.redhat.com> <1114934129.24014.96.camel@localhost.localdomain> Message-ID: <1114966557.21901.49.camel@tortoise.toronto.redhat.com> On Sun, 2005-05-01 at 08:55 +0100, David Woodhouse wrote: > On Sat, 2005-04-30 at 20:25 -0400, Thomas Fitzsimmons wrote: > > Yes, gcjwebplugin should be able to access https sites now. > > Seems it can do https but apparently not SSLv3. > > Do I need to install OpenSSL Java bindings (or a complete > reimplementation of SSL in Java) for it? ... > > $ gcjappletviewer "https://service.oneaccount.com/Service/Status/int001p.config_check?brand=2" > raw arguments: > https://service.oneaccount.com/Service/Status/int001p.config_check?brand=2 > java.lang.RuntimeException: error instantiating default socket factory: java.security.NoSuchAlgorithmException: SSLv3 > at javax.net.ssl.SSLSocketFactory.getDefault() (/usr/lib/libgcj.so.6.0.0) > Update jessie to 1.0.0-6. Tom From perbj at stanford.edu Sun May 1 17:28:23 2005 From: perbj at stanford.edu (Per Bjornsson) Date: Sun, 01 May 2005 10:28:23 -0700 Subject: ARTSD In-Reply-To: <20050501052638.A18884@tiki-lounge.com> References: <042920052009.431.427294910005280B000001AF2200748184020A04049A030C03A103079B@comcast.net> <200504292214.06093.ml-fedora@fathomssen.de> <000a01c54d0e$aebefc00$0201a8c0@Greeney> <200504301854.44620.rjune@bravegnuworld.com> <20050501052638.A18884@tiki-lounge.com> Message-ID: <1114968503.6971.11.camel@localhost.localdomain> On Sun, 2005-05-01 at 05:26 -0700, Toshio Kuratomi wrote: > AFAIK ALSA is Linux specific whereas gnome and kde are cross-platform. So > the important thing isn't to have alsa do network audio. It's to have a > cross platform API that addresses the present shortcomings (Alan posts > about those nearly every time this comes up.) Then it can be implemented on > Linux and other platforms in whatever form makes sense. Well, the solution that GNOME seems to have converged on is to use Gstreamer and use whatever default output method Gstreamer is set up for. On many desktop systems, using the Gstreamer alsasink output plugin would be the best; in other cases you can either use e.g. esd or some other sound server which can provide network transparency. At least some KDE programs use Gstreamer as well, and apparently there is talk of using Gstreamer pretty much as the default audio solution for KDE 4. So it seems that Gstreamer can fulfill the role of being _the_ audio API that desktop apps use. /Per -- Per Bjornsson Ph.D. Candidate, Department of Applied Physics, Stanford University From dwmw2 at infradead.org Sun May 1 20:49:20 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Sun, 01 May 2005 21:49:20 +0100 Subject: rawhide report: 20050430 changes In-Reply-To: <1114966557.21901.49.camel@tortoise.toronto.redhat.com> References: <200504301544.j3UFiMTg030812@porkchop.devel.redhat.com> <1114877909.24014.84.camel@localhost.localdomain> <1114907104.21901.31.camel@tortoise.toronto.redhat.com> <1114934129.24014.96.camel@localhost.localdomain> <1114966557.21901.49.camel@tortoise.toronto.redhat.com> Message-ID: <1114980560.3388.2.camel@localhost.localdomain> On Sun, 2005-05-01 at 12:55 -0400, Thomas Fitzsimmons wrote: > Update jessie to 1.0.0-6. Yay. Now the appletviewer tells me that showDocument is not implemented in standalone mode. Running it in Firefox or Mozilla I get a grey box where the app should be, which turns white after a few seconds. Any tips on debugging it? -- dwmw2 From drepper at redhat.com Sun May 1 20:58:48 2005 From: drepper at redhat.com (Ulrich Drepper) Date: Sun, 01 May 2005 13:58:48 -0700 Subject: gcc-4 inline assembler question (no register in class 'GENERAL_REGS' error) In-Reply-To: <87pswbcd9r.fsf@kosh.bigo.ensc.de> References: <87pswbcd9r.fsf@kosh.bigo.ensc.de> Message-ID: <42754308.5030207@redhat.com> Enrico Scholz wrote: > asm(" fs movl %0, %%eax\n" > " fs movl %1, %%ebx\n" > " fs movl %2, %%ecx\n" > " fs movl %3, %%edx\n" > " fs movl %4, %%esp\n" > " fs movl %5, %%ebp\n" > " fs movl %6, %%esi\n" // ** > : > : "m" (Xoffsetof(CPUState, regs[0])), > "m" (Xoffsetof(CPUState, regs[1])), > "m" (Xoffsetof(CPUState, regs[2])), > "m" (Xoffsetof(CPUState, regs[3])), > "m" (Xoffsetof(CPUState, regs[4])), > "m" (Xoffsetof(CPUState, regs[5])), // ** > "m" (Xoffsetof(CPUState, regs[6])) This form is calling for trouble anyway. Just use the following which should work well with older gcc versions as well: asm(" fs movl %c0, %%eax\n" " fs movl %c1, %%ebx\n" " fs movl %c2, %%ecx\n" " fs movl %c3, %%edx\n" " fs movl %c4, %%esp\n" " fs movl %c5, %%ebp\n" " fs movl %c6, %%esi\n" : : "i" (offsetof(CPUState, regs[0])), "i" (offsetof(CPUState, regs[1])), "i" (offsetof(CPUState, regs[2])), "i" (offsetof(CPUState, regs[3])), "i" (offsetof(CPUState, regs[4])), "i" (offsetof(CPUState, regs[5])), "i" (offsetof(CPUState, regs[6])) ); -- ? Ulrich Drepper ? Red Hat, Inc. ? 444 Castro St ? Mountain View, CA ? -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From stransky at redhat.com Mon May 2 07:24:06 2005 From: stransky at redhat.com (Martin Stransky) Date: Mon, 02 May 2005 09:24:06 +0200 Subject: rawhide report: 20050428 changes In-Reply-To: <20050429151100.GB6299@thacker.dyndns.org> References: <200504281200.j3SC0wIp029505@porkchop.devel.redhat.com> <200504281622.58702.fedora-devel-list@cygnusx-1.org> <003801c54c7f$5a5bf260$0201a8c0@Greeney> <1114760835.5061.52.camel@ignacio.ignacio.lan> <1114785992.10618.6.camel@mlenzdesktop> <20050429151100.GB6299@thacker.dyndns.org> Message-ID: <4275D596.5000301@redhat.com> Yes, I'll improve ainit or other sound tools for easy enable/disable dmix, because people with good sound card realy don't need it. Ma. >>i'll have to check that out on my audigy2 tonight. assuming I can get >>it to work, it hasn't worked since fc4t2 install. >> >>If that is the case then the logic behind enabling dmix should look for >>cards that already support hardware audio multiplexing and not enable >>dmix. better yet, the alsa mixer should have an option to enable and >>disable software mixing. > > > It's supported hardware mixing just fine on my live card for a long > time. The recent changes resulted in much scratchier sound for me > when using applications that don't use a direct ALSA output. (Sound > output from the proprietary Flash plugin, for example.) It looks > to me like ainit is turning on dmix even for my card which supports > hardware mixing; I'd rather it didn't. > > John Thacker > From pzad at pobox.sk Mon May 2 07:38:05 2005 From: pzad at pobox.sk (Peter Zubaj) Date: Mon, 2 May 2005 09:38:05 +0200 Subject: rawhide report: 20050428 changes Message-ID: <200505020738.j427c1fw005753@www4.pobox.sk> Hmm, AFAIK ALSA 1.0.9rc2 uses dmix as default for cards which do not support hw mixing. Cards which support hw mixing do not use dmix. This is done in card config files (compare CMI8338.conf and Audigy2.conf). Peter Zubaj >Yes, I'll improve ainit or other sound tools for easy enable/disable >dmix, because people with good sound card realy don't need it. ____________________________________ Vsetko o SuperStar http://superstar.atlas.sk From stransky at redhat.com Mon May 2 08:37:49 2005 From: stransky at redhat.com (Martin Stransky) Date: Mon, 02 May 2005 10:37:49 +0200 Subject: rawhide report: 20050428 changes In-Reply-To: <200505020738.j427c1fw005753@www4.pobox.sk> References: <200505020738.j427c1fw005753@www4.pobox.sk> Message-ID: <4275E6DD.8070001@redhat.com> Ok, I'll investigate it. Ma. > Hmm, AFAIK ALSA 1.0.9rc2 uses dmix as default for cards which do not > support hw mixing. > Cards which support hw mixing do not use dmix. This is done in card > config files (compare CMI8338.conf and Audigy2.conf). > > Peter Zubaj > > >>Yes, I'll improve ainit or other sound tools for easy enable/disable >>dmix, because people with good sound card realy don't need it. > > > > > ____________________________________ > Vsetko o SuperStar > http://superstar.atlas.sk > > From dwmw2 at infradead.org Mon May 2 08:38:14 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Mon, 02 May 2005 09:38:14 +0100 Subject: rawhide report: 20050428 changes In-Reply-To: <4275E6DD.8070001@redhat.com> References: <200505020738.j427c1fw005753@www4.pobox.sk> <4275E6DD.8070001@redhat.com> Message-ID: <1115023094.3287.5.camel@localhost.localdomain> On Mon, 2005-05-02 at 10:37 +0200, Martin Stransky wrote: > Ok, I'll investigate it. Btw can we have alsa-lib-1.0.9rc2-2 in dist-fc4 please? The version which fills ~/.asoundrc with 0xFF until the disk is full is somewhat suboptimal :) -- dwmw2 From stransky at redhat.com Mon May 2 09:08:15 2005 From: stransky at redhat.com (Martin Stransky) Date: Mon, 02 May 2005 11:08:15 +0200 Subject: rawhide report: 20050428 changes In-Reply-To: <1115023094.3287.5.camel@localhost.localdomain> References: <200505020738.j427c1fw005753@www4.pobox.sk> <4275E6DD.8070001@redhat.com> <1115023094.3287.5.camel@localhost.localdomain> Message-ID: <4275EDFF.8040204@redhat.com> You can download it from http://people.redhat.com/stransky/alsa/ Ma. > On Mon, 2005-05-02 at 10:37 +0200, Martin Stransky wrote: > >>Ok, I'll investigate it. > > > Btw can we have alsa-lib-1.0.9rc2-2 in dist-fc4 please? The version > which fills ~/.asoundrc with 0xFF until the disk is full is somewhat > suboptimal :) From thomasz at hostmaster.org Mon May 2 12:23:47 2005 From: thomasz at hostmaster.org (Thomas Zehetbauer) Date: Mon, 02 May 2005 14:23:47 +0200 Subject: Firefox printing problem Message-ID: <1115036627.28182.10.camel@hostmaster.org> Hi, I am having trouble printing with firefox-1.0.3-2 / mozilla-1.7.7-2 on current FC4 rawhide. When I print the following HTML page I get the postscript file linked below: http://beta.hostmaster.org/Economy/Flyer.html?print=1 http://www.hostmaster.org/~thomasz/mozilla.ps Note that there is a unexpected space after some umlauts while others are merged with their adjacent letters and the italic style is missing. Print preview looks OK and I wonder if someone else is seeing this. Tom -- T h o m a s Z e h e t b a u e r ( TZ251 ) PGP encrypted mail preferred - KeyID 96FFCB89 finger thomasz at hostmaster.org for key -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 481 bytes Desc: This is a digitally signed message part URL: From jnovy at redhat.com Mon May 2 12:38:06 2005 From: jnovy at redhat.com (Jindrich Novy) Date: Mon, 02 May 2005 14:38:06 +0200 Subject: Duplicated files in the pristine FC4t2 installation Message-ID: <1115037486.29438.54.camel@obelix.redhat.usu> Hello all, I've found some file duplicates when I browsed through the /usr directory tree in the pristine & complete FC4t2 installation what made me curious how many duplicates are there in total. This is not critical, so please take this as something for your information that some files would better be symlinked/hardlinked in order to not to waste disc space without a point. I know there's sometimes no other way that to duplicate a file, but the statistics I have is IMHO rather interesting: 206405 regular files found, 4149 MiB [5468 MiB] 15797 total dupes, 15705 non-zero sized. 96 MiB [161 MiB], 2.325% [2.951%] wasted by dupes, 13906 symlinks, 5042 hardlinks. So that 161 MiB is "wasted" physically in the /usr tree, what is about 3% in total from all the files within the /usr hierarchy. To let this information be somehow worth for the package maintainers, I'm adding a link to the list of all the duplicated files including their sizes and md5 sums and to what package they belong: http://people.redhat.com/jnovy/files/FC4t2-usr-dupes.gz This statistics was done by the "slink" utility I wrote some time ago. It's able to replace duplicates with symbolic links to save disc space [EXPERIMENTAL, but seems to work] or just display a statistics about duplicates for a given directory. If you want to give it a try, get it from: http://people.redhat.com/jnovy/files/slink-0.0.1-pre1.tar.bz2 It's interesting how many GPL "COPYING" clones we have in /usr/share/doc for instance. Unfortunately some of my packages are also affected ;) Regards, Jindrich -- Jindrich Novy , http://people.redhat.com/jnovy/ The worst evil in the world is refusal to think. From arjanv at redhat.com Mon May 2 12:43:07 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Mon, 2 May 2005 14:43:07 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115037486.29438.54.camel@obelix.redhat.usu> References: <1115037486.29438.54.camel@obelix.redhat.usu> Message-ID: <20050502124307.GA16882@devserv.devel.redhat.com> On Mon, May 02, 2005 at 02:38:06PM +0200, Jindrich Novy wrote: > > This statistics was done by the "slink" utility I wrote some time ago. we also ship the hardlink utility, which is probably a tiny bit nicer in that it makes the replacement more transparent for things like rpm From j.w.r.degoede at hhs.nl Mon May 2 12:52:02 2005 From: j.w.r.degoede at hhs.nl (Hans de Goede) Date: Mon, 02 May 2005 14:52:02 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115037486.29438.54.camel@obelix.redhat.usu> References: <1115037486.29438.54.camel@obelix.redhat.usu> Message-ID: <42762272.1080102@hhs.nl> Hmm, Lots of pyo pyc duplicates, this should be somehow fixed in python can RPM handle hardlinks iow can an rpm contain a file and hardlink to the file instead of 2 copies of the file? If rpm can handle hardlinks then this should be fixable preferrably python should just create a hardlink when the pyc and pyo are the same. About licenses what about a licenses.rpm which gets installed by default which contains most common licenses and which is an obligatory part of the base system. And then just put a ptr to the license in the rpm -qi info? That would also cleanout /usr/share/doc since quite a few dirs there only contain a copy of the GPL / MPL / whatever license Regards, Hans Jindrich Novy wrote: > Hello all, > > I've found some file duplicates when I browsed through the /usr > directory tree in the pristine & complete FC4t2 installation what made > me curious how many duplicates are there in total. This is not critical, > so please take this as something for your information that some files > would better be symlinked/hardlinked in order to not to waste disc space > without a point. I know there's sometimes no other way that to duplicate > a file, but the statistics I have is IMHO rather interesting: > > 206405 regular files found, 4149 MiB [5468 MiB] > 15797 total dupes, 15705 non-zero sized. > 96 MiB [161 MiB], 2.325% [2.951%] wasted by dupes, 13906 symlinks, 5042 > hardlinks. > > So that 161 MiB is "wasted" physically in the /usr tree, what is about > 3% in total from all the files within the /usr hierarchy. > > To let this information be somehow worth for the package maintainers, > I'm adding a link to the list of all the duplicated files including > their sizes and md5 sums and to what package they belong: > > http://people.redhat.com/jnovy/files/FC4t2-usr-dupes.gz > > This statistics was done by the "slink" utility I wrote some time ago. > It's able to replace duplicates with symbolic links to save disc space > [EXPERIMENTAL, but seems to work] or just display a statistics about > duplicates for a given directory. If you want to give it a try, get it > from: > > http://people.redhat.com/jnovy/files/slink-0.0.1-pre1.tar.bz2 > > It's interesting how many GPL "COPYING" clones we have in /usr/share/doc > for instance. Unfortunately some of my packages are also affected ;) > > Regards, > Jindrich > From arjanv at redhat.com Mon May 2 12:58:20 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Mon, 02 May 2005 08:58:20 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <42762272.1080102@hhs.nl> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> Message-ID: <1115038700.6098.0.camel@localhost.localdomain> > About licenses what about a licenses.rpm which gets installed by default > which contains most common licenses and which is an obligatory part of > the base system. And then just put a ptr to the license in the rpm -qi > info? That would also cleanout /usr/share/doc since quite a few dirs > there only contain a copy of the GPL / MPL / whatever license problem is that the GPL at least requires you to ship the license with the software. And RPMs get used and for sure distributed outside fedora... -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From j.w.r.degoede at hhs.nl Mon May 2 13:07:09 2005 From: j.w.r.degoede at hhs.nl (Hans de Goede) Date: Mon, 02 May 2005 15:07:09 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115038700.6098.0.camel@localhost.localdomain> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> <1115038700.6098.0.camel@localhost.localdomain> Message-ID: <427625FD.2090408@hhs.nl> Putting on evil hat: I already was afraid that would be used as an argument, taking this over the top glibc-devel and glibc-common don't contain COPYING, only glibc does. That can be explained in 2 ways: -all (L)GPL devel rpms should be modified to include COPYING -we can create a licenses.rpm as long as its Requires: by rpms which fall under one of the licenses in licenses.rpm Regards, Hans Arjan van de Ven wrote: >>About licenses what about a licenses.rpm which gets installed by default >>which contains most common licenses and which is an obligatory part of >>the base system. And then just put a ptr to the license in the rpm -qi >>info? That would also cleanout /usr/share/doc since quite a few dirs >>there only contain a copy of the GPL / MPL / whatever license > > > problem is that the GPL at least requires you to ship the license with > the software. And RPMs get used and for sure distributed outside > fedora... > > > ------------------------------------------------------------------------ > > -- > Fedora-maintainers mailing list > Fedora-maintainers at redhat.com > https://www.redhat.com/mailman/listinfo/fedora-maintainers From jnovy at redhat.com Mon May 2 13:09:46 2005 From: jnovy at redhat.com (Jindrich Novy) Date: Mon, 02 May 2005 15:09:46 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115038700.6098.0.camel@localhost.localdomain> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> <1115038700.6098.0.camel@localhost.localdomain> Message-ID: <1115039386.29438.63.camel@obelix.redhat.usu> On Mon, 2005-05-02 at 08:58 -0400, Arjan van de Ven wrote: > > About licenses what about a licenses.rpm which gets installed by default > > which contains most common licenses and which is an obligatory part of > > the base system. And then just put a ptr to the license in the rpm -qi > > info? That would also cleanout /usr/share/doc since quite a few dirs > > there only contain a copy of the GPL / MPL / whatever license > > problem is that the GPL at least requires you to ship the license with > the software. And RPMs get used and for sure distributed outside > fedora... This wouldn't be a drawback if some utility would search and replace the duplicates in /usr/share/doc for instance after installation of all the selected packages. But I'm not sure if rpm will be happy with it... -- Jindrich Novy , http://people.redhat.com/jnovy/ The worst evil in the world is refusal to think. From lfarkas at bppiac.hu Mon May 2 14:15:20 2005 From: lfarkas at bppiac.hu (Farkas Levente) Date: Mon, 02 May 2005 16:15:20 +0200 Subject: can't restart ntpd Message-ID: <427635F8.7040303@bppiac.hu> hi, can't (re)start ntpd server on fc3. ie. after: --------------------------------------- # /etc/init.d/ntpd restart Shutting down ntpd: [FAILED] ntpd: Synchronizing with time server: [ OK ] Starting ntpd: [ OK ] # ps axu|grep ntp root 23165 0.0 0.0 4992 688 pts/0 S+ 14:40 0:00 grep ntp --------------------------------------- after i found it just test it: --------------------------------------- # ntpd -n -u ntp:ntp -p /var/run/ntpd.pid Segmentation fault --------------------------------------- so i debug ntpd itself: --------------------------------------- gdb ntpd GNU gdb Red Hat Linux (6.1post-1.20040607.43rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-redhat-linux-gnu"...(no debugging symbols found)...Using host libthread_db library "/lib/tls/libthread_db.so.1". (gdb) set args -n -u ntp:ntp -p /var/run/ntpd.pid (gdb) run Starting program: /usr/sbin/ntpd -n -u ntp:ntp -p /var/run/ntpd.pid (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)... Program received signal SIGSEGV, Segmentation fault. 0x00143de9 in _dl_load_cache_lookup () from /lib/ld-linux.so.2 (gdb) bt #0 0x00143de9 in _dl_load_cache_lookup () from /lib/ld-linux.so.2 #1 0x001434b8 in _dl_map_object () from /lib/ld-linux.so.2 #2 0x0024eb58 in dl_open_worker () from /lib/tls/libc.so.6 #3 0x00000000 in ?? () --------------------------------------- after installing the ntp-debuginfo i've got this, so seems to be a bug in ntpd: --------------------------------------- (gdb) run Starting program: /usr/sbin/ntpd -n -u ntp:ntp -p /var/run/ntpd.pid Program received signal SIGSEGV, Segmentation fault. 0x00cbed30 in vfprintf () from /lib/tls/libc.so.6 (gdb) bt #0 0x00cbed30 in vfprintf () from /lib/tls/libc.so.6 #1 0x00cdf536 in vsnprintf () from /lib/tls/libc.so.6 #2 0x008a834a in msyslog (level=-1074460208, fmt=0xbff509d0 "\uffff\005") at msyslog.c:165 #3 0x008721a3 in loop_config (item=-1074460208, freq=0) at ntp_loopfilter.c:858 (gdb) list 858 msyslog(LOG_INFO, 859 "kernel time sync status %04x", 860 pll_status); 861 } 862 #endif /* KERNEL_PLL */ 863 #endif /* LOCKCLOCK */ 864 break; 865 866 case LOOP_DRIFTCOMP: 867 --------------------------------------- what else can i do? yours. https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156603 -- Levente "Si vis pacem para bellum!" From alan at balclutha.org Mon May 2 14:49:56 2005 From: alan at balclutha.org (Alan Milligan) Date: Tue, 03 May 2005 00:49:56 +1000 Subject: rhnsd internals Message-ID: <42763E14.5080303@balclutha.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I'm attempting to figure out how rhnsd works and am having some rather fundamental issues I'd appreciate explanations for. I'm specifically talking about the queue.get() and queue.submit() XML-RPC requests in rhn_check. From looking at code, this process looks esoteric (but possibly very cunning) to say the least. Firstly, this seems to work like a queue whereby each system, identified by ids may have a list of action objects awaiting retrieval, based upon uptime timestamps. I'm trying to visualise what these queue 'actions' may look like. Obviously, there isn't one waiting for each known system, this must be calculated in a generalised fashion. How does one then know when each 'action' has served it's purpose and can be removed from the queue? I'm also trying to figure out the format these actions are transmitted to the rhnsd client. It appears to be unmarshallable as a consequent XML-RPC call for the client to make. I see that each action is dynamically constructing a functor in /usr/share/rhn/actions. But am a bit baffled as to how its all supposed to work, parameters instantiated and passed into the methods. An example of actual action responses sent to the client would go a long way to clarifying this for me. Also, it would appear the rhnsd client continues polling for actions until it receives an empty dict. Perhaps the explanation of what constitutes an action will prove that this is eminently sensible behaviour. TIA, Alan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFCdj4UCfroLk4EZpkRAn33AJ9z6abFqKbNJnqWyCj6V9LoaZ6/5QCeMh4S aSe03b31wVPdQmXIRkglyUQ= =DD72 -----END PGP SIGNATURE----- From buildsys at redhat.com Mon May 2 15:42:36 2005 From: buildsys at redhat.com (Build System) Date: Mon, 2 May 2005 11:42:36 -0400 Subject: rawhide report: 20050502 changes Message-ID: <200505021542.j42FgakN028731@porkchop.devel.redhat.com> Updated Packages: From fitzsim at redhat.com Mon May 2 17:43:06 2005 From: fitzsim at redhat.com (Thomas Fitzsimmons) Date: Mon, 02 May 2005 13:43:06 -0400 Subject: rawhide report: 20050430 changes In-Reply-To: <1114980560.3388.2.camel@localhost.localdomain> References: <200504301544.j3UFiMTg030812@porkchop.devel.redhat.com> <1114877909.24014.84.camel@localhost.localdomain> <1114907104.21901.31.camel@tortoise.toronto.redhat.com> <1114934129.24014.96.camel@localhost.localdomain> <1114966557.21901.49.camel@tortoise.toronto.redhat.com> <1114980560.3388.2.camel@localhost.localdomain> Message-ID: <1115055786.2884.2.camel@rh-ibm-t41> On Sun, 2005-05-01 at 21:49 +0100, David Woodhouse wrote: > On Sun, 2005-05-01 at 12:55 -0400, Thomas Fitzsimmons wrote: > > Update jessie to 1.0.0-6. > > Yay. Now the appletviewer tells me that showDocument is not implemented > in standalone mode. Running it in Firefox or Mozilla I get a grey box > where the app should be, which turns white after a few seconds. Any tips > on debugging it? > Run firefox with the -g option. Tom From jlb17 at duke.edu Mon May 2 18:32:00 2005 From: jlb17 at duke.edu (Joshua Baker-LePain) Date: Mon, 2 May 2005 14:32:00 -0400 (EDT) Subject: 3w-9xxx module version in FC4 In-Reply-To: <20050430040047.16D397365B@hormel.redhat.com> References: <20050430040047.16D397365B@hormel.redhat.com> Message-ID: On Sat, 30 Apr 2005 at 9:11am, jkeating at j2solutions.net wrote > On Fri, 2005-04-29 at 11:18 -0400, Joshua Baker-LePain wrote: > > > > Err, really? I saw exactly the opposite. Now, this was on centos 4, but > > these numbers are for a software RAID0 across 2 9500-12 cards in 11 disk > > RAID5 mode. I'd also done 'blockdev --setra 16384' on both 3ware > > devices, and this is with a kernel modified to include XFS: > > I can't help but wonder if the software RAID 0 is interfering with your > setup. Any chance you can test w/out the software RAID 0 involved? > Just a straight RAID 5 across 11 disks? No difference. This time, I tested with the stock kernel and ext3 (just so no one can argue that XFS and/or 8K stacks were the issue). So, this is tiobench on centos 4, 2.6.9-5.0.3smp, with the 9.1.5.2 firmware/stock driver vs. 9.2 firmware/driver: 9.1.5.2/stock: Read, 1 thread: 270.89 Read, 2 thread: 176.04 Read, 4 thread: 167.48 Read, 8 thread: 162.20 Write, 1 thread: 63.97 Write, 2 thread: 66.38 Write, 4 thread: 61.64 Write, 8 thread: 63.21 9.2: Read, 1 thread: 269.70 Read, 2 thread: 199.13 Read, 4 thread: 177.69 Read, 8 thread: 153.60 Write, 1 thread: 18.26 Write, 2 thread: 18.74 Write, 4 thread: 18.31 Write, 8 thread: 18.77 For completeness' sake, here are the XFS numbers (as usual, better than ext3): 9.1.5.2/stock: Read, 1 thread: 239.74 Read, 2 thread: 360.29 Read, 4 thread: 324.12 Read, 8 thread: 276.95 Write, 1 thread: 90.10 Write, 2 thread: 94.25 Write, 4 thread: 90.69 Write, 8 thread: 83.18 9.2: Read, 1 thread: 367.44 Read, 2 thread: 338.09 Read, 4 thread: 306.44 Read, 8 thread: 240.73 Write, 1 thread: 23.33 Write, 2 thread: 22.89 Write, 4 thread: 19.11 Write, 8 thread: 19.43 -- Joshua Baker-LePain Department of Biomedical Engineering Duke University From pjones at redhat.com Mon May 2 19:19:10 2005 From: pjones at redhat.com (Peter Jones) Date: Mon, 02 May 2005 15:19:10 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <42762272.1080102@hhs.nl> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> Message-ID: <1115061550.12093.0.camel@localhost.localdomain> On Mon, 2005-05-02 at 14:52 +0200, Hans de Goede wrote: > Hmm, > > Lots of pyo pyc duplicates, this should be somehow fixed in python can > RPM handle hardlinks iow can an rpm contain a file and hardlink to the > file instead of 2 copies of the file? > > If rpm can handle hardlinks then this should be fixable preferrably > python should just create a hardlink when the pyc and pyo are the same. You could make rpm run "hardlinks" on a directory, but the results are pretty painful -- you wind up running it per-package, and that means everything takes forever. -- Peter From arjanv at redhat.com Mon May 2 19:23:46 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Mon, 2 May 2005 21:23:46 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115061550.12093.0.camel@localhost.localdomain> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> <1115061550.12093.0.camel@localhost.localdomain> Message-ID: <20050502192346.GA19481@devserv.devel.redhat.com> On Mon, May 02, 2005 at 03:19:10PM -0400, Peter Jones wrote: > On Mon, 2005-05-02 at 14:52 +0200, Hans de Goede wrote: > > Hmm, > > > > Lots of pyo pyc duplicates, this should be somehow fixed in python can > > RPM handle hardlinks iow can an rpm contain a file and hardlink to the > > file instead of 2 copies of the file? > > > > If rpm can handle hardlinks then this should be fixable preferrably > > python should just create a hardlink when the pyc and pyo are the same. > > You could make rpm run "hardlinks" on a directory, but the results are > pretty painful -- you wind up running it per-package, and that means > everything takes forever. or run it from cron.weekly on /usr/share/doc ;) From fedora at camperquake.de Mon May 2 19:23:19 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Mon, 2 May 2005 21:23:19 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115061550.12093.0.camel@localhost.localdomain> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> <1115061550.12093.0.camel@localhost.localdomain> Message-ID: <20050502212319.09396f1a@nausicaa.camperquake.de> Hi. Peter Jones wrote: > You could make rpm run "hardlinks" on a directory, but the results are > pretty painful -- you wind up running it per-package, and that means > everything takes forever. Please don't. Having to do that on kernel-devel installs is bad enough. -- Posting at the top because that's where the cursor happened to be is like shitting in your pants because that's where your asshole happened to be. From pjones at redhat.com Mon May 2 19:26:05 2005 From: pjones at redhat.com (Peter Jones) Date: Mon, 02 May 2005 15:26:05 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <20050502212319.09396f1a@nausicaa.camperquake.de> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> <1115061550.12093.0.camel@localhost.localdomain> <20050502212319.09396f1a@nausicaa.camperquake.de> Message-ID: <1115061965.12093.6.camel@localhost.localdomain> On Mon, 2005-05-02 at 21:23 +0200, Ralf Ertzinger wrote: > Hi. > > Peter Jones wrote: > > > You could make rpm run "hardlinks" on a directory, but the results are > > pretty painful -- you wind up running it per-package, and that means > > everything takes forever. > > Please don't. Having to do that on kernel-devel installs is bad enough. That's my point, yes. -- Peter From roland at redhat.com Mon May 2 19:28:11 2005 From: roland at redhat.com (Roland McGrath) Date: Mon, 2 May 2005 12:28:11 -0700 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: Peter Jones's message of Monday, 2 May 2005 15:26:05 -0400 <1115061965.12093.6.camel@localhost.localdomain> Message-ID: <200505021928.j42JSB1F022621@magilla.sf.frob.com> I think what one clearly wants is for rpm to maintain an installed file indexed keyed by md5sum. Then you can have a tool that just uses this database to identify duplicates (and doesn't take forever), or have rpm do so itself when installing new files. From j.w.r.degoede at hhs.nl Mon May 2 19:30:23 2005 From: j.w.r.degoede at hhs.nl (Hans de Goede) Date: Mon, 02 May 2005 21:30:23 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <200505021928.j42JSB1F022621@magilla.sf.frob.com> References: <200505021928.j42JSB1F022621@magilla.sf.frob.com> Message-ID: <42767FCF.3050609@hhs.nl> Roland McGrath wrote: > I think what one clearly wants is for rpm to maintain an installed file > indexed keyed by md5sum. Then you can have a tool that just uses this > database to identify duplicates (and doesn't take forever), or have rpm do > so itself when installing new files. > Hmm, what about hash collisions, that would be really really BAD Regards, Hans From kyrre at solution-forge.net Mon May 2 19:32:05 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Mon, 02 May 2005 21:32:05 +0200 Subject: Firefox printing problem In-Reply-To: <1115036627.28182.10.camel@hostmaster.org> References: <1115036627.28182.10.camel@hostmaster.org> Message-ID: <1115062325.3343.33.camel@localhost.localdomain> man, 02.05.2005 kl. 14.23 skrev Thomas Zehetbauer: > Hi, > > I am having trouble printing with firefox-1.0.3-2 / mozilla-1.7.7-2 on > current FC4 rawhide. When I print the following HTML page I get the > postscript file linked below: > http://beta.hostmaster.org/Economy/Flyer.html?print=1 > http://www.hostmaster.org/~thomasz/mozilla.ps > > Note that there is a unexpected space after some umlauts while others > are merged with their adjacent letters and the italic style is missing. > Print preview looks OK and I wonder if someone else is seeing this. > > Tom I can see it (comparing your print to your postscript). There is another similar problem when printing from mozilla (and actually also KHTML (long since i have used it so i don't know if its still there) - it seems to have something to do with and "dotted frames", where text get written on top of each other making it really hard to read. It also looks fine in print preview. Anybody knows what i am talking about? Kyrre From shiva at sewingwitch.com Mon May 2 19:32:04 2005 From: shiva at sewingwitch.com (Kenneth Porter) Date: Mon, 02 May 2005 12:32:04 -0700 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <42767FCF.3050609@hhs.nl> References: <200505021928.j42JSB1F022621@magilla.sf.frob.com> <42767FCF.3050609@hhs.nl> Message-ID: --On Monday, May 02, 2005 9:30 PM +0200 Hans de Goede wrote: > Hmm, what about hash collisions, that would be really really BAD You could be paranoid and check the contents after the hashes identify the candidates for linking. From roland at redhat.com Mon May 2 19:35:52 2005 From: roland at redhat.com (Roland McGrath) Date: Mon, 2 May 2005 12:35:52 -0700 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: Hans de Goede's message of Monday, 2 May 2005 21:30:23 +0200 <42767FCF.3050609@hhs.nl> Message-ID: <200505021935.j42JZqNm022649@magilla.sf.frob.com> > Roland McGrath wrote: > > I think what one clearly wants is for rpm to maintain an installed file > > indexed keyed by md5sum. Then you can have a tool that just uses this > > database to identify duplicates (and doesn't take forever), or have rpm do > > so itself when installing new files. > > > > Hmm, what about hash collisions, that would be really really BAD If you are concerned about them you can still compare contents before declaring two files identical. But using the hashes as the main detector makes it fast, since you only examine the data of files that are 99.999% likely to be identical. From mharris at redhat.com Mon May 2 19:41:17 2005 From: mharris at redhat.com (Mike A. Harris) Date: Mon, 02 May 2005 15:41:17 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <20050502192346.GA19481@devserv.devel.redhat.com> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> <1115061550.12093.0.camel@localhost.localdomain> <20050502192346.GA19481@devserv.devel.redhat.com> Message-ID: <4276825D.7090204@redhat.com> Arjan van de Ven wrote: > On Mon, May 02, 2005 at 03:19:10PM -0400, Peter Jones wrote: > >>On Mon, 2005-05-02 at 14:52 +0200, Hans de Goede wrote: >> >>>Hmm, >>> >>>Lots of pyo pyc duplicates, this should be somehow fixed in python can >>>RPM handle hardlinks iow can an rpm contain a file and hardlink to the >>>file instead of 2 copies of the file? >>> >>>If rpm can handle hardlinks then this should be fixable preferrably >>>python should just create a hardlink when the pyc and pyo are the same. >> >>You could make rpm run "hardlinks" on a directory, but the results are >>pretty painful -- you wind up running it per-package, and that means >>everything takes forever. > > > or run it from cron.weekly on /usr/share/doc ;) God please no. We already have too many cron jobs that turn a machine into a slug. I think it is 100% wrong to mark files as duplicates because they are the same "now". There is no guarantee they will be the same in a future update. Excluding a GPL license COPYING file from one package and linking it to another central copy fails the second someone decides to use GPLv3 for that package. Or if they add text to the top of the file or something. It is bad practice to be doing this with license files, for almost zero gain. A few months ago Warren scanned the entire OS to see what would be gained if we were to do what is being proposed here. The results were negligible. Lets take 10 steps back and try to see the forest for a minute, ignoring the trees for the time being. If the goal is to make the distribution take up less space - lets focus on analyzing what exactly is taking up the most space on the distribution after install. Find the top 10 space consumers, and begin analyzing how we might be able to reduce the space they're using. I suspect solving that high-level problem will result in a disk space savings 10-20 times any savings we might gain from hardlinking GPL "COPYING" files. Let's focus on the real problem rather than coming up with solutions and then looking for problems we can solve with them. From pjones at redhat.com Mon May 2 20:12:50 2005 From: pjones at redhat.com (Peter Jones) Date: Mon, 02 May 2005 16:12:50 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <4276825D.7090204@redhat.com> References: <1115037486.29438.54.camel@obelix.redhat.usu> <42762272.1080102@hhs.nl> <1115061550.12093.0.camel@localhost.localdomain> <20050502192346.GA19481@devserv.devel.redhat.com> <4276825D.7090204@redhat.com> Message-ID: <1115064770.12093.35.camel@localhost.localdomain> On Mon, 2005-05-02 at 15:41 -0400, Mike A. Harris wrote: > I think it is 100% wrong to mark files as duplicates because > they are the same "now". There is no guarantee they will be > the same in a future update. That doesn't matter, updating has the nice property of running unlink(2) on it if it's not supposed to be the same any more. But I think the whole problem is silly as well, FWIW. -- Peter From pjones at redhat.com Mon May 2 20:16:11 2005 From: pjones at redhat.com (Peter Jones) Date: Mon, 02 May 2005 16:16:11 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <200505021928.j42JSB1F022621@magilla.sf.frob.com> References: <200505021928.j42JSB1F022621@magilla.sf.frob.com> Message-ID: <1115064971.12093.38.camel@localhost.localdomain> On Mon, 2005-05-02 at 12:28 -0700, Roland McGrath wrote: > I think what one clearly wants is for rpm to maintain an installed file > indexed keyed by md5sum. Then you can have a tool that just uses this > database to identify duplicates (and doesn't take forever), or have rpm do > so itself when installing new files. You already have sufficient data in the rpmdb, the problem is you've either got to: a) change rpm to do all the work itself, and probably make this an option you can override with a either your environment/macros or some flag in the package -- the rules there get complicated quickly... or b) make it possible and safe to query the rpm db from a %post. Either of those is a bit "ewww". -- Peter From pjones at redhat.com Mon May 2 20:25:57 2005 From: pjones at redhat.com (Peter Jones) Date: Mon, 02 May 2005 16:25:57 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <200505021935.j42JZqNm022649@magilla.sf.frob.com> References: <200505021935.j42JZqNm022649@magilla.sf.frob.com> Message-ID: <1115065557.12093.48.camel@localhost.localdomain> On Mon, 2005-05-02 at 12:35 -0700, Roland McGrath wrote: > > Roland McGrath wrote: > > > I think what one clearly wants is for rpm to maintain an installed file > > > indexed keyed by md5sum. Then you can have a tool that just uses this > > > database to identify duplicates (and doesn't take forever), or have rpm do > > > so itself when installing new files. > > > > > > > Hmm, what about hash collisions, that would be really really BAD > > If you are concerned about them you can still compare contents before > declaring two files identical. But using the hashes as the main detector > makes it fast, since you only examine the data of files that are 99.999% > likely to be identical. And in the vast majority of cases, there's a simpler heuristic you can use first: is the basename the same? But really, this is 160MB of wasted space. We don't support installing onto USB, so from glancing at pricewatch, the smallest disk they list that we support installing onto would appear to be an 18GB SCSI drive for $23. There are larger, cheaper drives, too. So we're talking about saving just under 1% of the least-desirable supported install target currently being sold. Let's just stop? -- Peter From roland at redhat.com Mon May 2 20:27:08 2005 From: roland at redhat.com (Roland McGrath) Date: Mon, 2 May 2005 13:27:08 -0700 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: Peter Jones's message of Monday, 2 May 2005 16:12:50 -0400 <1115064770.12093.35.camel@localhost.localdomain> Message-ID: <200505022027.j42KR8mM022835@magilla.sf.frob.com> > But I think the whole problem is silly as well, FWIW. When Warren brought this up on IRC a while back, I wrote the following script and rand it on a rawhide everything install. This fails to take into account files that are already hardlinked, so and its results might well be significantly inflated. (Someone who cares could hack it further to check installed names of a duplicate file for being the same inode.) Total 408578931 bytes in 43107 inodes That's a max of < 400M on an install that is something 8.5-9G. So the issue is worth at most on the order of 5% of disk space, and that is probably a very high estimate. rpm -qa --qf '[%{FILEMD5S} %{FILENAMES} %{FILESIZES} %{SOURCERPM}\n]' | awk ' NF < 4 { next } # directory { md5_name[$1] = $2; md5_srpm[$1] = $4; info = $2 " " $4; if ($1 in sizes) { if ($3 != sizes[$1]) print "!!!", $1 ":", info, "VS", md5[info] } else { sizes[$1] = $3; } if ($1 in md5) { if (info == md5[$1]) next; for (i = 1; i < dups[$1]; ++i) if (dupinfo[$1 "," i] == info) next; dups[$1]++; dupinfo[$1 "," dups[$1]] = info; } else { md5[$1] = info; } } END { dupsize = dupcount = 0; for (sum in dups) { n = dups[sum]; dupcount += n; dupsize += n * sizes[sum]; print n, "dups:", sum, " ==> ", (n * sizes[sum]); print "\t" md5[sum]; for (i = 1; i <= n; ++i) print "\t" dupinfo[sum "," i]; } print "Total", dupsize, "bytes in", dupcount, "inodes"; } ' From rgammon at real.com Mon May 2 20:34:07 2005 From: rgammon at real.com (Ryan Gammon) Date: Mon, 02 May 2005 13:34:07 -0700 Subject: OS platform for Helix Player 2.0 Message-ID: <42768EBF.1090307@real.com> Hi guys, I sent the attached email out to player-dev at hc.org re: the platform for balto (helix player 2.0). Feedback appreciated, particularly if this is going to cause probems for your distribution. Please include the player-dev at helixcommunity.org list if possible. 2.0's coming out later in the year. -- Ryan Gammon rgammon at real.com Developer for Helix Player https://player.helixcommunity.org -------------- next part -------------- An embedded message was scrubbed... From: Ryan Gammon Subject: OS platform for Helix Player 2.0 Date: Mon, 02 May 2005 13:18:34 -0700 Size: 1147 URL: From roland at redhat.com Mon May 2 20:41:00 2005 From: roland at redhat.com (Roland McGrath) Date: Mon, 2 May 2005 13:41:00 -0700 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: Peter Jones's message of Monday, 2 May 2005 16:16:11 -0400 <1115064971.12093.38.camel@localhost.localdomain> Message-ID: <200505022041.j42Kf0Vi022871@magilla.sf.frob.com> > a) change rpm to do all the work itself, and probably make this an > option you can override with a either your environment/macros or some > flag in the package -- the rules there get complicated quickly... This definitely gets you into a dicey area, doing this automatically in rpm. You certainly don't want it done for config files, but one should be at least moderately afraid of linking together two files that the user winds up modifying one of and not expecting to have modified the other. > or b) make it possible and safe to query the rpm db from a %post. I don't really know what the issues with this are. or c) do it in a separate pass later for those who care. This has the nice property that noone gets it who didn't ask for it, and so the a) quagmire of options is only a burden to those who want to deal with it. Not being built in to rpm or any such thing, it also has the nice property that whoever cares can go off and do it on their own, put it in Extras and have users love it and clamor for it in Core, rather than arguing in the abstract about whether it's worth doing. Doing this as the occasional cron job is pretty reasonable (if you care to do it at all). I mean, my frickin' awk script only takes a few minutes for 1917 installed rpms, so a real implementation in C using rpmlib directly should be pretty quick. The other wrinkle about doing this is the file attributes. Pretty much all the copies of COPYING have the same owner and permissions, but each one has a different modtime. (If files differ in attributes other than modtime, you shouldn't merge them into a single inode anyway.) To make rpm -V happy, the tool would have to update the rpmdb to make modtimes in the entries for duplicate names all match. This won't fix rpm -Vp complaints. For that, we would have to harmonize all the rpm contents across the distribution. For example, the tree-composing crapola could find the duplicates and tweak all the modtimes in the rpms to match. This seems like a pretty ludicrous amount of effort for the benefit it has. From Axel.Thimm at ATrpms.net Mon May 2 22:05:37 2005 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Tue, 3 May 2005 00:05:37 +0200 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <4275495E.7080901@redhat.com> References: <4275495E.7080901@redhat.com> Message-ID: <20050502220537.GA11448@neu.nirvana> On Sun, May 01, 2005 at 11:25:50AM -1000, Warren Togami wrote: > Unfortunately we made a mistake with the original FC3 x86_64 and shipped > i386 perl by accident when it was never meant to be multilib. This > causes updates to clash and fail on FC3 x86_64. It is wrong to add i386 > perl to that repository, so the only way to recover is to use: > > yum remove perl.i386 Aren't these kind of bugs things you have to live with for the rest of the distro's EOL? E.g. keep perl multilib in FC3, have it fixed in FC4 (and its anaconda) and bite the bullet and provide i386 updates for another 6 months? > We are deeply sorry about this inconvenience and will try to avoid this > from ever happening again in the future. Fixing bugs by introducing bigger ones is bad. How many users will really be notified and act accordingly? This will just add more stress to supporting entities like endless threads on mailing lists rendering them even more useless. -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From katzj at redhat.com Mon May 2 22:21:40 2005 From: katzj at redhat.com (Jeremy Katz) Date: Mon, 02 May 2005 18:21:40 -0400 Subject: Request for Testing of Installer X Configuration on PPC Message-ID: <1115072500.5982.9.camel@bree.local.net> Hi. I've got an ISO for quick testing of anaconda's X configuration setup available that's a little bit smaller than a full tree. All you should have to do is download the ISO, burn it to a CD, boot and see if X starts up. Then, you can reboot and go back to your normal doings. Note that this is entirely non-destructive. So you can even run it if you are running OS X. *ANY* results would be helpful. Please be sure to include what you see happen as well as what type of hardware you're using (cpu, video, monitor). The iso is located at http://people.redhat.com/~katzj/mac-x-test.iso. The md5sum is 34234b3cdf2d89e4e94031a50f819285 Thanks in advance for testing this and helping to make test3 (and thus, the Fedora Core 4 release) better on PPC hardware! Jeremy From wtogami at redhat.com Mon May 2 22:31:49 2005 From: wtogami at redhat.com (Warren Togami) Date: Mon, 02 May 2005 12:31:49 -1000 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <20050502220537.GA11448@neu.nirvana> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> Message-ID: <4276AA55.5060101@redhat.com> Axel Thimm wrote: > On Sun, May 01, 2005 at 11:25:50AM -1000, Warren Togami wrote: > >>Unfortunately we made a mistake with the original FC3 x86_64 and shipped >>i386 perl by accident when it was never meant to be multilib. This >>causes updates to clash and fail on FC3 x86_64. It is wrong to add i386 >>perl to that repository, so the only way to recover is to use: >> >>yum remove perl.i386 > > > Aren't these kind of bugs things you have to live with for the rest of > the distro's EOL? E.g. keep perl multilib in FC3, have it fixed in FC4 > (and its anaconda) and bite the bullet and provide i386 updates for > another 6 months? https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156658 Anaconda upgrade should remove perl.i386 is filed here. It wont be an easy fix unfortunately. From Axel.Thimm at ATrpms.net Mon May 2 22:40:18 2005 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Tue, 3 May 2005 00:40:18 +0200 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <4276AA55.5060101@redhat.com> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> Message-ID: <20050502224018.GA12248@neu.nirvana> On Mon, May 02, 2005 at 12:31:49PM -1000, Warren Togami wrote: > Axel Thimm wrote: > >On Sun, May 01, 2005 at 11:25:50AM -1000, Warren Togami wrote: > > > >>Unfortunately we made a mistake with the original FC3 x86_64 and shipped > >>i386 perl by accident when it was never meant to be multilib. This > >>causes updates to clash and fail on FC3 x86_64. It is wrong to add i386 > >>perl to that repository, so the only way to recover is to use: > >> > >>yum remove perl.i386 > > > > > >Aren't these kind of bugs things you have to live with for the rest of > >the distro's EOL? E.g. keep perl multilib in FC3, have it fixed in FC4 > >(and its anaconda) and bite the bullet and provide i386 updates for > >another 6 months? > > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156658 > Anaconda upgrade should remove perl.i386 is filed here. It wont be an > easy fix unfortunately. That still leaves FC3 (and RHEL4 according to your bug report) broken w/o explicit user intervention. The fix cannot be to hope users will read (and remember) this announcement. -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From wtogami at redhat.com Mon May 2 22:42:55 2005 From: wtogami at redhat.com (Warren Togami) Date: Mon, 02 May 2005 12:42:55 -1000 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <20050502224018.GA12248@neu.nirvana> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> Message-ID: <4276ACEF.50001@redhat.com> Axel Thimm wrote: > > That still leaves FC3 (and RHEL4 according to your bug report) broken > w/o explicit user intervention. The fix cannot be to hope users will > read (and remember) this announcement. > Find a better solution for existing users if you think you can do better, then convince notting. Warren From arjanv at redhat.com Mon May 2 23:11:34 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Mon, 02 May 2005 19:11:34 -0400 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <4276ACEF.50001@redhat.com> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> <4276ACEF.50001@redhat.com> Message-ID: <1115075495.6501.0.camel@localhost.localdomain> On Mon, 2005-05-02 at 12:42 -1000, Warren Togami wrote: > Axel Thimm wrote: > > > > That still leaves FC3 (and RHEL4 according to your bug report) broken > > w/o explicit user intervention. The fix cannot be to hope users will > > read (and remember) this announcement. > > > > Find a better solution for existing users if you think you can do > better, then convince notting. the solution is to have an arch "versioned" obsoletes: and have the x86-64 perl obsolete the i386 one in the rpm :) and no I don't volunteer ;) -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From Axel.Thimm at ATrpms.net Mon May 2 23:16:59 2005 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Tue, 3 May 2005 01:16:59 +0200 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <4276ACEF.50001@redhat.com> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> <4276ACEF.50001@redhat.com> Message-ID: <20050502231659.GC12248@neu.nirvana> On Mon, May 02, 2005 at 12:42:55PM -1000, Warren Togami wrote: > Axel Thimm wrote: > > > >That still leaves FC3 (and RHEL4 according to your bug report) broken > >w/o explicit user intervention. The fix cannot be to hope users will > >read (and remember) this announcement. > > > > Find a better solution for existing users if you think you can do > better, then convince notting. So, why did you trim my proposed solution "live-with-it" in your reply? -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skvidal at phy.duke.edu Tue May 3 00:05:45 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Mon, 02 May 2005 20:05:45 -0400 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <1115075495.6501.0.camel@localhost.localdomain> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> <4276ACEF.50001@redhat.com> <1115075495.6501.0.camel@localhost.localdomain> Message-ID: <1115078745.7569.21.camel@cutter> On Mon, 2005-05-02 at 19:11 -0400, Arjan van de Ven wrote: > On Mon, 2005-05-02 at 12:42 -1000, Warren Togami wrote: > > Axel Thimm wrote: > > > > > > That still leaves FC3 (and RHEL4 according to your bug report) broken > > > w/o explicit user intervention. The fix cannot be to hope users will > > > read (and remember) this announcement. > > > > > > > Find a better solution for existing users if you think you can do > > better, then convince notting. > > > the solution is to have an arch "versioned" obsoletes: and have the > x86-64 perl obsolete the i386 one in the rpm :) so you want to have a way of saying "this packages obsoletes that package, but only if that package is that arch"? let me offer a profound 'ugh' -sv From arjanv at redhat.com Tue May 3 00:08:29 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Tue, 3 May 2005 02:08:29 +0200 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <1115078745.7569.21.camel@cutter> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> <4276ACEF.50001@redhat.com> <1115075495.6501.0.camel@localhost.localdomain> <1115078745.7569.21.camel@cutter> Message-ID: <20050503000828.GA23194@devserv.devel.redhat.com> On Mon, May 02, 2005 at 08:05:45PM -0400, seth vidal wrote: > On Mon, 2005-05-02 at 19:11 -0400, Arjan van de Ven wrote: > > On Mon, 2005-05-02 at 12:42 -1000, Warren Togami wrote: > > > Axel Thimm wrote: > > > > > > > > That still leaves FC3 (and RHEL4 according to your bug report) broken > > > > w/o explicit user intervention. The fix cannot be to hope users will > > > > read (and remember) this announcement. > > > > > > > > > > Find a better solution for existing users if you think you can do > > > better, then convince notting. > > > > > > the solution is to have an arch "versioned" obsoletes: and have the > > x86-64 perl obsolete the i386 one in the rpm :) > > so you want to have a way of saying "this packages obsoletes that > package, but only if that package is that arch"? Obsoletes: perl = 5.8.1-1.4.1(i386) or something > > let me offer a profound 'ugh' thanks ;) From perbj at stanford.edu Tue May 3 00:26:42 2005 From: perbj at stanford.edu (Per Bjornsson) Date: Mon, 02 May 2005 17:26:42 -0700 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <20050503000828.GA23194@devserv.devel.redhat.com> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> <4276ACEF.50001@redhat.com> <1115075495.6501.0.camel@localhost.localdomain> <1115078745.7569.21.camel@cutter> <20050503000828.GA23194@devserv.devel.redhat.com> Message-ID: <1115080002.4763.8.camel@localhost.localdomain> On Tue, 2005-05-03 at 02:08 +0200, Arjan van de Ven wrote: > > so you want to have a way of saying "this packages obsoletes that > > package, but only if that package is that arch"? > > Obsoletes: perl = 5.8.1-1.4.1(i386) > or something Well actually, isn't it OK if it obsoletes _all_ perl-5.8.1-1.4.1 versions, both i386 and x86_64? You want to upgrade the package, not parallel-install after all. So what does obsolete do now in terms of multilib? Will it only obsolete the same architecture, or all arches? /Per -- Per Bjornsson Ph.D. Candidate, Department of Applied Physics, Stanford University From skvidal at phy.duke.edu Tue May 3 00:30:29 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Mon, 02 May 2005 20:30:29 -0400 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <1115080002.4763.8.camel@localhost.localdomain> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> <4276ACEF.50001@redhat.com> <1115075495.6501.0.camel@localhost.localdomain> <1115078745.7569.21.camel@cutter> <20050503000828.GA23194@devserv.devel.redhat.com> <1115080002.4763.8.camel@localhost.localdomain> Message-ID: <1115080229.7569.23.camel@cutter> On Mon, 2005-05-02 at 17:26 -0700, Per Bjornsson wrote: > On Tue, 2005-05-03 at 02:08 +0200, Arjan van de Ven wrote: > > > so you want to have a way of saying "this packages obsoletes that > > > package, but only if that package is that arch"? > > > > Obsoletes: perl = 5.8.1-1.4.1(i386) > > or something > > Well actually, isn't it OK if it obsoletes _all_ perl-5.8.1-1.4.1 > versions, both i386 and x86_64? You want to upgrade the package, not > parallel-install after all. So what does obsolete do now in terms of > multilib? Will it only obsolete the same architecture, or all arches? that depends on what version of rpm you're using. -sv From mharris at www.linux.org.uk Tue May 3 00:35:34 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Mon, 02 May 2005 20:35:34 -0400 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115065557.12093.48.camel@localhost.localdomain> References: <200505021935.j42JZqNm022649@magilla.sf.frob.com> <1115065557.12093.48.camel@localhost.localdomain> Message-ID: <4276C756.4080708@www.linux.org.uk> Peter Jones wrote: > On Mon, 2005-05-02 at 12:35 -0700, Roland McGrath wrote: > >>>Roland McGrath wrote: >>> >>>>I think what one clearly wants is for rpm to maintain an installed file >>>>indexed keyed by md5sum. Then you can have a tool that just uses this >>>>database to identify duplicates (and doesn't take forever), or have rpm do >>>>so itself when installing new files. >>>> >>> >>>Hmm, what about hash collisions, that would be really really BAD >> >>If you are concerned about them you can still compare contents before >>declaring two files identical. But using the hashes as the main detector >>makes it fast, since you only examine the data of files that are 99.999% >>likely to be identical. > > > And in the vast majority of cases, there's a simpler heuristic you can > use first: is the basename the same? > > But really, this is 160MB of wasted space. We don't support installing > onto USB, so from glancing at pricewatch, the smallest disk they list > that we support installing onto would appear to be an 18GB SCSI drive > for $23. There are larger, cheaper drives, too. > > So we're talking about saving just under 1% of the least-desirable > supported install target currently being sold. Let's just stop? Seconded. The time wasted on the thread would be better spent if people would look at the top 10 real space wasters and tried to find solutions to those. Even then, personally I believe if we could cut the OS full install from 9Gb down to 1Gb -> who cares. Like you just stated, disks are massive nowadays, and cheap to boot. If there are people out there who cant or don't want to buy a new disk, and the install size is stopping them from using Fedora, then again - finding the _biggest_ space wasters and trying to resolve them instead of wasting time talking peanuts of duplicated files would be a bigger saver. I can think of several things people could volunteer for to make install footprint smaller: - Join X.Org Development and help keithp and juliusz get bitmap fonts converted from .bdf/pcf to .ttf bitmaps, and finish off the tools needed to make this happen. That shaves off quite a number of megs off the CDROMs in theory, although the same holds true as above as far as disk prices are concerned. - Examine packages for overzealous rpm dependancies. This includes finding things that link to libraries that aren't used, causing unnecessary deps. Finding and fixing those will help lower install footprint, although it wont save CD space. - Find things we can just totally throw away completely, or move to Fedora Extras, or some other repository. Things that really do not need to be in "Core" for general purpose OS. Since this is a touchy subject for many who want their favourite apps to be included in the "Core" by default for convenience, I'll suggest some that I personally use myself that I'd like to see remain in core, but which probably don't belong there: "mc", "iptraf", "pinfo". - Examine large packages like openoffice, X, and others to see what things consume the most space, determine wether there might be a better short or long term way to package the stuff, or change the way the underlying technology works. etc. From david at fubar.dk Tue May 3 00:51:57 2005 From: david at fubar.dk (David Zeuthen) Date: Mon, 02 May 2005 20:51:57 -0400 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115072500.5982.9.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> Message-ID: <1115081517.3398.7.camel@daxter.boston.redhat.com> On Mon, 2005-05-02 at 18:21 -0400, Jeremy Katz wrote: > Hi. I've got an ISO for quick testing of anaconda's X configuration > setup available that's a little bit smaller than a full tree. All you > should have to do is download the ISO, burn it to a CD, boot and see if > X starts up. Then, you can reboot and go back to your normal doings. > > Note that this is entirely non-destructive. So you can even run it if > you are running OS X. *ANY* results would be helpful. Please be sure > to include what you see happen as well as what type of hardware you're > using (cpu, video, monitor). > > The iso is located at http://people.redhat.com/~katzj/mac-x-test.iso. > The md5sum is 34234b3cdf2d89e4e94031a50f819285 > > Thanks in advance for testing this and helping to make test3 (and thus, > the Fedora Core 4 release) better on PPC hardware! OK, I tried this on my Powerbook G4 12" 867MHz (the 1st generation PB12) and X starts fine in 800x600x16 bit using the nv driver. Probing for video card: NVIDIA GeForce 4 (generic) Probing for monitor type: Unknown monitor Probing for mouse type: Generic - 3 Button Mouse (PS/2) Great work. Cheers, David From perbj at stanford.edu Tue May 3 01:49:23 2005 From: perbj at stanford.edu (Per Bjornsson) Date: Mon, 02 May 2005 18:49:23 -0700 Subject: (perl update clash) FC3 x86_64 Important Announcement In-Reply-To: <1115080229.7569.23.camel@cutter> References: <4275495E.7080901@redhat.com> <20050502220537.GA11448@neu.nirvana> <4276AA55.5060101@redhat.com> <20050502224018.GA12248@neu.nirvana> <4276ACEF.50001@redhat.com> <1115075495.6501.0.camel@localhost.localdomain> <1115078745.7569.21.camel@cutter> <20050503000828.GA23194@devserv.devel.redhat.com> <1115080002.4763.8.camel@localhost.localdomain> <1115080229.7569.23.camel@cutter> Message-ID: <1115084963.4763.12.camel@localhost.localdomain> On Mon, 2005-05-02 at 20:30 -0400, seth vidal wrote: > On Mon, 2005-05-02 at 17:26 -0700, Per Bjornsson wrote: > > Well actually, isn't it OK if it obsoletes _all_ perl-5.8.1-1.4.1 > > versions, both i386 and x86_64? You want to upgrade the package, not > > parallel-install after all. So what does obsolete do now in terms of > > multilib? Will it only obsolete the same architecture, or all arches? > > that depends on what version of rpm you're using. Yay. Well, the Perl problem at hand concerns FC3, right? 4.3.2 then (rpm-4.3.2-21). I can't seem to figure out which versions do what... /Per -- Per Bjornsson Ph.D. Candidate, Department of Applied Physics, Stanford University From gauravp at hclcomnet.co.in Tue May 3 06:37:19 2005 From: gauravp at hclcomnet.co.in (gaurav) Date: Tue, 03 May 2005 12:07:19 +0530 Subject: Stateless Debian Project Message-ID: <42771C1F.4040206@hclcomnet.co.in> Hi, We have just initiated Stateless Debian Project and we are looking for active volunteers/developers Detail of project are given below Summary This project was started by Fedora (Redhat) but is no longer in active development and will not be included in next release of fc4 , this projects aims to provide longterm commitment to the concept & port it to 100% community based distro like Debian Stateless Linux converts normal Linux desktop/clients to Stateless machines or appliances, which means if throw your computer out of window you still will be able to get exactly same same settings/data when you log from any other pc in the network ....A single administrator can easily manage network thousands of desktops ...Stateless Linux centralizes the state in a Gold server (different from CFengine) and rest of clients are updated regularly from it . This is different from thin clients as local processing power and memory of clients is used (or cached client) http://sourceforge.net/projects/statelessdebian/ Project Goals 1) Port relevant parts of RH State Linux (http://fedora.redhat.com/projects/stateless/) to Debian 2)Extend/integrate/enhance Stateless Linux thru existing projects like DBRL http://drbl.sourceforge.net/debian/wiki-view/pmwiki-view.php/DRBL/WhatIsDRBL and Debian CCD infrastructure, FAI (Fully Automated Installation,http://www.informatik.uni-koeln.de/fai/) 3) Add cluster support and node monitoring and autonomic capabilities 4)Meets most of OSDL Desktop Linux specification 5) Long term community based , vendor neutral project 6) Member of "Move 2 Debian" project 7) Latter (2nd phase )Add p2p support for directory support(DHT), file storage and retrieval(tuples), state transfer (Bittorent) to elinimate any need of centralized servers/infrastructure and make it highly scalable (to millions!!) Right now this project is in pre beta stage and we are preparing Stateless Linux White paper and started porting portions of RH Stateless Linux to Debian . Any Suggestions/Ideas will appreciated :-) ..if anyone is interested from this list then pl let me know Regards, Gaurav From s.mako at gmx.net Tue May 3 07:01:18 2005 From: s.mako at gmx.net (Zoltan Kota) Date: Tue, 3 May 2005 09:01:18 +0200 (CEST) Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115072500.5982.9.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> Message-ID: On Mon, 2 May 2005, Jeremy Katz wrote: > Hi. I've got an ISO for quick testing of anaconda's X configuration > setup available that's a little bit smaller than a full tree. All you > should have to do is download the ISO, burn it to a CD, boot and see if > X starts up. Then, you can reboot and go back to your normal doings. It doesn't really boot for me. iMac PowerPC G5, Nvidia GeForce FX 5200, (LCD 1440x900) ---- boot: linux ... loading ramdisk ... ok opening display /pci at 0,f0000000/NVDA,Parent at 10/NVDA,Display-B at 1 ... ok copying OF device tree ... done Initializing fake screen: NVDA,Display-A Calling quiesce... returning 0x01400000 from prom_init Invalid memory access at &SRR0: 00000000.01403b7c &SRR1:10000000.00083030 Apple PowerMac8, 1 5.2.2f2 BootRom built on ... Copyright... All Rights... Welcome to Open Firmware, ... To continue booting, type "mac-boot" and press return. To shut down, type "shut-down" and press return. Release keys to continue! ---- Zoltan From dwmw2 at infradead.org Tue May 3 08:09:31 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 03 May 2005 09:09:31 +0100 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: References: <1115072500.5982.9.camel@bree.local.net> Message-ID: <1115107772.8508.25.camel@localhost.localdomain> On Tue, 2005-05-03 at 09:01 +0200, Zoltan Kota wrote: > It doesn't really boot for me. > iMac PowerPC G5, Nvidia GeForce FX 5200, (LCD 1440x900) That's a kernel issue -- the 32-bit kernel doesn't run on the G5; you need the 64-bit kernel there. Ideally we'd have the bootloader detect the machine type and automatically pick the right kernel. At the moment, though, you need a separate boot.iso for the G5. I'm not sure if Jeremy built a 64-bit version. -- dwmw2 From wbeebe at gmail.com Tue May 3 11:50:05 2005 From: wbeebe at gmail.com (William Beebe) Date: Tue, 3 May 2005 07:50:05 -0400 Subject: Stateless Debian Project In-Reply-To: <42771C1F.4040206@hclcomnet.co.in> References: <42771C1F.4040206@hclcomnet.co.in> Message-ID: How about putting this boundless energy into more timely releases of Debian? And did you ever stop to think that if stateless was dropped that there was a good reason to drop it? Why does Debian think it has to become the technological rat hole for abandoned fringe projects? Oh, I forgot. Debian is a fringe project. On 5/3/05, gaurav wrote: > > > Hi, > > We have just initiated Stateless Debian Project and we are looking for > active volunteers/developers > Detail of project are given below > > Summary > This project was started by Fedora (Redhat) but is no longer in active > development and will not be included in next release of fc4 , this > projects aims to provide longterm commitment to the concept & port it to > 100% community based distro like Debian > > Stateless Linux converts normal Linux desktop/clients to Stateless > machines or appliances, which means if throw your computer out of window > you still will be able to get exactly same same settings/data when you log > from any other pc in the network ....A single administrator can easily > manage network thousands of desktops ...Stateless Linux centralizes the > state in a Gold server (different from CFengine) and rest of clients > are updated regularly from it . This is different from thin clients as > local processing power and memory of clients is used (or cached client) > > http://sourceforge.net/projects/statelessdebian/ > Project Goals > > 1) Port relevant parts of RH State Linux (http://fedora.redhat.com/projects/stateless/) to Debian > 2)Extend/integrate/enhance Stateless Linux thru existing projects like DBRL > http://drbl.sourceforge.net/debian/wiki-view/pmwiki-view.php/DRBL/WhatIsDRBL > and Debian CCD infrastructure, FAI (Fully Automated Installation,http://www.informatik.uni-koeln.de/fai/) > 3) Add cluster support and node monitoring and autonomic capabilities > 4)Meets most of OSDL Desktop Linux specification > 5) Long term community based , vendor neutral project > 6) Member of "Move 2 Debian" project > 7) Latter (2nd phase )Add p2p support for directory support(DHT), file storage and retrieval(tuples), state transfer (Bittorent) > to elinimate any need of centralized servers/infrastructure and make it highly scalable (to millions!!) > > Right now this project is in pre beta stage and we are preparing Stateless Linux White paper and started porting portions of RH Stateless Linux to Debian . > Any Suggestions/Ideas will appreciated :-) ..if anyone is interested from this list then pl let me know > > Regards, > Gaurav > > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list > From arjanv at redhat.com Tue May 3 11:58:54 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Tue, 03 May 2005 07:58:54 -0400 Subject: Stateless Debian Project In-Reply-To: References: <42771C1F.4040206@hclcomnet.co.in> Message-ID: <1115121534.6077.10.camel@localhost.localdomain> On Tue, 2005-05-03 at 07:50 -0400, William Beebe wrote: > How about putting this boundless energy into more timely releases of > Debian? And did you ever stop to think that if stateless was dropped > that there was a good reason to drop it? Why does Debian think it has > to become the technological rat hole for abandoned fringe projects? > Oh, I forgot. Debian is a fringe project. please consider responding in a more respectful and adult manner. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From buildsys at redhat.com Tue May 3 12:10:31 2005 From: buildsys at redhat.com (Build System) Date: Tue, 3 May 2005 08:10:31 -0400 Subject: rawhide report: 20050503 changes Message-ID: <200505031210.j43CAVwY027741@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.3-20050426.134031.FC4.9 ----------------------------------------- anaconda-10.2.0.58-1 -------------------- * Mon May 02 2005 Jeremy Katz - 10.2.0.58-1 - Beep on CD insertion, not after - Fix language support selection (clumens) - Fix nfsiso (clumens) - Misc X config fixes for ppc. Boot with "usefbx" to use fbdev instead again (#149188) chkconfig-1.3.19-1 ------------------ * Fri Apr 29 2005 Bill Nottingham 1.3.19-1 - build with updated translations cman-kernel-2.6.11.3-20050425.154843.FC4.5 ------------------------------------------ comps-extras-10.2-1 ------------------- * Mon May 02 2005 Bill Nottingham - 10.2-1 - add some icons dlm-kernel-2.6.11.3-20050425.154843.FC4.6 ----------------------------------------- gnbd-kernel-2.6.11.2-20050420.133124.FC4.10 ------------------------------------------- initscripts-8.10-1 ------------------ * Fri Apr 29 2005 Bill Nottingham 8.10-1 - fix hang on stale GDM sockets (#156355) kernel-2.6.11-1.1282_FC4 ------------------------ * Sun May 01 2005 Dave Jones - Fix yesterdays IDE fixes. - Blacklist another brainless SCSI scanner. (#155457) * Sun May 01 2005 David Woodhouse - Fix EHCI port power switching * Sun May 01 2005 Dave Jones - Enable usbmon & debugfs. (#156489) kudzu-1.1.116-1 --------------- * Fri Apr 29 2005 Bill Nottingham 1.1.116-1 - fix nforce4 fix - some readlink() usage fixes perl-3:5.8.6-10 --------------- * Thu Apr 28 2005 Ville Skytt?? - 3:5.8.6-10 - Apply fixes for CAN-2004-0452, CAN-2005-0155 and CAN-2005-0156 (#156128). * Tue Apr 26 2005 Warren Togami - 3:5.8.6-7 - Updating CGI.pm from version 3.05 to 3.08 (mod_perl 2.0.0 RC5). (#155839) rhpl-0.161-1 ------------ * Mon May 02 2005 Jeremy Katz - 0.161-1 - Only default to fbdev X server on ppc64pseries, not all ppc (#149188) - Set UseFBDev on ppc boxes using ati video cards (#149188) samba-0:3.0.14a-2 ----------------- * Mon May 02 2005 Jay Fenlason 3.0.14a-2 - New upstream release. - the -64bit-timestamps, -clitar, -establish_trust, user_rights_v1, winbind_find_dc_v2 patches are now obsolete. * Thu Apr 07 2005 Jay Fenlason 3.0.13-2 - New upstream release - add my -quoting patch, to fix swat with strings that contain html meta-characters, and to use correct quote characters in lists, closing bz#134310 - include the upstream winbindd_2k3sp1 patch - include the -smbclient patch. - include the -hang patch from upstream. selinux-policy-strict-1.23.14-2 ------------------------------- * Mon May 02 2005 Dan Walsh 1.23.14-2 - Allow all domains on ppc execmem priv, otherwise it crashes * Mon May 02 2005 Dan Walsh 1.23.14-1 - Update to latest from NSA * Added afs policy from Andrew Reisse. * Merged patch from Lorenzo Hern??ndez Garc??a-Hierro which defines execstack and execheap permissions. The patch excludes these permissions from general_domain_access and updates the macros for X, legacy binaries, users, and unconfined domains. * Added nlmsg_relay permisison where netlink_audit_socket class is used. Added nlmsg_readpriv permission to auditd_t and auditctl_t. * Merged some minor cleanups from Russell Coker and David Hampton. * Merged patch from Dan Walsh. Many changes made to allow targeted policy to run closer to strict and now almost all of non-userspace is protected via SELinux. Kernel is now in unconfined_domain for targeted and runs as root:system_r:kernel_t. Added transitionbool to daemon_sub_domain, mainly to turn off httpd_suexec transitioning. Implemented web_client_domain name_connect rules. Added yp support for cups. Now the real hotplug, udev, initial_sid_contexts are used for the targeted policy. Other minor cleanups and fixes. Auditd fixes by Paul Moore. * Fri Apr 29 2005 Dan Walsh 1.23.13-8 - Fixes for consoletype, kudzu reading proc_t - Add label /dev/adb - Fixes for hal selinux-policy-targeted-1.23.14-2 --------------------------------- * Mon May 02 2005 Dan Walsh 1.23.14-2 - Allow all domains on ppc execmem priv, otherwise it crashes * Mon May 02 2005 Dan Walsh 1.23.14-1 - Update to latest from NSA * Added afs policy from Andrew Reisse. * Merged patch from Lorenzo Hern??ndez Garc??a-Hierro which defines execstack and execheap permissions. The patch excludes these permissions from general_domain_access and updates the macros for X, legacy binaries, users, and unconfined domains. * Added nlmsg_relay permisison where netlink_audit_socket class is used. Added nlmsg_readpriv permission to auditd_t and auditctl_t. * Merged some minor cleanups from Russell Coker and David Hampton. * Merged patch from Dan Walsh. Many changes made to allow targeted policy to run closer to strict and now almost all of non-userspace is protected via SELinux. Kernel is now in unconfined_domain for targeted and runs as root:system_r:kernel_t. Added transitionbool to daemon_sub_domain, mainly to turn off httpd_suexec transitioning. Implemented web_client_domain name_connect rules. Added yp support for cups. Now the real hotplug, udev, initial_sid_contexts are used for the targeted policy. Other minor cleanups and fixes. Auditd fixes by Paul Moore. * Fri Apr 29 2005 Dan Walsh 1.23.13-8 - Fixes for consoletype, kudzu reading proc_t - Add label /dev/adb - Fixes for hal From dwmw2 at infradead.org Tue May 3 12:12:15 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 03 May 2005 13:12:15 +0100 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115072500.5982.9.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> Message-ID: <1115122336.4446.4.camel@hades.cambridge.redhat.com> On Mon, 2005-05-02 at 18:21 -0400, Jeremy Katz wrote: > Hi. I've got an ISO for quick testing of anaconda's X configuration > setup available that's a little bit smaller than a full tree. All you > should have to do is download the ISO, burn it to a CD, boot and see > if > X starts up. Then, you can reboot and go back to your normal doings. > > Note that this is entirely non-destructive. So you can even run it if > you are running OS X. *ANY* results would be helpful. Please be sure > to include what you see happen as well as what type of hardware you're > using (cpu, video, monitor). > > The iso is located at http://people.redhat.com/~katzj/mac-x-test.iso. > The md5sum is 34234b3cdf2d89e4e94031a50f819285 PowerBook1,1 (Rage 128): Works fine. PowerBook5,3 (Radeon 9600): Kills machine hard. The kernel has enabled dynamic clocks and unless X also does so, I see this... (II) RADEON(0): Dynamic Clock Scaling Disabled (WW) RADEON(0): Direct rendering not yet supported on Radeon 9500 and newer cards (EE) RADEON(0): Idle timed out, resetting engine... (EE) RADEON(0): Idle timed out, resetting engine... ... and then it dies. Ben, is this expected behaviour if the kernel and X disagree about the use of dynamic clocks? Is this addressed by one of the patches you have outstanding at http://gate.crashing.org/~benh/xorg/ ? The second problem with the PB5,3 is that rhpl explicitly specifies 800x600 instead of letting X honour the DDC information. I can see that rhpl has code to parse this, but it doesn't seem to be doing so... (II) RADEON(0): Supported additional Video Mode: (II) RADEON(0): clock: 96.2 MHz Image Size: 367 x 230 mm (II) RADEON(0): h_active: 1440 h_sync: 1504 h_sync_end 1536 h_blank_end 1760 h_border: 0 (II) RADEON(0): v_active: 900 v_sync: 903 v_sync_end 906 v_blanking: 912 v_border: 0 <...> (II) RADEON(0): Panel size found from DDC: 1440x900 (II) RADEON(0): Valid Mode from Detailed timing table: 1440x900 This patch to the rhpl-generated XConfig.test makes it work: --- XConfig.test.orig 2005-05-03 12:51:18.000000000 +0100 +++ XConfig.test 2005-05-03 12:51:05.000000000 +0100 @@ -104,2 +104,3 @@ Option "UseFBDev" "true" + Option "DynamicClocks" "true" EndSection @@ -114,3 +115,2 @@ Depth 16 - Modes "800x600" "640x480" EndSubSection Full log at http://david.woodhou.se/Xorg.1.log -- dwmw2 From gauravp at hclcomnet.co.in Tue May 3 13:38:31 2005 From: gauravp at hclcomnet.co.in (gaurav) Date: Tue, 03 May 2005 19:08:31 +0530 Subject: Stateless Debian Project In-Reply-To: References: <42771C1F.4040206@hclcomnet.co.in> Message-ID: <42777ED7.3060502@hclcomnet.co.in> William Beebe wrote: >How about putting this boundless energy into more timely releases of >Debian? > pl note that Debian has versions Stable (1-2 yrs....like RHEL 18 months) , testing (yearly), unstable(few months ...like fedora).... >And did you ever stop to think that if stateless was dropped >that there was a good reason to drop it? > As per RH not enough volunteers ....lots of people other people (me included)... there is lot of potential in the concept..I wrote many mails regarding the status of Stateless Linux and offering to volunteer but ..... > Why does Debian think it has >to become the technological rat hole for abandoned fringe projects? >Oh, I forgot. Debian is a fringe project. > > > No, my friend your mistaken ....Debian provides good platform ...thats why most Distro have their base on Debian than any other Distro (pl visit distro watch) Pl note this not porting project , we aim extend the existing RH Stateless Linux Project and code will run on any Distro (like fedora or Redhat)...Debian just provides us stable platform (and many distro are based on it) Since this project was started by RH I like will welcome anyone who is interested in joining this project :-) >On 5/3/05, gaurav wrote: > > >>Hi, >> >>We have just initiated Stateless Debian Project and we are looking for >>active volunteers/developers >>Detail of project are given below >> >>Summary >>This project was started by Fedora (Redhat) but is no longer in active >>development and will not be included in next release of fc4 , this >>projects aims to provide longterm commitment to the concept & port it to >>100% community based distro like Debian >> >>Stateless Linux converts normal Linux desktop/clients to Stateless >>machines or appliances, which means if throw your computer out of window >>you still will be able to get exactly same same settings/data when you log >>from any other pc in the network ....A single administrator can easily >>manage network thousands of desktops ...Stateless Linux centralizes the >>state in a Gold server (different from CFengine) and rest of clients >>are updated regularly from it . This is different from thin clients as >>local processing power and memory of clients is used (or cached client) >> >>http://sourceforge.net/projects/statelessdebian/ >>Project Goals >> >>1) Port relevant parts of RH State Linux (http://fedora.redhat.com/projects/stateless/) to Debian >>2)Extend/integrate/enhance Stateless Linux thru existing projects like DBRL >>http://drbl.sourceforge.net/debian/wiki-view/pmwiki-view.php/DRBL/WhatIsDRBL >>and Debian CCD infrastructure, FAI (Fully Automated Installation,http://www.informatik.uni-koeln.de/fai/) >>3) Add cluster support and node monitoring and autonomic capabilities >>4)Meets most of OSDL Desktop Linux specification >>5) Long term community based , vendor neutral project >>6) Member of "Move 2 Debian" project >>7) Latter (2nd phase )Add p2p support for directory support(DHT), file storage and retrieval(tuples), state transfer (Bittorent) >>to elinimate any need of centralized servers/infrastructure and make it highly scalable (to millions!!) >> >>Right now this project is in pre beta stage and we are preparing Stateless Linux White paper and started porting portions of RH Stateless Linux to Debian . >>Any Suggestions/Ideas will appreciated :-) ..if anyone is interested from this list then pl let me know >> >>Regards, >>Gaurav >> >>-- >>fedora-devel-list mailing list >>fedora-devel-list at redhat.com >>http://www.redhat.com/mailman/listinfo/fedora-devel-list >> >> >> > > > From mihamina.rakotomandimby at etu.univ-orleans.fr Tue May 3 14:28:34 2005 From: mihamina.rakotomandimby at etu.univ-orleans.fr (Rakotomandimby (R12y) Mihamina) Date: Tue, 03 May 2005 16:28:34 +0200 Subject: What dependencies does it need ? Message-ID: <1115130514.6589.0.camel@fctmp> Hi, I made a findlib package from the altlinux src.rpm, in order to make a ocamlnet package. ocamlnet depends on findlib. You'll find at the bottom the specfile i used. It builds without any errors. But when I try to install the thing: # rpm -i /usr/src/redhat/RPMS/i386/findlib-1.0.4-fc3.i386.rpm error: Failed dependencies: libc.so.6(GLIBC_PRIVATE) is needed by findlib-1.0.4-fc3.i386 perl(gtkdoc-common.pl) is needed by findlib-1.0.4-fc3.i386 What did I do wrong? Here is the specfile: ======================== specfile ==================== %define ocamlver 3.08.3 Name: findlib Version: 1.0.4 Release: fc3 Group: Development/ML Summary: A module packaging tool for OCaml License: GPL Packager: Mihamina Rakotomandimby Url: http://www.ocaml-programming.de/packages/documentation/findlib/ Source: http://www.ocaml-programming.de/packages/%name-%version.tar.gz Requires: ocaml = %ocamlver # Automatically added by buildreq on Mon Oct 13 2003 BuildRequires: camlp4 labltk ncurses-devel # ocaml-runtime BuildRequires: ocaml = %ocamlver %package -n ocamlfind-mini Summary: Minimal findlib script to be distributed with user libraries Group: Development/ML Requires: %name = %version-%release %description The "findlib" library provides a scheme to manage reusable software components (packages), and includes tools that support this scheme. Packages are collections of OCaml modules for which metainformation can be stored. The packages are kept in the filesystem hierarchy, but with strict directory structure. The library contains functions to look the directory up that stores a package, to query metainformation about a package, and to retrieve dependency information about multiple packages. There is also a tool that allows the user to enter queries on the command-line. In order to simplify compilation and linkage, there are new frontends of the various OCaml compilers that can directly deal with packages. %description -n ocamlfind-mini ocamlfind-mini is an O'Caml script that implements a subset of the full functionality of ocamlfind. It consists only of one file, so it is easy to distribute it with any software. The subset is normally sufficient to compile a library and to install the library; but it is insufficient to link the library into an executable. %prep %setup -q %build ./configure -mandir /usr/share/man - config /usr/lib/ocaml/etc/findlib.conf -with-toolbox make all opt %install %define _compress_method skip make prefix=$RPM_BUILD_ROOT install mkdir -p $RPM_BUILD_ROOT%_docdir/%name-%version cp README INSTALL LICENSE $RPM_BUILD_ROOT%_docdir/%name-%version/ mkdir -p $RPM_BUILD_ROOT/usr/lib/ocaml/ocamlfind-mini/ cp mini/* $RPM_BUILD_ROOT/usr/lib/ocaml/ocamlfind-mini/ %files -n ocamlfind-mini %_libdir/ocaml/ocamlfind-mini/* %files %_bindir/* %_libdir/ocaml/etc/* %_libdir/ocaml/topfind %_libdir/ocaml/site-lib/* %_mandir/man?/* %doc README doc/ INSTALL LICENSE %changelog * Mon Apr 2 2005 Mihamina Rakotomandimby 1.0.4-fc3 - Vitaly Lugovsky package modification to suit it to Fedora Core 3 - rebuild ===================================== end of spec ================== -- Get a fully managed dedicated server for ?200/month ($257/month) No time limit for taking care of your server. You keep the "root" acces if you want. Billing periods are 3 months. See the conditions at http://aspo.rktmb.org/activities/managed_servers From dcbw at redhat.com Tue May 3 14:31:32 2005 From: dcbw at redhat.com (Dan Williams) Date: Tue, 03 May 2005 10:31:32 -0400 Subject: What dependencies does it need ? In-Reply-To: <1115130514.6589.0.camel@fctmp> References: <1115130514.6589.0.camel@fctmp> Message-ID: <1115130692.13829.4.camel@dcbw.boston.redhat.com> On Tue, 2005-05-03 at 16:28 +0200, Rakotomandimby (R12y) Mihamina wrote: > perl(gtkdoc-common.pl) is needed by findlib-1.0.4-fc3.i386 Add a "Requires: gtk-doc" for this one at least, that's where gtkdoc-common.pl lives. Dan From mihamina.rakotomandimby at etu.univ-orleans.fr Tue May 3 14:40:17 2005 From: mihamina.rakotomandimby at etu.univ-orleans.fr (Rakotomandimby (R12y) Mihamina) Date: Tue, 03 May 2005 16:40:17 +0200 Subject: What dependencies does it need ? In-Reply-To: <1115130692.13829.4.camel@dcbw.boston.redhat.com> References: <1115130514.6589.0.camel@fctmp> <1115130692.13829.4.camel@dcbw.boston.redhat.com> Message-ID: <1115131217.1859.18.camel@ngeza> On Tue, 2005-05-03 at 10:31 -0400, Dan Williams wrote: > Add a "Requires: gtk-doc" for this one at least, that's where > gtkdoc-common.pl lives. Okay. I guess the libc issue is stilla mytery... :-) PS: I do apoogize for the multiple posting, but I just changed email address. So that I seem to be confused on which was used... -- ASPO Infog?rance http://aspo.rktmb.org/activites/infogerance Unofficial FAQ fcolc http://faq.fcolc.eu.org/ LUG sur Orl?ans et alentours (France). T?l : 02 34 08 26 04 / 06 33 26 13 14 From katzj at redhat.com Tue May 3 14:41:54 2005 From: katzj at redhat.com (Jeremy Katz) Date: Tue, 03 May 2005 10:41:54 -0400 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115122336.4446.4.camel@hades.cambridge.redhat.com> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> Message-ID: <1115131314.5982.28.camel@bree.local.net> On Tue, 2005-05-03 at 13:12 +0100, David Woodhouse wrote: > The second problem with the PB5,3 is that rhpl explicitly specifies > 800x600 instead of letting X honour the DDC information. I can see that > rhpl has code to parse this, but it doesn't seem to be doing so... We only parse it if the X startup fails. If the X startup _succeeds_, then parsing the log, killing X, and restarting it may lead to X not starting back up, which ends up being worse. But this basically sounds like us not being worse off on ppc laptops than on x86 ones. This gets back to the discussion we had a few weeks ago about X configuration :-) Jeremy From mricon at gmail.com Tue May 3 14:59:46 2005 From: mricon at gmail.com (Konstantin Ryabitsev) Date: Tue, 3 May 2005 10:59:46 -0400 Subject: rawhide report: 20050503 changes In-Reply-To: <200505031210.j43CAVwY027741@porkchop.devel.redhat.com> References: <200505031210.j43CAVwY027741@porkchop.devel.redhat.com> Message-ID: On 5/3/05, Build System wrote: > kernel-2.6.11-1.1282_FC4 > ------------------------ > * Sun May 01 2005 Dave Jones > - Fix yesterdays IDE fixes. > - Blacklist another brainless SCSI scanner. (#155457) > > * Sun May 01 2005 David Woodhouse > - Fix EHCI port power switching > > * Sun May 01 2005 Dave Jones > - Enable usbmon & debugfs. (#156489) My USB mouse doesn't work with this kernel (Logitech USB MouseMan). Probably the result of one of the above. Regards, -- Konstantin Ryabitsev Zlotniks, INC From paul at city-fan.org Tue May 3 15:13:01 2005 From: paul at city-fan.org (Paul Howarth) Date: Tue, 03 May 2005 16:13:01 +0100 Subject: What dependencies does it need ? In-Reply-To: <1115131217.1859.18.camel@ngeza> References: <1115130514.6589.0.camel@fctmp> <1115130692.13829.4.camel@dcbw.boston.redhat.com> <1115131217.1859.18.camel@ngeza> Message-ID: <427794FD.1070901@city-fan.org> Rakotomandimby (R12y) Mihamina wrote: > On Tue, 2005-05-03 at 10:31 -0400, Dan Williams wrote: > >>Add a "Requires: gtk-doc" for this one at least, that's where >>gtkdoc-common.pl lives. > > > Okay. > I guess the libc issue is stilla mytery... :-) One of the binaries in the package is probably linked against a "private" function in glibc. You may be able to find the problematic binary using "objdump -p" on each binary in the package and look for GLIBC_PRIVATE in the output. You would then need to figure out which symbol was causing the problem and fix the build process to not detect that symbol in glibc (e.g. by forcing the configure script to use an alternative implementation). I had a similar issue with libspf2 finding the private symbol __ns_get16 in glibc. Good luck! Paul. From jnovy at redhat.com Tue May 3 15:28:08 2005 From: jnovy at redhat.com (Jindrich Novy) Date: Tue, 03 May 2005 17:28:08 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <1115065557.12093.48.camel@localhost.localdomain> References: <200505021935.j42JZqNm022649@magilla.sf.frob.com> <1115065557.12093.48.camel@localhost.localdomain> Message-ID: <1115134088.29438.133.camel@obelix.redhat.usu> On Mon, 2005-05-02 at 16:25 -0400, Peter Jones wrote: > On Mon, 2005-05-02 at 12:35 -0700, Roland McGrath wrote: > > > Roland McGrath wrote: > > > > I think what one clearly wants is for rpm to maintain an installed file > > > > indexed keyed by md5sum. Then you can have a tool that just uses this > > > > database to identify duplicates (and doesn't take forever), or have rpm do > > > > so itself when installing new files. > > > > > > > > > > Hmm, what about hash collisions, that would be really really BAD > > > > If you are concerned about them you can still compare contents before > > declaring two files identical. But using the hashes as the main detector > > makes it fast, since you only examine the data of files that are 99.999% > > likely to be identical. > > And in the vast majority of cases, there's a simpler heuristic you can > use first: is the basename the same? The easiest way seems to be only to stat all the files to be compared, put all info to some array of pointers to the info structures, sort the array by size [this will automagically detect all zero-sized files that won't be linked and are skipped] then just go from top to bottom in the array and check in-depth all the files with equal size, i.e. byte-by- byte compare during the md5sum is calculated. This avoids all the md5sum collisions. This is how it's done in the slink utility, the md5sums are printed in the log just FYI and isn't used as a measure of file equality. The basename heuristics seems less reliable and more calculation-time/design expensive to me. Jindrich -- Jindrich Novy , http://people.redhat.com/jnovy/ The worst evil in the world is refusal to think. From dwmw2 at infradead.org Tue May 3 15:38:57 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 03 May 2005 16:38:57 +0100 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115131314.5982.28.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> <1115131314.5982.28.camel@bree.local.net> Message-ID: <1115134737.16187.9.camel@hades.cambridge.redhat.com> On Tue, 2005-05-03 at 10:41 -0400, Jeremy Katz wrote: > We only parse it if the X startup fails. If the X startup _succeeds_, > then parsing the log, killing X, and restarting it may lead to X not > starting back up, which ends up being worse. But this basically sounds > like us not being worse off on ppc laptops than on x86 ones. Couldn't we just refrain from specifying a mode in this case, since we _know_ we're just pulling it out of our wossname? Or if we _must_ specify a mode, what's wrong with the one we get from 'fbset -x'? -- dwmw2 From selinux at gmail.com Tue May 3 15:50:23 2005 From: selinux at gmail.com (Tom London) Date: Tue, 3 May 2005 08:50:23 -0700 Subject: rawhide report: 20050503 changes In-Reply-To: References: <200505031210.j43CAVwY027741@porkchop.devel.redhat.com> Message-ID: <4c4ba1530505030850654cb1fc@mail.gmail.com> On 5/3/05, Konstantin Ryabitsev wrote: > My USB mouse doesn't work with this kernel (Logitech USB MouseMan). > Probably the result of one of the above. > > Regards, > -- > Konstantin Ryabitsev > Zlotniks, INC > Yeah. My mouse 'points and clicks', but the wheel and middle button are dead. tom -- Tom London From link at pobox.com Tue May 3 16:00:04 2005 From: link at pobox.com (Terje Bless) Date: Tue, 3 May 2005 18:00:04 +0200 Subject: Duplicated files in the pristine FC4t2 installation In-Reply-To: <4276825D.7090204@redhat.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike A. Harris wrote: >I think it is 100% wrong to mark files as duplicates because they are >the same "now". There is no guarantee they will be the same in a future >update. Excluding a GPL license COPYING file from one package and >linking it to another central copy fails the second someone decides to >use GPLv3 for that package. So make virtual provides for each license in the licenses.rpm and depend on that. Not caring much for the mentioned 160MB on install (allthough the contribution to number of install media might be relevant), but looking at it from askew; the license is _metadata_ and belongs in the metadata section of the rpm -- i.e. License plus a Requires -- with a pointer to the real license file. cf. Creative Commons; you don't include the full license in every little web page under a CC license, you point to the license in the metadata. Among other things, this enables automated discovery and processing of license information in addition to the marginal benefit of reducing disk space wasted. Playing tricks with hardlink or softlinking seems pointless and silly to me. >Or if they add text to the top of the file or something. At which point the packager will have to take a look at it to determine what to do -- which s/he would have to anyway when the license changed -- possibly including removing the Requires and putting the COPYING file back in the package. - -- Interviewer: "In what language do you write your algorithms?" Abigail: English. Interviewer: "What would you do if, say, Telnet didn't work?" Abigail: Look at the error message. -----BEGIN PGP SIGNATURE----- Version: PGP SDK 3.2.2 iQA/AwUBQnegA6PyPrIkdfXsEQLx2QCffhq6InBzuBE2r5dBqUjQ8RNBEzsAoO9/ dKx9b5kggTDiFH5gWZB5OUUV =la8q -----END PGP SIGNATURE----- From katzj at redhat.com Tue May 3 16:04:45 2005 From: katzj at redhat.com (Jeremy Katz) Date: Tue, 03 May 2005 12:04:45 -0400 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115134737.16187.9.camel@hades.cambridge.redhat.com> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> <1115131314.5982.28.camel@bree.local.net> <1115134737.16187.9.camel@hades.cambridge.redhat.com> Message-ID: <1115136285.5982.49.camel@bree.local.net> On Tue, 2005-05-03 at 16:38 +0100, David Woodhouse wrote: > On Tue, 2005-05-03 at 10:41 -0400, Jeremy Katz wrote: > > We only parse it if the X startup fails. If the X startup _succeeds_, > > then parsing the log, killing X, and restarting it may lead to X not > > starting back up, which ends up being worse. But this basically sounds > > like us not being worse off on ppc laptops than on x86 ones. > > Couldn't we just refrain from specifying a mode in this case, since we > _know_ we're just pulling it out of our wossname? Or if we _must_ > specify a mode, what's wrong with the one we get from 'fbset -x'? The problem is that adding a "if this looks like this, and this looks like this and this looks like something else" to do a workaround like not specifying the modes does *not* lead to the code being maintainable. It's already bad enough, there's no need to make things worse. :-) But, doing a hack to look at the fb mode on ppc and try that resolution probably isn't terrible to do in anaconda [1]. I'm afraid it still isn't going to help with some of the radeon problems, though :/ Jeremy [1] In fact, it's not -- committed the following Index: anaconda =================================================================== RCS file: /usr/local/CVS/anaconda/anaconda,v retrieving revision 1.340 diff -u -u -r1.340 anaconda --- anaconda 3 May 2005 02:00:45 -0000 1.340 +++ anaconda 3 May 2005 15:54:51 -0000 @@ -911,6 +911,14 @@ # if no monitor probed lets guess based on runres hsync = monitorhw.getMonitorHorizSync() vsync = monitorhw.getMonitorVertSync() + + # if we're on a pmac, try to use the fb resolution + if not runres_override and iutil.getPPCMachine() == "PMac": + fbinfo = isys.fbinfo() + fbres = "%sx%s" %(fbinfo[0], fbinfo[1]) + if monitor.monitor_supports_mode(hsync, vsync, fbres): + runres = fbres + res_supported = monitor.monitor_supports_mode(hsync, vsync, runres) # XXX: this is a bit of a hack, but there are a reasonable number of From sundaram at redhat.com Tue May 3 16:08:10 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Tue, 03 May 2005 21:38:10 +0530 Subject: Stateless Debian Project In-Reply-To: <42777ED7.3060502@hclcomnet.co.in> References: <42771C1F.4040206@hclcomnet.co.in> <42777ED7.3060502@hclcomnet.co.in> Message-ID: <4277A1EA.5020502@redhat.com> Hi > pl note that Debian has versions Stable (1-2 yrs....like RHEL 18 > months) , testing (yearly), unstable(few months ...like fedora).... If you really insist on a comparison, Debian unstable is more like Fedora development tree. Fedora project itself does test releases periodically between every major release which has been about every 6 months so far. Updates are provided till the second test release of second subsequent major version following it. You can use Fedora legacy to extend this lifecyle even further > As per RH not enough volunteers ....lots of people other people (me > included)... there is lot of potential in the concept..I wrote many > mails regarding the status of Stateless Linux and offering to > volunteer but ..... You are still welcome to work on it and produce actual code. Fedora extras CVS is open to community contributors regards Rahul From fedora at camperquake.de Tue May 3 16:08:30 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Tue, 3 May 2005 18:08:30 +0200 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115072500.5982.9.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> Message-ID: <20050503180830.1570c70e@nausicaa.camperquake.de> Hi. Jeremy Katz wrote: > Hi. I've got an ISO for quick testing of anaconda's X configuration > setup available that's a little bit smaller than a full tree. All you > should have to do is download the ISO, burn it to a CD, boot and see if > X starts up. Then, you can reboot and go back to your normal doings. Hardware: processor : 0 cpu : 750CXe temperature : 38 C (uncalibrated) clock : 499MHz revision : 34.21 (pvr 0008 2215) bogomips : 995.32 machine : PowerBook4,1 motherboard : PowerBook4,1 MacRISC2 MacRISC Power Macintosh detected as : 257 (iBook 2) pmac flags : 0000001b L2 cache : 256K unified memory : 384MB pmac-generation : NewWorld iBook2, 500MHz, ATI Rage 128 Mobility The installer comes up, in 800x600. The display does not like that too much, the native resolution is 1024x768. The installer is displayed in the upper left corner, a ~40 pixel high bar of garbage (in varying colors) below, and the installer screen repeated below that. The right part of the screen is black. So, it's basically all right if the installer chooses 1024x768. -- Sturgeon's Law: 90% of science fiction is junk. But then, 90% of everything is junk. From dwmw2 at infradead.org Tue May 3 16:12:43 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 03 May 2005 17:12:43 +0100 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115122336.4446.4.camel@hades.cambridge.redhat.com> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> Message-ID: <1115136764.16187.15.camel@hades.cambridge.redhat.com> On Tue, 2005-05-03 at 13:12 +0100, David Woodhouse wrote: > PowerBook5,3 (Radeon 9600): Kills machine hard. The kernel has enabled > dynamic clocks and unless X also does so, I see this... > > (II) RADEON(0): Dynamic Clock Scaling Disabled > (WW) RADEON(0): Direct rendering not yet supported on Radeon 9500 and newer cards > (EE) RADEON(0): Idle timed out, resetting engine... > (EE) RADEON(0): Idle timed out, resetting engine... > ... and then it dies. This is actually a caused by a patch we have in our xorg-x11 RPM. Stock 6.8.2 works fine (albeit at 800x600) with dynamic clocks off. cf. https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=152648 -- dwmw2 From dwmw2 at infradead.org Tue May 3 16:16:32 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 03 May 2005 17:16:32 +0100 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115136285.5982.49.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> <1115131314.5982.28.camel@bree.local.net> <1115134737.16187.9.camel@hades.cambridge.redhat.com> <1115136285.5982.49.camel@bree.local.net> Message-ID: <1115136992.16187.19.camel@hades.cambridge.redhat.com> On Tue, 2005-05-03 at 12:04 -0400, Jeremy Katz wrote: > The problem is that adding a "if this looks like this, and this looks > like this and this looks like something else" to do a workaround like > not specifying the modes does *not* lead to the code being maintainable. > It's already bad enough, there's no need to make things worse. :-) The logic I was thinking of would be "If we have information about supported modes, then specify a mode. Else keep quiet and let X work it out for itself". > But, doing a hack to look at the fb mode on ppc and try that resolution > probably isn't terrible to do in anaconda [1]. Does it need to be ppc-specific? If DDC fails, just try to get a sane modeline from the framebuffer. > I'm afraid it still isn't going to help with some of the radeon > problems, though :/ See separate mail re the dynamic clock lockup. -- dwmw2 From aoliva at redhat.com Tue May 3 16:23:43 2005 From: aoliva at redhat.com (Alexandre Oliva) Date: 03 May 2005 13:23:43 -0300 Subject: What dependencies does it need ? In-Reply-To: <1115131217.1859.18.camel@ngeza> References: <1115130514.6589.0.camel@fctmp> <1115130692.13829.4.camel@dcbw.boston.redhat.com> <1115131217.1859.18.camel@ngeza> Message-ID: On May 3, 2005, "Rakotomandimby (R12y) Mihamina" wrote: > On Tue, 2005-05-03 at 10:31 -0400, Dan Williams wrote: >> Add a "Requires: gtk-doc" for this one at least, that's where >> gtkdoc-common.pl lives. > Okay. > I guess the libc issue is stilla mytery... :-) Presumably the program references glibc-internal symbols, which it shouldn't do. -- Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org} Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org} From davej at redhat.com Tue May 3 16:32:31 2005 From: davej at redhat.com (Dave Jones) Date: Tue, 3 May 2005 12:32:31 -0400 Subject: rawhide report: 20050503 changes In-Reply-To: <4c4ba1530505030850654cb1fc@mail.gmail.com> References: <200505031210.j43CAVwY027741@porkchop.devel.redhat.com> <4c4ba1530505030850654cb1fc@mail.gmail.com> Message-ID: <20050503163230.GB26161@redhat.com> On Tue, May 03, 2005 at 08:50:23AM -0700, Tom London wrote: > On 5/3/05, Konstantin Ryabitsev wrote: > > My USB mouse doesn't work with this kernel (Logitech USB MouseMan). > > Probably the result of one of the above. > > > > Regards, > > -- > > Konstantin Ryabitsev > > Zlotniks, INC > > > Yeah. > > My mouse 'points and clicks', but the wheel and middle button are dead. Yeah, known bug. I'll get this fixed up even if it means dropping this change for now. Dave From katzj at redhat.com Tue May 3 16:47:39 2005 From: katzj at redhat.com (Jeremy Katz) Date: Tue, 03 May 2005 12:47:39 -0400 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115136992.16187.19.camel@hades.cambridge.redhat.com> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> <1115131314.5982.28.camel@bree.local.net> <1115134737.16187.9.camel@hades.cambridge.redhat.com> <1115136285.5982.49.camel@bree.local.net> <1115136992.16187.19.camel@hades.cambridge.redhat.com> Message-ID: <1115138861.5982.51.camel@bree.local.net> On Tue, 2005-05-03 at 17:16 +0100, David Woodhouse wrote: > > But, doing a hack to look at the fb mode on ppc and try that resolution > > probably isn't terrible to do in anaconda [1]. > > Does it need to be ppc-specific? If DDC fails, just try to get a sane > modeline from the framebuffer. We use vga16fb on i386/x86_64/ia64 to get CJK text consoles. And trying to get fbinfo from that isn't going to give results that we want or expect :) Jeremy From katzj at redhat.com Tue May 3 17:06:25 2005 From: katzj at redhat.com (Jeremy Katz) Date: Tue, 03 May 2005 13:06:25 -0400 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <20050503180830.1570c70e@nausicaa.camperquake.de> References: <1115072500.5982.9.camel@bree.local.net> <20050503180830.1570c70e@nausicaa.camperquake.de> Message-ID: <1115139986.5982.54.camel@bree.local.net> On Tue, 2005-05-03 at 18:08 +0200, Ralf Ertzinger wrote: > Jeremy Katz wrote: > > Hi. I've got an ISO for quick testing of anaconda's X configuration > > setup available that's a little bit smaller than a full tree. All you > > should have to do is download the ISO, burn it to a CD, boot and see if > > X starts up. Then, you can reboot and go back to your normal doings. [snip] > iBook2, 500MHz, ATI Rage 128 Mobility > > The installer comes up, in 800x600. The display does not like that too much, > the native resolution is 1024x768. The installer is displayed in the upper > left corner, a ~40 pixel high bar of garbage (in varying colors) below, > and the installer screen repeated below that. The right part of the screen > is black. > > So, it's basically all right if the installer chooses 1024x768. So booting with resolution=1024x768 helps? Also, does the fb come up in 1024x768 normally? (you can check by running fbset on the console and seeing what it says the resolution is). If so, then the patch I just committed should help. Jeremy From dwmw2 at infradead.org Tue May 3 17:08:24 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 03 May 2005 18:08:24 +0100 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115138861.5982.51.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> <1115131314.5982.28.camel@bree.local.net> <1115134737.16187.9.camel@hades.cambridge.redhat.com> <1115136285.5982.49.camel@bree.local.net> <1115136992.16187.19.camel@hades.cambridge.redhat.com> <1115138861.5982.51.camel@bree.local.net> Message-ID: <1115140104.16187.31.camel@hades.cambridge.redhat.com> On Tue, 2005-05-03 at 12:47 -0400, Jeremy Katz wrote: > We use vga16fb on i386/x86_64/ia64 to get CJK text consoles. And trying > to get fbinfo from that isn't going to give results that we want or > expect :) Hm. Use what we get from the fb if it's better than 640x480, perhaps? Or if the driver _isn't_ vga16fb? But we're rapidly descending into the realm of the not-pretty, I agree. -- dwmw2 From fedora at camperquake.de Tue May 3 17:15:41 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Tue, 3 May 2005 19:15:41 +0200 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115139986.5982.54.camel@bree.local.net> References: <1115072500.5982.9.camel@bree.local.net> <20050503180830.1570c70e@nausicaa.camperquake.de> <1115139986.5982.54.camel@bree.local.net> Message-ID: <20050503191541.6d6be140@nausicaa.camperquake.de> Hi. Jeremy Katz wrote: > So booting with resolution=1024x768 helps? That produces a 800x600 installer with grey borders, extending it to 1024x768. > Also, does the fb come up in > 1024x768 normally? fbset claims 1024x768. -- And all the roads that lead you there are winding And all the lights that light the way are blinding -- Oasis, "Wonderwall" From katzj at redhat.com Tue May 3 17:19:43 2005 From: katzj at redhat.com (Jeremy Katz) Date: Tue, 03 May 2005 13:19:43 -0400 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <20050503191541.6d6be140@nausicaa.camperquake.de> References: <1115072500.5982.9.camel@bree.local.net> <20050503180830.1570c70e@nausicaa.camperquake.de> <1115139986.5982.54.camel@bree.local.net> <20050503191541.6d6be140@nausicaa.camperquake.de> Message-ID: <1115140783.5982.56.camel@bree.local.net> On Tue, 2005-05-03 at 19:15 +0200, Ralf Ertzinger wrote: > Jeremy Katz wrote: > > > So booting with resolution=1024x768 helps? > > That produces a 800x600 installer with grey borders, extending it to 1024x768. > > > Also, does the fb come up in > > 1024x768 normally? > > fbset claims 1024x768. Perfect. Then the change I made before lunch should fix that case. Maybe we're just down to dwmw2's dynamic clocks problem and the g5 not working with that. Which would be a significant improvement over where we were 24 hours ago :) Jeremy From mihamina.rakotomandimby at etu.univ-orleans.fr Tue May 3 17:26:02 2005 From: mihamina.rakotomandimby at etu.univ-orleans.fr (Rakotomandimby (R12y) Mihamina) Date: Tue, 03 May 2005 19:26:02 +0200 Subject: What dependencies does it need ? In-Reply-To: References: <1115130514.6589.0.camel@fctmp> <1115130692.13829.4.camel@dcbw.boston.redhat.com> <1115131217.1859.18.camel@ngeza> Message-ID: <1115141162.1860.22.camel@ngeza> On Tue, 2005-05-03 at 13:23 -0300, Alexandre Oliva wrote: > Presumably the program references glibc-internal symbols, which it > shouldn't do. I'm not very good in programming yet. Do you think Paul Howarth's indications are good for me to follow? May be there are some more newbie way to manage it. I dont know. I will also ask on usenet, then we'll see. :-) Thanks for your -- ASPO Infog?rance http://aspo.rktmb.org/activites/infogerance Unofficial FAQ fcolc http://faq.fcolc.eu.org/ LUG sur Orl?ans et alentours (France). T?l : 02 34 08 26 04 / 06 33 26 13 14 From rmo at sunnmore.net Tue May 3 18:25:21 2005 From: rmo at sunnmore.net (Roy-Magne Mo) Date: Tue, 3 May 2005 20:25:21 +0200 Subject: rawhide report: 20050503 changes In-Reply-To: <200505031210.j43CAVwY027741@porkchop.devel.redhat.com> References: <200505031210.j43CAVwY027741@porkchop.devel.redhat.com> Message-ID: <20050503182521.GA8694@slogen.sunnmore.net> On Tue, May 03, 2005 at 08:10:31AM -0400, Build System wrote: > kernel-2.6.11-1.1282_FC4 > ------------------------ > * Sun May 01 2005 Dave Jones > - Fix yesterdays IDE fixes. > - Blacklist another brainless SCSI scanner. (#155457) > > * Sun May 01 2005 David Woodhouse > - Fix EHCI port power switching > > * Sun May 01 2005 Dave Jones > - Enable usbmon & debugfs. (#156489) Seems like this kernel update has fixed my "long" standing bug, BZ #152298. I'll update the bugzilla. -- Roy-Magne Mo From benh at kernel.crashing.org Tue May 3 23:32:20 2005 From: benh at kernel.crashing.org (Benjamin Herrenschmidt) Date: Wed, 04 May 2005 09:32:20 +1000 Subject: Request for Testing of Installer X Configuration on PPC In-Reply-To: <1115136764.16187.15.camel@hades.cambridge.redhat.com> References: <1115072500.5982.9.camel@bree.local.net> <1115122336.4446.4.camel@hades.cambridge.redhat.com> <1115136764.16187.15.camel@hades.cambridge.redhat.com> Message-ID: <1115163140.7627.30.camel@gaston> On Tue, 2005-05-03 at 17:12 +0100, David Woodhouse wrote: > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=152648 Well, this is weird. But first of all, why are you disabling Dynamic Clocks ? Isn't this a laptop ? You should disable them most of the time. Also, the driver is supposed not to "touch" dynamic clocks at all on non-mobility chips. There are several issues adressed by my patches, I can't say for sure about this one, but you should certainly merge them if you want things like working VGA output etc... I'll try current CVS head with dynclk off on Paulus 5,3 today. There used to be a bug in the dynclk code "off" that would mixup "OUPLL" and "OUTREG" that I fixed at some point, I'm not sure if that fix is in your tree. There may be some other problem as well. Ben. From mandreiana.lists at gmail.com Wed May 4 05:59:52 2005 From: mandreiana.lists at gmail.com (Marius Andreiana) Date: Wed, 04 May 2005 08:59:52 +0300 Subject: applications which depend on esound Message-ID: <1115186392.4047.10.camel@marte.biciclete.ro> With alsa 1.0.9rc3, dmix is enabled by default and it works (with gstreamer too). There are several applications which depend on esound which seems weird, as I thought these use gstreamer: libesd.so.0 is needed by (installed) nautilus-2.10.0-3.i386 libesd.so.0 is needed by (installed) gnome-media-2.10.0-2.i386 libesd.so.0 is needed by (installed) gok-1.0.3-1.i386 libesd.so.0 is needed by (installed) control-center-2.10.0-4.i386 libesd.so.0 is needed by (installed) smpeg-0.4.4-9.i386 libesd.so.0 is needed by (installed) libgnome-2.10.0-2.i386 libesd.so.0 is needed by (installed) gnome-session-2.10.0-2.i386 esound >= 1:0.2.27 is needed by (installed) libgnomeui-2.10.0-1.i386 Should I file upstream bugs for these to remove direct esound depencies and require only gstreamer? These depencies seem ok to me until esound is dropped, as these are sound systems: libesd.so.0 is needed by (installed) arts-1.4.0-1.i386 libesd.so.0 is needed by (installed) gstreamer-plugins-0.8.8-5.i386 Note1: alsa-lib 1.0.9rc3 will be in updates soon, now only at http://people.redhat.com/stransky/alsa-lib/ Note2: dmix doesn't solve "sound over network" problem. This still requires a sound server (e.g. MAS, Jack, esd), but the sound server can be implemented at gstreamer framework level, not application level. -- Marius Andreiana From aani_avni at yahoo.com Wed May 4 07:15:02 2005 From: aani_avni at yahoo.com (avni thacker) Date: Wed, 4 May 2005 00:15:02 -0700 (PDT) Subject: i want to volunteer the stateless Linux Message-ID: <20050504071503.89410.qmail@web54002.mail.yahoo.com> Hi i m doing my BE in IT(Information Technology), i m an open source enthu, please guide me so that i can make some real time application in open source, i want to do something exciting and programming is my passion. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From gauravp at hclcomnet.co.in Wed May 4 08:32:14 2005 From: gauravp at hclcomnet.co.in (gaurav) Date: Wed, 04 May 2005 14:02:14 +0530 Subject: i want to volunteer the stateless Linux In-Reply-To: <20050504071503.89410.qmail@web54002.mail.yahoo.com> References: <20050504071503.89410.qmail@web54002.mail.yahoo.com> Message-ID: <4278888E.4020703@hclcomnet.co.in> avni thacker wrote: > Hi > i m doing my BE in IT(Information Technology), i m an open source > enthu, please guide me so that i can make some real time application > in open source, i want to do something exciting and programming is my > passion. > stateless linux != Real Time Linux ....if you want to make realtime linux pl check http://source.mvista.com/linux_2_6_RT.html .....if you are intrested in stateless linux (and beyond) ...you may join our project http://sourceforge.net/projects/statelessdebian/ > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > From rbultje at ronald.bitfreak.net Wed May 4 10:10:28 2005 From: rbultje at ronald.bitfreak.net (Ronald S. Bultje) Date: Wed, 04 May 2005 12:10:28 +0200 Subject: applications which depend on esound In-Reply-To: <1115186392.4047.10.camel@marte.biciclete.ro> References: <1115186392.4047.10.camel@marte.biciclete.ro> Message-ID: <1115201427.3021.682.camel@tux.lan> Hi, On Wed, 2005-05-04 at 07:59, Marius Andreiana wrote: > There are several applications which depend on esound which seems weird, > as I thought these use gstreamer: > libesd.so.0 is needed by (installed) nautilus-2.10.0-3.i386 > libesd.so.0 is needed by (installed) gnome-media-2.10.0-2.i386 [..] > esound >= 1:0.2.27 is needed by (installed) > libgnomeui-2.10.0-1.i386 [..] > Should I file upstream bugs for these to remove direct esound depencies > and require only gstreamer? No. The only actual dependency is in libgnome/&ui, and the others solely depend on esd because libgnome/&ui do. The libgnome/&ui people are aware of the problem and may at some point be ported to GStreamer; the problem isn't so much the patching itself, because patches are available, but the amount on which we want to drop an unstable API of a huge library into the GNOME developer platform (*). Once ported, that will remove the esd dependency from the others, too. Ronald (*) that doesn't mean it won't be accepted; it just means it'll take some time. ;). -- Ronald S. Bultje From buildsys at redhat.com Wed May 4 12:15:12 2005 From: buildsys at redhat.com (Build System) Date: Wed, 4 May 2005 08:15:12 -0400 Subject: rawhide report: 20050504 changes Message-ID: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.4-20050503.144108.FC4.1 ----------------------------------------- * Tue May 03 2005 Chris Feist - Build new sources and remove patch. HelixPlayer-1:1.0.4-4 --------------------- * Tue May 03 2005 John (J5) Palmieri 1:1.0.4-4 - remove compat-gcc32 requirement alsa-lib-1.0.9rc2-3 ------------------- * Tue May 03 2005 Martin Stransky 1.0.9rc2-3 - fixed ainit (#156278, #156505) * Thu Apr 28 2005 David Woodhouse 1.0.9rc2-2 - Fix bogus use of fgetc() in ainit. (#156278) anaconda-10.2.0.59-1 -------------------- * Tue May 03 2005 Jeremy Katz - 10.2.0.59-1 - Try to use the fb res on pmac - Always reset terminal attrs on ppc (notting, #156411) - Remove bogus preexisting LVM info when doing kickstart installs (clumens, #156283) ant-0:1.6.2-3jpp_5fc -------------------- * Fri Apr 29 2005 Gary Benson 0:1.6.2-3jpp_5fc - BC-compile the two core jars. - Own /usr/share/java/ant. * Fri Apr 08 2005 Andrew Overholt 0:1.6.2-3jpp_4fc - Back out ant-apache-javac-ecj.patch and ant.orig changes as they stop eclipse from building. * Tue Mar 29 2005 Gary Benson 0:1.6.2-3jpp_3fc - Add NOTICE file as per Apache License version 2.0. - Remove some now unnecessary gcj workarounds. - Add the jsch subpackage since we now ship jsch. audit-0.7.3-2 ------------- * Tue May 03 2005 Steve Grubb 0.7.3-1 - Add code to get watch list to auditctl - Get -f & -hn working in ausearch - Added search by terminal, exe, and syscall to ausearch program - Added -w parameter to match whole word in ausearch * Wed Apr 27 2005 Steve Grubb 0.7.2-1 - Allow ausearch uid & gid to be non-numeric (root, wheel, etc) - Fix problems with changing run level - Added new code for logging shutdown reason credentials - Update DAEMON messages to use better timestamp bluez-utils-2.15-7 ------------------ * Mon May 02 2005 David Woodhouse 2.15-7 - Run hidd by default (#155831) * Mon May 02 2005 Miloslav Trmac - 2.15-6 - Add missing backslashes to init script descriptions (#154494, patch by Ville Skytt??) dbus-0.33-2 ----------- * Sun May 01 2005 John (J5) Palmieri - 0.33-2 - Backport patch from CVS that fixes int32's being marshaled as uint16's in the python bindings dhcp-10:3.0.2-12 ---------------- * Tue May 03 2005 Jason Vas Dias 10:3.0.2-12 - Rebuild for new glibc - Fix dhcdbd set for multiple interfaces dosfstools-2.10-3 ----------------- * Thu Apr 28 2005 Peter Vrabec 2.10-3 - if HDIO_GETGEO fails, print a warning and default to H=255,S=63 (#155950) e2fsprogs-1.37-3 ---------------- * Wed May 04 2005 Jeremy Katz - 1.37-3 - fix cramfs detection bug in libblkid eclipse-1:3.1.0_fc-0.M6.14 -------------------------- * Mon May 02 2005 Ben Konrath 3.1.0_fc-0.M6.14 - Replace temporary patch to debug.ui with upstream patch to swt (rh#155853). * Sat Apr 30 2005 Ben Konrath 3.1.0_fc-0.M6.13 - Add patch to temporarily stop an NPE in debug.ui (rh#155853). * Thu Apr 28 2005 Phil Muldoon - Allow multiple optional arguments in eclipse-copy-platform.sh eclipse-bugzilla-1:0.1.0_fc-14 ------------------------------ * Mon May 02 2005 Jeff Pound 0.1.0_fc-14 - Fix default bugzilla.redhat.com url. evolution-2.2.2-3 ----------------- * Sat Apr 30 2005 David Malcolm - 2.2.2-3 - updated mozilla_build_version to 1.7.7 * Sat Apr 30 2005 David Malcolm - 2.2.2-2 - Finished porting conduits to pilot-link-0.12 API; re-enabled pilot support (#152172) foomatic-3.0.2-19 ----------------- * Tue May 03 2005 Tim Waugh 3.0.2-19 - Add IEEE 1284 ID for Epson Stylus CX5400 (bug #156661). * Tue Apr 12 2005 Tim Waugh - Fix Postscript driver (bug #151645). - Add IEEE 1284 ID for HP DeskJet 5150 (bug #154518). - Add IEEE 1284 ID for HP LaserJet 2420 (bug #114191). * Thu Mar 24 2005 Tim Waugh - Add a hook to remove any foomatic data cached by system-config-printer. gdb-6.3.0.0-1.21 ---------------- * Tue May 03 2005 Jeff Johnston 6.3.0.0-1.21 - Bump up release number. * Tue May 03 2005 Jeff Johnston 6.3.0.0-1.20 - Bump up release number. * Tue May 03 2005 Jeff Johnston 6.3.0.0-1.19 - Fix for partial die in cache error - Bugzilla 137904 gdm-1:2.6.0.8-12 ---------------- * Tue May 03 2005 Ray Strode 1:2.6.0.8-12 - Fix processing of non-ascii characters that got broken in 2.6.0.8-11, found by Miloslav Trmac , (bug 156590). gjdoc-0.7.4-4 ------------- * Thu Apr 28 2005 Andrew Overholt 0.7.4-4 - Add patches to ignore option case and deal with some error cases (Julian Scheid). gnome-media-2.10.2-1 -------------------- * Mon May 02 2005 John (J5) Palmieri 2.10.2-1 - Update to 2.10.2 gnome-panel-2.10.1-8 -------------------- * Mon May 02 2005 Mark McLoughlin 2.10.1-8 - Update to new OpenOffice.org .desktop file locations in openoffice.org-1.9.97-3 (bug #156064) * Wed Apr 27 2005 Mark McLoughlin - 2.10.1-7 - Add patch to clamp the size of the icons on the panel at 48x48. Fixes "moved the panel to the side, can't move it back" issue (rh #141743) * Wed Apr 27 2005 Mark McLoughlin 2.10.1-6 - Reference the OpenOffice.org Impress .desktop file correctly gnome-pilot-2.0.13-2 -------------------- * Fri Apr 29 2005 David Malcolm - 2.0.13-2 - move .desktop file from /usr/share/control-center-2.0/capplets to /usr/share/applications (#149228) gnome-vfs2-2.10.0-5 ------------------- * Fri Apr 29 2005 David Zeuthen 2.10.0-5 - Tweak the patch a bit so it applies * Fri Apr 29 2005 David Zeuthen 2.10.0-4 - Make local neon copy support gssapi correctly (#150132) gnupg-1.4.1-2 ------------- * Thu Apr 28 2005 Nalin Dahyabhai 1.4.1-2 - add -Wa,--noexecstack back to CFLAGS when invoking configure, the --enable-noexecstack flag only seems to affect asm modules gnutls-1.0.25-1 --------------- * Sat Apr 30 2005 Tomas Mraz 1.0.25-1 - new upstream version fixes potential DOS attack gstreamer-0.8.10-1 ------------------ * Tue May 03 2005 John (J5) Palmieri 0.8.10-1 - Update to upstream 0.8.10 * Thu Mar 17 2005 Colin Walters 0.8.9-4 - Rebuild to make it through beehive * Thu Mar 03 2005 John (J5) Palmieri 0.8.9-3 - add gstreamer-0.8.9-cast-fix.patch which casts the variable before sending it into the macro - update openjade hack to refrence xml-dtd-4.2-1.0-26 jwhois-3.2.2-14 --------------- * Sat Apr 30 2005 Miloslav Trmac - 3.2.2-14 - Add an AfriNIC range (#156178) kernel-2.6.11-1.1284_FC4 ------------------------ * Tue May 03 2005 Dave Jones - Disable usbmon/debugfs again for now until SELinux policy is fixed. * Mon May 02 2005 David Woodhouse - Make kallsyms include platform-specific symbols - Fix might_sleep warning in pbook clock-spreading fix libgnomeprint22-2.10.1-3 ------------------------ * Fri Apr 29 2005 John (J5) Palmieri - 2.10.1-3 - Added patch to make sure glyphs are layed out right (Bug #154939) libgtop2-2.10.1-1 ----------------- * Fri Apr 29 2005 David Zeuthen - 2.10.1-1 - New upstream version (#155188) * Fri Mar 18 2005 David Zeuthen - 2.10.0-2 - Rebuilt * Fri Mar 18 2005 David Zeuthen - 2.10.0-1 - Even newer upstream version libuser-0.53.7-1 ---------------- * Sat Apr 30 2005 Miloslav Trmac - 0.53.7-1 - Rebuild with updated translations, add missing translations. libwpd-0.8.1-1 -------------- * Fri Apr 29 2005 Caolan McNamara 0.8.1-1 - bump to latest version kudos Fridrich Strba - drop integrated patch mkinitrd-4.2.10-1 ----------------- * Tue May 03 2005 Peter Jones - 4.2.10-1 - set umask explicitly mod_jk-0:1.2.6-3jpp_2fc ----------------------- * Thu Apr 28 2005 Gary Benson - 0:1.2.6-3jpp_2fc - Actually build more than just an empty debuginfo package (#151415). mx4j-1:2.1.0-1jpp_5fc --------------------- * Fri Apr 29 2005 Gary Benson 0:2.1.0-1jpp_5fc - BC-compile the combined jarfile. pam-0.79-8 ---------- * Mon May 02 2005 Tomas Mraz 0.79-8 - pam_console: support loading .perms files in the console.perms.d (#156069) patch-2.5.4-23 -------------- * Fri Apr 29 2005 Tim Waugh 2.5.4-23 - Applied patch from Toshio Kuratomi to avoid problems with DOS-format newlines (bug #154283). * Wed Mar 02 2005 Tim Waugh 2.5.4-22 - Rebuild for new GCC. * Wed Feb 09 2005 Tim Waugh 2.5.4-21 - Rebuilt. perl-Frontier-RPC-0.06-39 ------------------------- * Sat Apr 30 2005 Jose Pedro Oliveira - 0.06-39 - Source URL: using the Comprehensive Perl Arcana Society Tapestry address (Frontier::RPC version 0.06 no longer available in CPAN mirrors). - spec cleanup (#156480) policycoreutils-1.23.6-2 ------------------------ * Fri Apr 29 2005 Dan Walsh 1.23.6-2 - Change -f flag in fixfiles to remove stuff from /tmp - Change -F flag to pass -F flag to restorecon/fixfiles. (IE Force relabel). qt-1:3.3.4-12 ------------- * Thu Apr 14 2005 Than Ngo 1:3.3.4-12 - fix bad symlink #154086 redhat-artwork-0.122-5 ---------------------- * Tue May 03 2005 John (J5) Palmieri 0.122-5 - Add a symlink for ftp share icons rpm-4.4.1-17 ------------ * Tue May 03 2005 Paul Nasrat - 4.4.1-17 - Fix typo * Tue May 03 2005 Paul Nasrat - 4.4.1-16 - Yet more matchpathcon * Tue May 03 2005 Paul Nasrat - 4.4.1-15 - Some more matchpathcon work shadow-utils-2:4.0.7-7 ---------------------- * Fri Apr 29 2005 Jeremy Katz - 2:4.0.7-7 - don't assume selinux is enabled if is_selinux_enabled() returns -1 * Mon Apr 18 2005 Peter Vrabec 2:4.0.7-6 - fix chage -l option (#109499, #137498) * Mon Apr 04 2005 Peter Vrabec 2:4.0.7-5 - fix memory leak, and CPU spinning when grp_update() and duplicate group entries in /etc/group (#151484) tcpdump-14:3.8.2-12 ------------------- * Thu Apr 28 2005 Martin Stransky - 14:3.8.2-12 - fix for CAN-2005-1280 Multiple DoS issues in tcpdump (CAN-2005-1279 CAN-2005-1278), #156041 tzdata-2005i-2 -------------- * Sat Apr 30 2005 Jakub Jelinek 2005i-2 - 2005i - updates for Iran, Haiti and Nicaragua util-linux-2.12p-9.3 -------------------- * Wed May 04 2005 Jeremy Katz - 2.12p-9.3 - rebuild against new libe2fsprogs (and libblkid) to fix cramfs auto-detection * Mon May 02 2005 Karel Zak 2.12p-9.2 - rebuild * Mon May 02 2005 Karel Zak 2.12p-9 - fix #156597 - look - doesn't work with separators vte-0.11.13-1.fc4 ----------------- * Thu Apr 28 2005 Warren Togami 0.11.13-1 - 0.11.13, all patches are now upstream words-3.0-7 ----------- * Mon May 02 2005 Karel Zak 3-7 - sort with --dictionary-order - remove words with possessives ('s) xerces-j2-0:2.6.2-4jpp_4fc -------------------------- * Fri Apr 29 2005 Gary Benson 0:2.6.2-4jpp_4fc - BC-compile. xorg-x11-6.8.2-30 ----------------- * Sat Apr 30 2005 Mike A. Harris 6.8.2-30 - Disabled xfs.init-fc4-startearly.patch as it breaks systems that /usr is on NFS. (FC4Blocker #156413) xscreensaver-1:4.21-2 --------------------- * Tue May 03 2005 Ray Strode 1:4.21-2 - Use absolute filenames for screenhacks so we don't pull in screenhacks from PATH (bug 151677). - Don't try to ping in sonar screensaver (bug 139692). From ivazquez at ivazquez.net Wed May 4 12:19:17 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Wed, 04 May 2005 08:19:17 -0400 Subject: rawhide report: 20050504 changes In-Reply-To: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> References: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> Message-ID: <1115209157.5462.38.camel@ignacio.ignacio.lan> On Wed, 2005-05-04 at 08:15 -0400, Build System wrote: > Updated Packages: Whee! Update time... -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From peter.backlund at home.se Wed May 4 12:40:10 2005 From: peter.backlund at home.se (Peter Backlund) Date: Wed, 04 May 2005 14:40:10 +0200 Subject: rawhide report: 20050504 changes In-Reply-To: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> References: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> Message-ID: <1115210410.18426.24.camel@localhost.localdomain> > evolution-2.2.2-3 > ----------------- > * Sat Apr 30 2005 David Malcolm - 2.2.2-3 > - updated mozilla_build_version to 1.7.7 Any plans to use Firefox as Gecko-provider instead of Mozilla? A boatload of apps require mozilla-nspr/-nss, but afaics the same files are provided by firefox: [root at localhost ~]# for p in `rpm -ql mozilla-nss mozilla-nspr`; do rpm -ql firefox | grep `basename $p`; done; /usr/lib/firefox-1.0.3/libnss3.so /usr/lib/firefox-1.0.3/libnssckbi.so /usr/lib/firefox-1.0.3/libsmime3.so /usr/lib/firefox-1.0.3/libsoftokn3.so /usr/lib/firefox-1.0.3/libssl3.so /usr/lib/firefox-1.0.3/libnssckbi.so /usr/lib/firefox-1.0.3/libnspr4.so /usr/lib/firefox-1.0.3/libplc4.so /usr/lib/firefox-1.0.3/libplds4.so Also, apps like epiphany, galeon, yelp and liferea are built against the gecko provided by mozilla instead of firefox. /Peter From arjanv at redhat.com Wed May 4 12:47:24 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Wed, 04 May 2005 08:47:24 -0400 Subject: rawhide report: 20050504 changes In-Reply-To: <1115210410.18426.24.camel@localhost.localdomain> References: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> <1115210410.18426.24.camel@localhost.localdomain> Message-ID: <1115210844.6053.1.camel@localhost.localdomain> On Wed, 2005-05-04 at 14:40 +0200, Peter Backlund wrote: > > evolution-2.2.2-3 > > ----------------- > > * Sat Apr 30 2005 David Malcolm - 2.2.2-3 > > - updated mozilla_build_version to 1.7.7 > > Any plans to use Firefox as Gecko-provider instead of Mozilla? A > boatload of apps require mozilla-nspr/-nss, but afaics the same files > are provided by firefox: the real answer I guess is a libgecko RPM which is sort of independent from both firefox and mozilla and is used by all things in the distro. But that is a change that is probably too invasive for this stage in the fc4 cycle. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From fedora at camperquake.de Wed May 4 13:20:25 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Wed, 4 May 2005 15:20:25 +0200 Subject: rawhide report: 20050504 changes In-Reply-To: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> References: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> Message-ID: <20050504132025.GA25692@ryoko.camperquake.de> On Wed, May 04, 2005 at 08:15:12AM -0400, Build System wrote: > xorg-x11-6.8.2-30 > ----------------- > * Sat Apr 30 2005 Mike A. Harris 6.8.2-30 > - Disabled xfs.init-fc4-startearly.patch as it breaks systems that /usr is > on NFS. (FC4Blocker #156413) Does that mean 'no more early logon for noone'? From mike at navi.cx Wed May 4 13:37:52 2005 From: mike at navi.cx (Mike Hearn) Date: Wed, 04 May 2005 14:37:52 +0100 Subject: applications which depend on esound References: <1115186392.4047.10.camel@marte.biciclete.ro> <1115201427.3021.682.camel@tux.lan> Message-ID: On Wed, 04 May 2005 12:10:28 +0200, Ronald S. Bultje wrote: > No. The only actual dependency is in libgnome/&ui, and the others solely > depend on esd because libgnome/&ui do. In other words building with --as-needed would eliminate most of those esound dependencies. Unfortunately esound is exposed in libgnome so cannot be dropped unless a fully compatible replacement is produced. But that's no big deal. Esound isn't that large. thanks -mike From matthew at nocturnal.org Wed May 4 14:36:37 2005 From: matthew at nocturnal.org (Matthew Lenz) Date: Wed, 04 May 2005 09:36:37 -0500 Subject: applications which depend on esound In-Reply-To: References: <1115186392.4047.10.camel@marte.biciclete.ro> <1115201427.3021.682.camel@tux.lan> Message-ID: <1115217397.13542.4.camel@mlenzdesktop> On Wed, 2005-05-04 at 14:37 +0100, Mike Hearn wrote: > On Wed, 04 May 2005 12:10:28 +0200, Ronald S. Bultje wrote: > > No. The only actual dependency is in libgnome/&ui, and the others solely > > depend on esd because libgnome/&ui do. > > Unfortunately esound is exposed in libgnome so cannot > be dropped unless a fully compatible replacement is produced. But that's > no big deal. Esound isn't that large. Quoting Marius Andreiana ... "Note2: dmix doesn't solve "sound over network" problem. This still requires a sound server (e.g. MAS, Jack, esd), but the sound server can be implemented at gstreamer framework level, not application level." gnome has an esound replacement, gstreamer. gstreamer is what needs to talk to a network audio server. > thanks -mike > From caillon at redhat.com Wed May 4 15:02:28 2005 From: caillon at redhat.com (Christopher Aillon) Date: Wed, 04 May 2005 11:02:28 -0400 Subject: rawhide report: 20050504 changes In-Reply-To: <1115210410.18426.24.camel@localhost.localdomain> References: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> <1115210410.18426.24.camel@localhost.localdomain> Message-ID: <4278E404.2040000@redhat.com> Peter Backlund wrote: >>evolution-2.2.2-3 >>----------------- >>* Sat Apr 30 2005 David Malcolm - 2.2.2-3 >>- updated mozilla_build_version to 1.7.7 > > > Any plans to use Firefox as Gecko-provider instead of Mozilla? This will happen in the FC5 time frame. I am already working on splitting these out. > A boatload of apps require mozilla-nspr/-nss, but afaics the same files > are provided by firefox: > > [root at localhost ~]# for p in `rpm -ql mozilla-nss mozilla-nspr`; do rpm > -ql firefox | grep `basename $p`; done; > /usr/lib/firefox-1.0.3/libnss3.so > /usr/lib/firefox-1.0.3/libnssckbi.so > /usr/lib/firefox-1.0.3/libsmime3.so > /usr/lib/firefox-1.0.3/libsoftokn3.so > /usr/lib/firefox-1.0.3/libssl3.so > /usr/lib/firefox-1.0.3/libnssckbi.so > /usr/lib/firefox-1.0.3/libnspr4.so > /usr/lib/firefox-1.0.3/libplc4.so > /usr/lib/firefox-1.0.3/libplds4.so It installs the files, but does not Provide them. We strip out those provides in the spec > > Also, apps like epiphany, galeon, yelp and liferea are built against the > gecko provided by mozilla instead of firefox. And this is a problem because why? Firefox 1.0.x and Mozilla 1.7.x are built off of the same Gecko source (there are a few really minor differences, but for embedding purposes, the APIs are all the same). The major differences are in UI and a few UI components, which embeddors don't, specifically the ones you mentioned, don't care about anyway. Mozilla Drivers are keeping the two branches (firefox 1.0/aviary and mozilla 1.7.x) in sync, with embeddors in mind. From mbneto at gmail.com Wed May 4 15:38:48 2005 From: mbneto at gmail.com (mbneto) Date: Wed, 4 May 2005 11:38:48 -0400 Subject: Boot process stops at level 3 and 5 (or rawhide ate my baby) Message-ID: <5cf776b80505040838612d8310@mail.gmail.com> Since saturday my fc3+rawhide-updates system started to act strangely. When I boot runlevel 5 or 3 it hangs during the boot process. I can't ctrl-alt-del. It happens with all kernel versions that I have, even "old" ones that worked before saturday, like 2.6.11-1.1240_FC4. If I boot single mode, do a service network start and init 3 I can use the machine ok. Any ideas of what may be wrong ? or even how can I provide more information about it ? From selinux at gmail.com Wed May 4 15:45:24 2005 From: selinux at gmail.com (Tom London) Date: Wed, 4 May 2005 08:45:24 -0700 Subject: Boot process stops at level 3 and 5 (or rawhide ate my baby) In-Reply-To: <5cf776b80505040838612d8310@mail.gmail.com> References: <5cf776b80505040838612d8310@mail.gmail.com> Message-ID: <4c4ba15305050408457735ab52@mail.gmail.com> Booting with 'enforcing=0' on the grub line worked for me. -- Tom London From caillon at redhat.com Wed May 4 16:10:58 2005 From: caillon at redhat.com (Christopher Aillon) Date: Wed, 04 May 2005 12:10:58 -0400 Subject: rawhide report: 20050504 changes In-Reply-To: <1115210844.6053.1.camel@localhost.localdomain> References: <200505041215.j44CFC3t000449@porkchop.devel.redhat.com> <1115210410.18426.24.camel@localhost.localdomain> <1115210844.6053.1.camel@localhost.localdomain> Message-ID: <4278F412.7060205@redhat.com> Arjan van de Ven wrote: > On Wed, 2005-05-04 at 14:40 +0200, Peter Backlund wrote: > >>>evolution-2.2.2-3 >>>----------------- >>>* Sat Apr 30 2005 David Malcolm - 2.2.2-3 >>>- updated mozilla_build_version to 1.7.7 >> >>Any plans to use Firefox as Gecko-provider instead of Mozilla? A >>boatload of apps require mozilla-nspr/-nss, but afaics the same files >>are provided by firefox: > > > the real answer I guess is a libgecko RPM which is sort of independent > from both firefox and mozilla and is used by all things in the distro. > But that is a change that is probably too invasive for this stage in the > fc4 cycle. This is the current upstream plan: http://wiki.mozilla.org/XUL:Xul_Runner But, this is all FC5 time frame stuff... From dragoran at feuerpokemon.de Wed May 4 16:13:04 2005 From: dragoran at feuerpokemon.de (dragoran) Date: Wed, 04 May 2005 18:13:04 +0200 Subject: gcc4 slower than gcc3.4.3 ? Message-ID: <4278F490.5010905@feuerpokemon.de> http://www.coyotegulch.com/reviews/gcc4/index.html are this bugs in the compiler or does this apps needs to be optimized for gcc4? From sundaram at redhat.com Wed May 4 16:17:18 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Wed, 04 May 2005 21:47:18 +0530 Subject: gcc4 slower than gcc3.4.3 ? In-Reply-To: <4278F490.5010905@feuerpokemon.de> References: <4278F490.5010905@feuerpokemon.de> Message-ID: <4278F58E.4080205@redhat.com> dragoran wrote: > http://www.coyotegulch.com/reviews/gcc4/index.html > are this bugs in the compiler or does this apps needs to be optimized > for gcc4? > The infrastructure for optimisation [1] has already been merged into GCC with the new major release of 4.0 but significant performance improvements can only be expected in the subsequent minor revisions. regards Rahul [1] http://lwn.net/Articles/84888/ From mls at suse.de Wed May 4 18:26:44 2005 From: mls at suse.de (Michael Schroeder) Date: Wed, 4 May 2005 20:26:44 +0200 Subject: rpm's zlib and deltarpms Message-ID: <20050504182644.GA3024@suse.de> Hi folks, I just found out that the rsync patch applied to rpm's internal zlib library makes it produce zlib-incompatible payloads. That is, a decompress and recompress with normal zlib doesn't produce the same output. Normally this doesn't do much harm, but with deltarpms I need to reconstruct the exact compressed payload. So at the moment deltarpms won't work with FC4 rpms. Seems like I have to add the patched zlib version to the deltarpm package and somehow detect which version to use. (Btw, the zlib license states that altered source versions must be "plainly marked". Please do so. See zlib FAQ entry #24.) Sigh, Michael. -- Michael Schroeder mls at suse.de main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} From n3npq at nc.rr.com Wed May 4 18:48:25 2005 From: n3npq at nc.rr.com (Jeff Johnson) Date: Wed, 04 May 2005 14:48:25 -0400 Subject: rpm's zlib and deltarpms In-Reply-To: <20050504182644.GA3024@suse.de> References: <20050504182644.GA3024@suse.de> Message-ID: <427918F9.4040706@nc.rr.com> Michael Schroeder wrote: >Hi folks, > >I just found out that the rsync patch applied to rpm's internal >zlib library makes it produce zlib-incompatible payloads. That is, >a decompress and recompress with normal zlib doesn't produce the >same output. Normally this doesn't do much harm, but with deltarpms >I need to reconstruct the exact compressed payload. So at the >moment deltarpms won't work with FC4 rpms. > > Link with -lrpmio to use same zlib that rpm uses. >Seems like I have to add the patched zlib version to the deltarpm >package and somehow detect which version to use. > >(Btw, the zlib license states that altered source versions must be >"plainly marked". Please do so. See zlib FAQ entry #24.) > > From CHANGES, and has been in change logs: - add https://svn.uhulinux.hu/packages/dev/zlib/patches/02-rsync.patch Plaintome. 73 de Jeff From njp at santram.co.uk Wed May 4 18:48:29 2005 From: njp at santram.co.uk (Neil J. Patel) Date: Wed, 04 May 2005 19:48:29 +0100 Subject: Boot process stops at level 3 and 5 (or rawhide ate my baby) In-Reply-To: <4c4ba15305050408457735ab52@mail.gmail.com> References: <5cf776b80505040838612d8310@mail.gmail.com> <4c4ba15305050408457735ab52@mail.gmail.com> Message-ID: <1115232509.10335.30.camel@localhost.localdomain> On Wed, 2005-05-04 at 08:45 -0700, Tom London wrote: > Booting with 'enforcing=0' on the grub line worked for me. > > > -- > Tom London > Also, i had a similar problem in that boot would hang at Smartd, stopping that and sendmail using chkconfig worked for me. From mls at suse.de Wed May 4 19:15:17 2005 From: mls at suse.de (Michael Schroeder) Date: Wed, 4 May 2005 21:15:17 +0200 Subject: rpm's zlib and deltarpms In-Reply-To: <427918F9.4040706@nc.rr.com> References: <20050504182644.GA3024@suse.de> <427918F9.4040706@nc.rr.com> Message-ID: <20050504191516.GA26676@suse.de> On Wed, May 04, 2005 at 02:48:25PM -0400, Jeff Johnson wrote: > >I just found out that the rsync patch applied to rpm's internal > >zlib library makes it produce zlib-incompatible payloads. That is, > >a decompress and recompress with normal zlib doesn't produce the > >same output. Normally this doesn't do much harm, but with deltarpms > >I need to reconstruct the exact compressed payload. So at the > >moment deltarpms won't work with FC4 rpms. > > > > > > Link with -lrpmio to use same zlib that rpm uses. Doesn't help, it's not about the system where applydeltarpm or createdeltarpm is running on, but the system where the rpm got built. > From CHANGES, and has been in change logs: > - add https://svn.uhulinux.hu/packages/dev/zlib/patches/02-rsync.patch Please do take a look into zlib FAQ entry #24 ;-) Michael. -- Michael Schroeder mls at suse.de main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} From mbneto at gmail.com Wed May 4 19:44:46 2005 From: mbneto at gmail.com (mbneto) Date: Wed, 4 May 2005 15:44:46 -0400 Subject: Boot process stops at level 3 and 5 (or rawhide ate my baby) In-Reply-To: <1115232509.10335.30.camel@localhost.localdomain> References: <5cf776b80505040838612d8310@mail.gmail.com> <4c4ba15305050408457735ab52@mail.gmail.com> <1115232509.10335.30.camel@localhost.localdomain> Message-ID: <5cf776b80505041244810f611@mail.gmail.com> Since this seems to be a strange behaviour should we open a bug to grub ? From selinux at gmail.com Wed May 4 19:56:59 2005 From: selinux at gmail.com (Tom London) Date: Wed, 4 May 2005 12:56:59 -0700 Subject: Boot process stops at level 3 and 5 (or rawhide ate my baby) In-Reply-To: <5cf776b80505041244810f611@mail.gmail.com> References: <5cf776b80505040838612d8310@mail.gmail.com> <4c4ba15305050408457735ab52@mail.gmail.com> <1115232509.10335.30.camel@localhost.localdomain> <5cf776b80505041244810f611@mail.gmail.com> Message-ID: <4c4ba153050504125770ff5d1d@mail.gmail.com> On 5/4/05, mbneto wrote: > Since this seems to be a strange behaviour should we open a bug to grub ? > > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list > This is not caused by grub..... Something in the kernel or in the init scripts is causing this. With all the discussion, I presume the developers are fixing this now.... This looks like a bugzilla for it: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156860 tom -- Tom London From mihamina.rakotomandimby at etu.univ-orleans.fr Wed May 4 20:28:32 2005 From: mihamina.rakotomandimby at etu.univ-orleans.fr (Rakotomandimby (R12y) Mihamina) Date: Wed, 04 May 2005 22:28:32 +0200 Subject: What is wrong in this spec file? Message-ID: <1115238512.23655.6.camel@fctmp> Hi, Attached is one spec file, for the findlib package. It has been from a src.rpm of the altlinux distribution I made just minor changes to suit it to my fedora core3. Near the end of the packaging process, the stuff checks the "prelinking". Okay, but the problem is it say absolutely anything, as you see at the bottom of this message. That was the first thing. The second is that if a make a : # rpm -qlp /usr/src/redhat/RPMS/i386/findlib-1.0.4-fc3.i386.rpm It shows me my entire system!! I think this thing has embeded all my PATH into the rpm pacakge! I guess something would be wrong into my spec file, but I cant find where. Would you help me? ======= end of rpmbuild -bb findlib.spec ======== [.....] + mkdir -p /usr/share/doc/findlib-1.0.4 + cp README INSTALL LICENSE /usr/share/doc/findlib-1.0.4/ + mkdir -p /usr/lib/ocaml/ocamlfind-mini/ + cp mini/README mini/ocamlfind-mini /usr/lib/ocaml/ocamlfind-mini/ + /usr/lib/rpm/find-debuginfo.sh /usr/src/redhat/BUILD/findlib-1.0.4 0 blocks + /usr/lib/rpm/redhat/brp-compress + /usr/lib/rpm/redhat/brp-strip /usr/bin/strip + /usr/lib/rpm/redhat/brp-strip-static-archive /usr/bin/strip + /usr/lib/rpm/redhat/brp-strip-comment- note /usr/bin/strip /usr/bin/objdump Processing files: findlib-1.0.4-fc3 Executing(%doc): /bin/sh -e /var/tmp/rpm-tmp.89611 + umask 022 + cd /usr/src/redhat/BUILD + cd findlib-1.0.4 + DOCDIR=/usr/share/doc/findlib-1.0.4 + export DOCDIR + rm -rf /usr/share/doc/findlib-1.0.4 + /bin/mkdir -p /usr/share/doc/findlib-1.0.4 + cp -pr README doc/ INSTALL LICENSE /usr/share/doc/findlib-1.0.4 + exit 0 prelink: /usr/bin/bdfindex: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/disol: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/dltest: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/fmtest: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/isql: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/iusql: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/kban: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/ktest: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/odbcinst: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/vfperf: at least one of file's dependencies has changed since prelinking prelink: /usr/bin/vftest: at least one of file's dependencies has changed since prelinking Provides: xchat(EXPORTED) Requires(rpmlib): rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib (PartialHardlinkSets) <= 4.0.4-1 rpmlib(PayloadFilesHavePrefix) <= 4.0-1 Requires: /bin/awk /bin/bash /bin/sh /usr/bin/env /usr/bin/ocamlrun /usr/bin/perl /usr/bin/python gtk-doc lib-javax-activation-20030319.so lib-javax-mail-20031006.so lib-javax-xml-parsers-2.2.1.so lib-javax-xml-transform-2.4.1.so lib-org-apache-bcel-5.0.so lib-org-apache-commons-logging-1.0.2.so lib-org-apache-regexp-1.2.so lib-org-apache-tools-ant-1.5.2.so lib-org-apache-tools-bzip2-1.5.2.so lib-org-apache-tools-mail-1.5.2.so lib-org-apache-tools-tar-1.5.2.so lib-org-apache-tools-zip-1.5.2.so lib-org-apache-xalan-2.4.1.so lib-org-apache-xerces-2.2.1.so lib-org-xml-sax-2.2.1.so libFLAC.so.4 libGL.so.1 libGLU.so.1 libICE.so.6 libIDL-0.6.so.0 libIDL-2.so.0 libIIOP.so.0 libMagick.so.6 libORBit-2.so.0 libORBit.so.0 libORBitCosNaming-2.so.0 libORBitCosNaming.so.0 libORBitutil.so.0 libOggFLAC.so.1 libSDL-1.2.so.0 libSDL_image-1.2.so.0 libSDL_mixer-1.2.so.0 libSDL_net-1.2.so.0 libSM.so.6 libVFlib2.so.24 libX11.so.6 libXaw.so.7 libXaw3d.so.7 libXcursor.so.1 libXext.so.6 libXft.so.2 libXi.so.6 libXinerama.so.1 libXmu.so.6 libXpm.so.4 libXrandr.so.2 libXrender.so.1 libXss.so.1 libXt.so.6 libXtst.so.6 libXv.so.1 libXxf86dga.so.1 libXxf86vm.so.1 libaa.so.1 libacl.so.1 libacl.so.1(ACL_1.0) libao.so.2 libapr-0.so.0 libaprutil-0.so.0 libart_lgpl.so.2 libart_lgpl_2.so.2 libartsc.so.0 libartsflow.so.1 libartsflow_idl.so.1 libasound.so.2 libasound.so.2(ALSA_0.9) libasound.so.2(ALSA_0.9.0rc4) libasound.so.2(ALSA_0.9.0rc8) libasound.so.2(ALSA_0.9.5) libaspell.so.15 libatk-1.0.so.0 libattr.so.1 libattr.so.1(ATTR_1.0) libaudiofile.so.0 libbeecrypt.so.6 libbfd-2.15.92.0.2.so libbonobo-2.so.0 libbonobo-activation.so.4 libbonobo-print.so.2 libbonobo.so.2 libbonoboui-2.so.0 libbonobox.so.2 libbz2.so.1 libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1) libc.so.6(GLIBC_2.1.1) libc.so.6(GLIBC_2.1.2) libc.so.6(GLIBC_2.1.3) libc.so.6(GLIBC_2.2) libc.so.6(GLIBC_2.2.3) libc.so.6(GLIBC_2.2.4) libc.so.6(GLIBC_2.3) libc.so.6(GLIBC_2.3.2) libc.so.6(GLIBC_2.3.3) libc.so.6(GLIBC_2.3.4) libc.so.6(GLIBC_PRIVATE) libcapi20.so.2 libcdda_interface.so.0 libcdda_paranoia.so.0 libcddb-slave2.so.0 libcom_err.so.2 libcroco-0.6.so.3 libcrypt.so.1 libcrypt.so.1(GLIBC_2.0) libcrypto.so.4 libcspi.so.0 libcups.so.2 libcurl.so.3 libdb-3.3.so libdb-4.1.so libdb-4.2.so libdb.so.2 libdb.so.2(GLIBC_2.0) libdb.so.3 libdb.so.3(GLIBC_2.1) libdbus-1.so.0 libdbus-glib-1.so.0 libdes425.so.3 libdia.so libdl.so.2 libdl.so.2(GLIBC_2.0) libdl.so.2(GLIBC_2.1) libdns.so.16 libdv.so.4 libdvdread.so.3 libdw.so.1 libdw.so.1(ELFUTILS_1.0) libe2p.so.2 libecal.so.6 libedataserver.so.3 libeel-2.so.2 libefs.so.1 libelf.so.1 libelf.so.1(ELFUTILS_1.0) libelf.so.1(ELFUTILS_1.1) libemiscwidgets.so.0 libesd.so.0 libeshell.so.0 libestbase.so.1.2.2.1 libeststring.so.1.2 libeutil.so.0 libevolution-a11y.so.0 libevolution-importer.so.0 libevolution-widgets-a11y.so.0 libexif.so.9 libexpat.so.0 libexslt.so.0 libext2fs.so.2 libfaad.so.0 libfontconfig.so.1 libfreetype.so.6 libfribidi.so.0 libgailutil.so.17 libgaim-remote.so.0 libgal-2.2.so.1 libgal-a11y-2.2.so.1 libgal.so.23 libgcc_s.so.1 libgcc_s.so.1(GCC_3.0) libgcc_s.so.1(GCC_3.3) libgcc_s.so.1(GLIBC_2.0) libgcj.so.5 libgconf-1.so.1 libgconf-2.so.4 libgconf-gtk-1.so.1 libgcrypt.so.11 libgdbm.so.2 libgdk-1.2.so.0 libgdk-x11-2.0.so.0 libgdk_imlib.so.1 libgdk_pixbuf-2.0.so.0 libgdk_pixbuf.so.2 libgdk_pixbuf_ [..... there is still lot of crazyness but let's stop here ....] ==================================================================== -- Get a fully managed dedicated server for ?200/month ($257/month) No time limit for taking care of your server. You keep the "root" acces if you want. Billing periods are 3 months. See the conditions at http://aspo.rktmb.org/activities/managed_servers -------------- next part -------------- %define ocamlver 3.08.3 Name: findlib Version: 1.0.4 Release: fc3 Group: Development/ML Summary: A module packaging tool for OCaml License: GPL Packager: Mihamina Rakotomandimby Url: http://www.ocaml-programming.de/packages/documentation/findlib/ Source: http://www.ocaml-programming.de/packages/%name-%version.tar.gz Requires: ocaml = %ocamlver gtk-doc # Automatically added by buildreq on Mon Oct 13 2003 BuildRequires: camlp4 labltk ncurses-devel # ocaml-runtime BuildRequires: ocaml = %ocamlver %package -n ocamlfind-mini Summary: Minimal findlib script to be distributed with user libraries Group: Development/ML Requires: %name = %version-%release %description The "findlib" library provides a scheme to manage reusable software components (packages), and includes tools that support this scheme. Packages are collections of OCaml modules for which metainformation can be stored. The packages are kept in the filesystem hierarchy, but with strict directory structure. The library contains functions to look the directory up that stores a package, to query metainformation about a package, and to retrieve dependency information about multiple packages. There is also a tool that allows the user to enter queries on the command-line. In order to simplify compilation and linkage, there are new frontends of the various OCaml compilers that can directly deal with packages. %description -n ocamlfind-mini ocamlfind-mini is an O'Caml script that implements a subset of the full functionality of ocamlfind. It consists only of one file, so it is easy to distribute it with any software. The subset is normally sufficient to compile a library and to install the library; but it is insufficient to link the library into an executable. %prep %setup -q %build ./configure -mandir /usr/share/man -config /usr/lib/ocaml/etc/findlib.conf -with-toolbox make all opt %install %define _compress_method skip make prefix=$RPM_BUILD_ROOT install mkdir -p $RPM_BUILD_ROOT%_docdir/%name-%version cp README INSTALL LICENSE $RPM_BUILD_ROOT%_docdir/%name-%version/ mkdir -p $RPM_BUILD_ROOT/usr/lib/ocaml/ocamlfind-mini/ cp mini/* $RPM_BUILD_ROOT/usr/lib/ocaml/ocamlfind-mini/ %files -n ocamlfind-mini %_libdir/ocaml/ocamlfind-mini/* %files %_bindir/* %_libdir/ocaml/etc/* %_libdir/ocaml/topfind %_libdir/ocaml/site-lib/* %_mandir/man?/* %doc README doc/ INSTALL LICENSE %changelog * Mon Apr 2 2005 Mihamina Rakotomandimby 1.0.4-fc3 - Vitaly Lugovsky package modification to suit it to Fedora Core 3 - rebuild * Tue Oct 26 2004 Vitaly Lugovsky 1.0.4-alt1 - rebuild * Sat Jul 17 2004 Vitaly Lugovsky 1.0.4-alt0.2 - rebuild * Wed Jul 07 2004 Vitaly Lugovsky 1.0.4-alt0.1 - rebuild * Fri May 7 2004 Alexander V. Nikolaev 0.9-alt2.1 - Non-maintainer upload - Add "packager" to spec - Rebuild with glibc 2.3.x and ocaml 3.07-alt6.1 * Tue Jan 27 2004 Vitaly Lugovsky 0.9-alt2 threads library issue fixed * Tue Dec 16 2003 Vitaly Lugovsky 0.9-alt1.1 rebuild * Wed Oct 08 2003 Vitaly Lugovsky 0.9-alt1 - A new version * Wed Aug 27 2003 Vitaly Lugovsky 0.8-alt3s - rebuild * Thu Mar 06 2003 Vitaly Lugovsky 0.8-alt2s - added static libraries METAs * Fri Jan 31 2003 Vitaly Lugovsky 0.8-alt1s - rebuild with ocaml-3.06 [Shared] * Sun Oct 27 2002 Vitaly Lugovsky 0.8-alt1 - new version * Sun Aug 18 2002 Vitaly Lugovsky 0.7.2-alt1 - new release *Tue Jul 30 2002 Vitaly Lugovsky 0.7.1-alt1 - new release *Mon Jun 24 2002 Vitaly Lugovsky 0.7-alt1 - new version released *Tue Apr 16 2002 Vitaly Lugovsky 0.6.2-alt6 - Rebuild with 3.04+9 *Sat Mar 2 2002 Vitaly Lugovsky 0.6.2-alt4 - Rebuild with ocaml-3.04+7-alt1 *Sun Feb 17 2002 Vitaly Lugovsky - Rebuild with ocaml-3.04-alt4 (shared patch disabled) *Mon Jan 14 2002 Vitaly Lugovsky - First RPM release. From jos at xos.nl Wed May 4 20:31:55 2005 From: jos at xos.nl (Jos Vos) Date: Wed, 4 May 2005 22:31:55 +0200 Subject: What is wrong in this spec file? In-Reply-To: <1115238512.23655.6.camel@fctmp>; from mihamina.rakotomandimby@etu.univ-orleans.fr on Wed, May 04, 2005 at 10:28:32PM +0200 References: <1115238512.23655.6.camel@fctmp> Message-ID: <20050504223155.A15595@xos037.xos.nl> On Wed, May 04, 2005 at 10:28:32PM +0200, Rakotomandimby (R12y) Mihamina wrote: > I guess something would be wrong into my spec file, but I cant find > where. Would you help me? What about a missing "BuildRoot" tag? ;-) -- -- Jos Vos -- X/OS Experts in Open Systems BV | Phone: +31 20 6938364 -- Amsterdam, The Netherlands | Fax: +31 20 6948204 From mls at suse.de Wed May 4 20:32:48 2005 From: mls at suse.de (Michael Schroeder) Date: Wed, 4 May 2005 22:32:48 +0200 Subject: What is wrong in this spec file? In-Reply-To: <1115238512.23655.6.camel@fctmp> References: <1115238512.23655.6.camel@fctmp> Message-ID: <20050504203248.GA19407@suse.de> On Wed, May 04, 2005 at 10:28:32PM +0200, Rakotomandimby (R12y) Mihamina wrote: > # rpm -qlp /usr/src/redhat/RPMS/i386/findlib-1.0.4-fc3.i386.rpm > > It shows me my entire system!! > I think this thing has embeded all my PATH into the rpm pacakge! No, just: %_bindir/* %_libdir/ocaml/etc/* %_libdir/ocaml/topfind %_libdir/ocaml/site-lib/* %_mandir/man?/* > I guess something would be wrong into my spec file, but I cant find > where. Would you help me? You have to define a BuildRoot. Cheers, Michael. -- Michael Schroeder mls at suse.de main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} From paul at city-fan.org Wed May 4 20:33:56 2005 From: paul at city-fan.org (Paul Howarth) Date: Wed, 04 May 2005 21:33:56 +0100 Subject: What is wrong in this spec file? In-Reply-To: <1115238512.23655.6.camel@fctmp> References: <1115238512.23655.6.camel@fctmp> Message-ID: <1115238836.10489.205.camel@laurel.intra.city-fan.org> On Wed, 2005-05-04 at 22:28 +0200, Rakotomandimby (R12y) Mihamina wrote: > Hi, > > Attached is one spec file, for the findlib package. > It has been from a src.rpm of the altlinux distribution > I made just minor changes to suit it to my fedora core3. > Near the end of the packaging process, the stuff checks the > "prelinking". > Okay, but the problem is it say absolutely anything, as you see at the > bottom of this message. > That was the first thing. The second is that if a make a : > > # rpm -qlp /usr/src/redhat/RPMS/i386/findlib-1.0.4-fc3.i386.rpm > > It shows me my entire system!! > I think this thing has embeded all my PATH into the rpm pacakge! > > I guess something would be wrong into my spec file, but I cant find > where. Would you help me? The main problem is that there's no BuildRoot. You'd probably have spotted this yourself if you weren't building RPMs as root (which is not a good idea at all). Paul. -- Paul Howarth From jspaleta at gmail.com Wed May 4 20:42:50 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Wed, 4 May 2005 16:42:50 -0400 Subject: rpm's zlib and deltarpms In-Reply-To: <20050504191516.GA26676@suse.de> References: <20050504182644.GA3024@suse.de> <427918F9.4040706@nc.rr.com> <20050504191516.GA26676@suse.de> Message-ID: <604aa7910505041342578fd4dd@mail.gmail.com> On 5/4/05, Michael Schroeder wrote: > Please do take a look into zlib FAQ entry #24 ;-) Better yet, file a bug against rpm in fedora bugzilla including the trivial patch to include the zlib version change in the zlib.h header. -jef"votes for ZLIB_VERSION "1.2.2.2f-the-version-of-doom" "spaleta From mattdm at mattdm.org Wed May 4 20:45:04 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Wed, 4 May 2005 16:45:04 -0400 Subject: What is wrong in this spec file? In-Reply-To: <20050504203248.GA19407@suse.de> References: <1115238512.23655.6.camel@fctmp> <20050504203248.GA19407@suse.de> Message-ID: <20050504204504.GA19714@jadzia.bu.edu> On Wed, May 04, 2005 at 10:32:48PM +0200, Michael Schroeder wrote: > No, just: > %_bindir/* > %_libdir/ocaml/etc/* > %_libdir/ocaml/topfind > %_libdir/ocaml/site-lib/* > %_mandir/man?/* [...] > You have to define a BuildRoot. I also find it nicer to not use so many wildcards in the %files listing. That way, if one of the expected files doesn't get built right, the RPM build fails. Won't help this problem, though. In fact, might make it worse since you might not notice what's happening. Although, if you avoid building RPMs as root, that'll pretty much avoid it too (and be much safer). -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 79 degrees Fahrenheit. From n3npq at nc.rr.com Wed May 4 21:36:11 2005 From: n3npq at nc.rr.com (Jeff Johnson) Date: Wed, 04 May 2005 17:36:11 -0400 Subject: rpm's zlib and deltarpms In-Reply-To: <604aa7910505041342578fd4dd@mail.gmail.com> References: <20050504182644.GA3024@suse.de> <427918F9.4040706@nc.rr.com> <20050504191516.GA26676@suse.de> <604aa7910505041342578fd4dd@mail.gmail.com> Message-ID: <4279404B.9030004@nc.rr.com> Jeff Spaleta wrote: >On 5/4/05, Michael Schroeder wrote: > > >>Please do take a look into zlib FAQ entry #24 ;-) >> >> > >Better yet, file a bug against rpm in fedora bugzilla including the >trivial patch to include the zlib version change in the zlib.h header. > > Make sure you include the RFE to distribute zlib.h and zconf.h with rpm too. 73 de Jeff From mihamina.rakotomandimby at etu.univ-orleans.fr Wed May 4 22:30:02 2005 From: mihamina.rakotomandimby at etu.univ-orleans.fr (Rakotomandimby (R12y) Mihamina) Date: Thu, 05 May 2005 00:30:02 +0200 Subject: generate src.rpm Message-ID: <1115245802.3609.2.camel@ngeza> Hello, I saw a lot of documentation on how to rebuild rpm, building rpm from an src.rpm, but: How to generate the src.rpm? Because. I used the src.rpm of someone to generate my rpms, and I would like people to be abble to use my src.rpms too. But I dont know how to generate one, from the sources, the patches and the specfile. Would you help me? just indicate a link if you have one. -- ASPO Infog?rance http://aspo.rktmb.org/activites/infogerance Unofficial FAQ fcolc http://faq.fcolc.eu.org/ LUG sur Orl?ans et alentours (France). T?l : 02 34 08 26 04 / 06 33 26 13 14 From paul at city-fan.org Wed May 4 22:30:40 2005 From: paul at city-fan.org (Paul Howarth) Date: Wed, 04 May 2005 23:30:40 +0100 Subject: generate src.rpm In-Reply-To: <1115245802.3609.2.camel@ngeza> References: <1115245802.3609.2.camel@ngeza> Message-ID: <1115245840.10489.210.camel@laurel.intra.city-fan.org> On Thu, 2005-05-05 at 00:30 +0200, Rakotomandimby (R12y) Mihamina wrote: > Hello, > > I saw a lot of documentation on how to rebuild rpm, building rpm from an > src.rpm, but: > > How to generate the src.rpm? > > Because. I used the src.rpm of someone to generate my rpms, and I would > like people to be abble to use my src.rpms too. > > But I dont know how to generate one, from the sources, the patches and > the specfile. > > Would you help me? just indicate a link if you have one. If you use "rpmbuild -ba" instead of "rpmbuild -bb" to build your RPMs, you'll get a src.rpm as well as your binary RPMs. Alternatively, you can just do "rpmbuild -bs" to build a src.rpm. See "man rpmbuild" (of course). Paul. -- Paul Howarth From Axel.Thimm at ATrpms.net Wed May 4 22:33:35 2005 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Thu, 5 May 2005 00:33:35 +0200 Subject: Stateless Debian Project In-Reply-To: <42771C1F.4040206@hclcomnet.co.in> References: <42771C1F.4040206@hclcomnet.co.in> Message-ID: <20050504223335.GA9651@neu.nirvana> On Tue, May 03, 2005 at 12:07:19PM +0530, gaurav wrote: > We have just initiated Stateless Debian Project and we are looking for > active volunteers/developers > Detail of project are given below Debian already has FAI (http://www.informatik.uni-koeln.de/fai/). Perhaps it would make more sense to start to converge the two similar concepts? -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From jspaleta at gmail.com Wed May 4 23:30:11 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Wed, 4 May 2005 19:30:11 -0400 Subject: rpm's zlib and deltarpms In-Reply-To: <4279404B.9030004@nc.rr.com> References: <20050504182644.GA3024@suse.de> <427918F9.4040706@nc.rr.com> <20050504191516.GA26676@suse.de> <604aa7910505041342578fd4dd@mail.gmail.com> <4279404B.9030004@nc.rr.com> Message-ID: <604aa79105050416305ca2387e@mail.gmail.com> On 5/4/05, Jeff Johnson wrote: > Make sure you include the RFE to distribute zlib.h and zconf.h with rpm too. You don't consider the patched zlib source in the srpm for rpm a modified source distribution of zlib? Seems to me that zlib.h and zconf.h as distributed in the rpm srpm falls inside the scenario laid on in the zlib faq for modified source distributions... making any such RFE unecessarily. No real point in arguing about this legal hair-splitting in the lists. I'm sure this minor license compliance issue with regard to how zlib is maintained as part of the rpm sourcecode base will cross paths with a redhat legal rep at some point if it hasn't already happened. The only question I have is, if the version string in the zlib.h inside the rpm codebase is changed does that have a technical impact on any codebase that is currently interacting with rpm? -jef From n3npq at nc.rr.com Wed May 4 23:50:12 2005 From: n3npq at nc.rr.com (Jeff Johnson) Date: Wed, 04 May 2005 19:50:12 -0400 Subject: rpm's zlib and deltarpms In-Reply-To: <604aa79105050416305ca2387e@mail.gmail.com> References: <20050504182644.GA3024@suse.de> <427918F9.4040706@nc.rr.com> <20050504191516.GA26676@suse.de> <604aa7910505041342578fd4dd@mail.gmail.com> <4279404B.9030004@nc.rr.com> <604aa79105050416305ca2387e@mail.gmail.com> Message-ID: <42795FB4.2090702@nc.rr.com> Jeff Spaleta wrote: >On 5/4/05, Jeff Johnson wrote: > > >>Make sure you include the RFE to distribute zlib.h and zconf.h with rpm too. >> >> > >You don't consider the patched zlib source in the srpm for rpm a >modified source distribution of zlib? Seems to me that zlib.h and >zconf.h as distributed in the rpm srpm falls inside the scenario laid >on in the zlib faq for modified source distributions... making any >such RFE unecessarily. > >No real point in arguing about this legal hair-splitting in the lists. >I'm sure this minor license compliance issue with regard to how zlib >is maintained as part of the rpm sourcecode base will cross paths with >a redhat legal rep at some point if it hasn't already happened. The >only question I have is, if the version string in the zlib.h inside >the rpm codebase is changed does that have a technical impact on any >codebase that is currently interacting with rpm? > > > No interaction is possible unless zlib.h and zconf.h are distributed. Hence the RFE ... 73 de Jeff From buildsys at redhat.com Thu May 5 12:13:34 2005 From: buildsys at redhat.com (Build System) Date: Thu, 5 May 2005 08:13:34 -0400 Subject: rawhide report: 20050505 changes Message-ID: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.4-20050503.144108.FC4.5 ----------------------------------------- NetworkManager-0.4-10.cvs20050404 --------------------------------- * Wed May 04 2005 Dan Williams - 0.4-10.cvs20050404 - Fix leak of a socket in DHCP code * Wed May 04 2005 Dan Williams - 0.4-9.cvs20050404 - Fix some memory leaks (Tom Parker) - Join to threads rather than spinning for their completion (Tom Parker) - Fix misuse of a g_assert() (Colin Walters) - Fix return checking of an ioctl() (Bill Moss) - Better detection and matching of hidden access points (Bill Moss) - Don't use varargs, and therefore don't crash on PPC (Peter Jones) alsa-lib-1.0.9rc2-4 ------------------- * Wed May 04 2005 Than Ngo 1.0.9rc2-4 - apply patch to fix artsd daemon crash #156592 cman-kernel-2.6.11.3-20050425.154843.FC4.8 ------------------------------------------ curl-7.13.1-3 ------------- * Tue May 03 2005 Ivana Varekova 7.13.1-3 - fix bug 150768 - curl-7.12.3-2 breaks basic authentication used Daniel Stenberg patch (patch2) device-mapper-multipath-0.4.4-2.1 --------------------------------- * Wed May 04 2005 Alasdair Kergon - 0.4.4-2.1 - By default, disable the multipathd service. dlm-kernel-2.6.11.3-20050425.154843.FC4.10 ------------------------------------------ dump-0.4b40-1 ------------- * Tue May 03 2005 Jindrich Novy 0.4b40-1 - Updated to dump 0.4b40 - Dropped .ea and .asize patches (applied upstream) firefox-0:1.0.3-4 ----------------- * Tue May 03 2005 Christopher Aillon 0:1.0.3-4 - Rebuild * Tue May 03 2005 Christopher Aillon - Patch from Marcel Mol supporting launching with filenames containing whitespace. * Tue May 03 2005 Christopher Aillon 0:1.0.3-3 - Firefox script fixes to support multilib installs. - Add upstream patch to fix bidi justification of pango - Add patch to fix launching of helper applications glibc-2.3.5-6 ------------- * Wed May 04 2005 Jakub Jelinek 2.3.5-6 - update from CVS - fix cancellation on i?86 - add call frame information to i?86 assembly * Tue May 03 2005 Jakub Jelinek 2.3.5-5 - update from CVS - add some more UTF-8 locales (#156115) - clean up /lib64/tls instead of /lib/tls on x86-64, s390x and ppc64 in glibc_post_upgrade (#156656) - fix posix_fallocate{,64} (#156289) gnbd-kernel-2.6.11.2-20050420.133124.FC4.13 ------------------------------------------- gnome-desktop-2.10.0-2 ---------------------- * Wed May 04 2005 Ray Strode - 2.10.0-2 - run gettext initialization routines on startup (bug 155659). gnome-panel-2.10.1-9 -------------------- * Wed May 04 2005 Mark McLoughlin 2.10.1-9 - Fix crash with "Recent Documents" menu (bug #156633) gnome-terminal-2.10.0-2 ----------------------- * Wed May 04 2005 Ray Strode 2.10.0-2 - Fix ne translation (bug 152240). gzip-1.3.5-6 ------------ * Mon May 02 2005 Ivana Varekova 1.3.5-6 - rebuilt * Fri Apr 29 2005 Ivana Varekova 1.3.5-5 - fix bug 156269 - CAN-2005-1228 directory traversal bug (using the patch from Ulf Harnhammar) httpd-2.0.54-7 -------------- * Wed May 04 2005 Joe Orton 2.0.54-7 - mod_userdir: fix memory allocation issue (upstream #34588) - mod_ldap: fix memory corruption issue (Brad Nicholes, upstream #34618) jakarta-commons-beanutils-0:1.7.0-1jpp_2fc ------------------------------------------ * Wed May 04 2005 Gary Benson - 0:1.7.0-1jpp_2fc - BC-compile. jakarta-commons-collections-0:3.1-1jpp_2fc ------------------------------------------ * Wed May 04 2005 Gary Benson - 0:3.1-1jpp_2fc - BC-compile. - Do not fetch stuff from sun.com during javadoc generation. jakarta-commons-digester-0:1.6-2jpp_2fc --------------------------------------- * Wed May 04 2005 Gary Benson - 0:1.6-2jpp_2fc - BC-compile the main jarfile. jakarta-commons-logging-0:1.0.4-2jpp_2fc ---------------------------------------- * Wed May 04 2005 Gary Benson - 0:1.0.4-2jpp_2fc - BC-compile. * Tue Jan 11 2005 Gary Benson - 0:1.0.4-2jpp_1fc - Sync with RHAPS. * Thu Nov 04 2004 Gary Benson - 0:1.0.4-1jpp_2fc - Build into Fedora. jakarta-commons-modeler-0:1.1-3jpp_2fc -------------------------------------- * Wed May 04 2005 Gary Benson - 0:1.1-3jpp_2fc - BC-compile. * Mon Mar 07 2005 Gary Benson - 0:1.1-3jpp_1fc - Build into Fedora. * Fri Oct 22 2004 Fernando Nasser - 0:1.1-3jpp_1rh - Merge with upstream version kdelibs-6:3.4.0-6 ----------------- * Wed May 04 2005 Than Ngo 6:3.4.0-6 - kimgio input validation vulnerabilities, CAN-2005-1046 kdewebdev-6:3.4.0-3 ------------------- * Wed May 04 2005 Than Ngo 6:3.4.0-3 - apply patch to fix CAN-2005-0754, Kommander untrusted code execution, thanks to KDE security team kernel-2.6.11-1.1286_FC4 ------------------------ * Wed May 04 2005 Jeremy Katz - enable radeonfb and agp on ppc64 to fix X on the G5 latex2html-2002.2.1-3 --------------------- * Wed May 04 2005 Jindrich Novy 2002.2.1-3 - add latex2html, texexpand, pstoimg man pages (#60308) libselinux-1.23.10-2 -------------------- * Tue Apr 26 2005 Dan Walsh 1.23.10-2 - Add info to man page * Tue Apr 26 2005 Dan Walsh 1.23.10-1 - Update from NSA * Merged set_selinuxmnt patch from Bill Nottingham (Red Hat). * Rewrote get_ordered_context_list and helpers, including changing logic to allow variable MLS fields. mkinitrd-4.2.11-1 ----------------- * Wed May 04 2005 Peter Jones - 4.2.11-1 - don't copy "rw" option into the initrd; use "ro" instead if it's the only option. - Don't print "unmounting old ..." messages in nash if we're set to quiet mod_jk-0:1.2.6-3jpp_4fc ----------------------- * Wed May 04 2005 Joe Orton 0:1.2.6-3jpp_4fc - add dependency on httpd-mmn - package NOTICE in docdir per ASL2.0 requirements * Tue May 03 2005 Bill Nottingham - 0:1.2.6-3jpp_3fc - temporarily disable -tools subpackage (#156757) openoffice.org-1:1.9.100-1 -------------------------- * Wed May 04 2005 Caolan McNamara - 1:1.9.100-1 - bump to next version - drop finally integrated openoffice.org-1.9.75.ooo41904.singleton.patch * Tue May 03 2005 Caolan McNamara - 1:1.9.99-2 - add openoffice.org-1.9.99.gcc19870.gcjaccessproblem.filter.patch for gcj bug workaround - add openoffice.org-1.9.97.ooo48610.searchalltemplates.wizards.patch - help documentation for cs and et has landed - add openoffice.org-1.9.99.gcc19870.gcjaccessproblem.wizards.patch -> wizards work! * Tue May 03 2005 Caolan McNamara - 1:1.9.99-1 - add openoffice.org-1.9.97.ooo48600.rtfparseerror.svx.patch - drop openoffice.org-1.9.95.gcc21233.noquotesonjavaver.patch as of gcc 4.0.0-2 patch-2.5.4-24 -------------- * Wed May 04 2005 Tim Waugh 2.5.4-24 - Reverted last change (bug #154283, bug #156762). php-5.0.4-9 ----------- * Tue May 03 2005 Joe Orton 5.0.4-9 - build simplexml_import_dom even with shared dom (#156434) - prevent truncation of copied files to ~2Mb (#155916) - install /usr/bin/php from CLI build alongside CGI - enable sysvmsg extension (#142988) policycoreutils-1.23.7-1 ------------------------ * Fri Apr 29 2005 Dan Walsh 1.23.7-1 - Change -f flag in fixfiles to remove stuff from /tmp - Change -F flag to pass -F flag to restorecon/fixfiles. (IE Force relabel). * Thu Apr 14 2005 Dan Walsh 1.23.6-1 - Update to match NSA * Fixed signed/unsigned pointer bug in load_policy. * Reverted context validation patch for genhomedircon. * Wed Apr 13 2005 Dan Walsh 1.23.5-1 - Update to match NSA * Reverted load_policy is_selinux_enabled patch from Dan Walsh. Otherwise, an initial policy load cannot be performed using load_policy, e.g. for anaconda. redhat-artwork-0.122-6 ---------------------- * Tue May 03 2005 John (J5) Palmieri 0.122-6 - Add a symlink for openoffice icons so we have matching bluecurve icons instead of the ugly upstream ones rpm-4.4.1-18.1 -------------- * Wed May 04 2005 Paul Nasrat - 4.4.1-18.1 - Fix typo - Fix typo * Wed May 04 2005 Paul Nasrat - 4.4.1-18 - Add missing fsm.c from matchpathcon patches tomcat5-0:5.0.30-5jpp_2fc ------------------------- * Wed May 04 2005 Gary Benson 0:5.0.30-5jpp_2fc - BC-compile. xml-commons-0:1.0-0.b2.6jpp_11fc -------------------------------- * Tue May 03 2005 Gary Benson - 0:1.0-0.b2.6jpp_11fc - BC-compile the API jar. From dnjinc at wowway.com Thu May 5 12:40:29 2005 From: dnjinc at wowway.com (Demond James) Date: Thu, 05 May 2005 08:40:29 -0400 Subject: generate src.rpm In-Reply-To: <1115245802.3609.2.camel@ngeza> References: <1115245802.3609.2.camel@ngeza> Message-ID: <427A143D.7040706@wowway.com> Rakotomandimby (R12y) Mihamina wrote: >Hello, > >I saw a lot of documentation on how to rebuild rpm, building rpm from an >src.rpm, but: > >How to generate the src.rpm? > >Because. I used the src.rpm of someone to generate my rpms, and I would >like people to be abble to use my src.rpms too. > >But I dont know how to generate one, from the sources, the patches and >the specfile. > >Would you help me? just indicate a link if you have one. > > > I think you're a good canidate for http://www.gurulabs.com/goodies/guru+guides.php From gauravp at hclcomnet.co.in Thu May 5 12:53:56 2005 From: gauravp at hclcomnet.co.in (gaurav) Date: Thu, 05 May 2005 18:23:56 +0530 Subject: Stateless Debian Project In-Reply-To: <20050504223335.GA9651@neu.nirvana> References: <42771C1F.4040206@hclcomnet.co.in> <20050504223335.GA9651@neu.nirvana> Message-ID: <427A1764.801@hclcomnet.co.in> Axel Thimm wrote: >On Tue, May 03, 2005 at 12:07:19PM +0530, gaurav wrote: > > >>We have just initiated Stateless Debian Project and we are looking for >>active volunteers/developers >>Detail of project are given below >> >> > >Debian already has FAI (http://www.informatik.uni-koeln.de/fai/). > >Perhaps it would make more sense to start to converge the two similar >concepts? > > we plan to ....its mentioned in my initial mail :-) From bernd.bartmann at gmail.com Thu May 5 16:03:44 2005 From: bernd.bartmann at gmail.com (Bernd Bartmann) Date: Thu, 5 May 2005 18:03:44 +0200 Subject: Missing update announcements Message-ID: <6c18a4f05050509036a77b73e@mail.gmail.com> Again several updates/erratas for FC3 appeared on the download mirrors but were never announced. I've opened a bug for each missing announcement: perl-5.8.5-11.FC3, #156366 logwatch-5.2.2-1.FC3.1, #156367 devhelp-0.9.2-2.3.2, #156017 epiphany-1.4.4-4.3.2, #156016 mozilla-1.7.7-1.3.1, #156015 logwatch-5.2.2-1.FC3, #156014 firefox-1.0.3-1.3.1, #156013 postfix-2.1.5-5, #151588 nfs-utils-1.0.6-52, #149901 Can someone @redhat.com please take care that the announcements get posted? TIA, Bernd. From bernd.bartmann at gmail.com Thu May 5 16:58:42 2005 From: bernd.bartmann at gmail.com (Bernd Bartmann) Date: Thu, 5 May 2005 18:58:42 +0200 Subject: Missing update announcements In-Reply-To: <1115311053.20494.17.camel@finch.boston.redhat.com> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> Message-ID: <6c18a4f05050509584792d76b@mail.gmail.com> On 5/5/05, John Dennis wrote: > On Thu, 2005-05-05 at 18:03 +0200, Bernd Bartmann wrote: > > Again several updates/erratas for FC3 appeared on the download mirrors > > but were never announced. > > It's worthwhile to point out that the process for getting a new rpm > signed and pushed is entirely manual done via email requests. A > developer sends an email asking for the package to be pushed. It may > take between 1 hour and 2 days for the request to be honored. He cannot > generate the announcement until the sign/push occurs because the MD5 > sums won't be correct until the package is signed. So it is entirely > possible for a package to get signed while the developer is gone and he > may not be able to generate the announcement until he returns. Because > of the manual nature of the process and the inherent delays it is > entirely possible for several days to transpire between a package making > an appearance and the announcement. > > I'm not defending the process, but I trying to explain why it sometimes > seems like announcements are delayed. The process should be fixed, but > it is what it is :-( > > > I've opened a bug for each missing > > announcement: > > How long have you waited for the announcement? If its less than 2 days I > would give the developer a chance before opening a bug. I assume you > have some automated script that is detecting these missed announcements, > humans trying to juggle the process along with all the other things on > their plate can't compete with computers waiting to catch them. Every of the missing announcements is much longer than 2 days out. Here are the dates when the files appeared on the download mirror that I use: perl-5.8.5-11.FC3, released 2005-04-27, #156366 logwatch-5.2.2-1.FC3.1, released 2005-04-26, #156367 devhelp-0.9.2-2.3.2, released 2005-04-22, #156017 epiphany-1.4.4-4.3.2, released 2005-04-22, #156016 mozilla-1.7.7-1.3.1, released 2005-04-22, #156015 logwatch-5.2.2-1.FC3, released 2005-04-21, #156014 firefox-1.0.3-1.3.1, released 2005-04-19, #156013 postfix-2.1.5-5, released 2005-03-17, #151588 nfs-utils-1.0.6-52, released 2005-02-19, #149901 Up to now I have no script to monitor the missing announcement. I'm still hoping that this problem is addressed in a way so that it cannot happen again. I just take a look at my rsync mirror logs and compare them to the announcements that appeared on the mailing list. If no announcement comes out within two days after the rpms appear on the download mirror I open up a bug in bugzilla. So far most rpm packagers were helpful and sent out the announcement when they see the bugreport. But sometimes even this fails :-( Why is it not possible to automatically sent out the announcement after the rpms are signed and the MD5 sums are known? Is it possible to trigger an email to the packager after the package is signed? Something like a reminder "package is ready for distribution, please sent out an announcement". TIA, Bernd. From mrsam at courier-mta.com Thu May 5 17:18:11 2005 From: mrsam at courier-mta.com (Sam Varshavchik) Date: Thu, 05 May 2005 13:18:11 -0400 Subject: Missing update announcements References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> <6c18a4f05050509584792d76b@mail.gmail.com> Message-ID: Bernd Bartmann writes: > Why is it not possible to automatically sent out the announcement > after the rpms are signed and the MD5 sums are known? > > Is it possible to trigger an email to the packager after the package > is signed? Something like a reminder "package is ready for > distribution, please sent out an announcement". Why even care about MD5 sums? As long as the signature verifies OK that's all you need to know. The signature is really just another sum of the file's contents, that's encrypted by the private key. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mattdm at mattdm.org Thu May 5 17:54:26 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Thu, 5 May 2005 13:54:26 -0400 Subject: Missing update announcements In-Reply-To: <1115311053.20494.17.camel@finch.boston.redhat.com> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> Message-ID: <20050505175426.GA30155@jadzia.bu.edu> On Thu, May 05, 2005 at 12:37:33PM -0400, John Dennis wrote: > may not be able to generate the announcement until he returns. Because > of the manual nature of the process and the inherent delays it is > entirely possible for several days to transpire between a package making > an appearance and the announcement. > I'm not defending the process, but I trying to explain why it sometimes > seems like announcements are delayed. The process should be fixed, but > it is what it is :-( The manual nature of the current process causes some other problems too -- particularly, the format of the announcements often deviates from the norm (although people have gotten better about this). -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 77 degrees Fahrenheit. From matthew at nocturnal.org Thu May 5 18:31:33 2005 From: matthew at nocturnal.org (Matthew Lenz) Date: Thu, 05 May 2005 13:31:33 -0500 Subject: early-login status? Message-ID: <1115317893.16123.8.camel@mlenzdesktop> I've been trying to follow the early-login changes but don't seem much discussion on the lists about it, or in the bugzilla entry. Are the changes to syslog and xfs /etc/rc.d/init.d's still required or is it now simply a matter of s/rhgb/early-login/g on /etc/grub.conf and enabling the 3 early-login scripts in /etc/rc.d/init.d? Have any of you tried out recently released http://jw.dyndns.org/initng/ (InitNG)? Anyone see any major CONs to implementing it for FC5? Looks neater than heck from what I can see. From kyrre at solution-forge.net Thu May 5 18:36:21 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Thu, 05 May 2005 20:36:21 +0200 Subject: Two latest kernels and udev Message-ID: <1115318181.3343.14.camel@localhost.localdomain> I am not able to boot my computer using either of the two latest kernels (2.6.11-1.1286_FC4 and .1284). All that happens is during "Starting udev" i get the error "MAKEDEV mkdir: File exists". Yet udev gets the green "OK". It then hangs at "Initiating hardware...". Kernel .1282, .1276, and .1275 are all OK - they boot, no problem. (there is an error "mknod: failed to create /dev/"{"console", "zero", "null"}" 17" Anyone with similar experiences? This is a Dell Latitude c600/c500 laptop computer. Kyrre From kyrre at solution-forge.net Thu May 5 18:41:43 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Thu, 05 May 2005 20:41:43 +0200 Subject: early-login status? In-Reply-To: <1115317893.16123.8.camel@mlenzdesktop> References: <1115317893.16123.8.camel@mlenzdesktop> Message-ID: <1115318502.3343.17.camel@localhost.localdomain> tor, 05.05.2005 kl. 20.31 skrev Matthew Lenz: > I've been trying to follow the early-login changes but don't seem much > discussion on the lists about it, or in the bugzilla entry. Are the > changes to syslog and xfs /etc/rc.d/init.d's still required or is it now > simply a matter of s/rhgb/early-login/g on /etc/grub.conf and enabling > the 3 early-login scripts in /etc/rc.d/init.d? > > Have any of you tried out recently released http://jw.dyndns.org/initng/ > (InitNG)? Anyone see any major CONs to implementing it for FC5? Looks > neater than heck from what I can see. And there is also the Apple launchd, which is (as far as i know) F/OSS and *very* fast (15 secounds or something from OS start to desktop... And no, that is *not* suspend!) From ville.skytta at iki.fi Thu May 5 18:44:03 2005 From: ville.skytta at iki.fi (Ville =?ISO-8859-1?Q?Skytt=E4?=) Date: Thu, 05 May 2005 21:44:03 +0300 Subject: Two latest kernels and udev In-Reply-To: <1115318181.3343.14.camel@localhost.localdomain> References: <1115318181.3343.14.camel@localhost.localdomain> Message-ID: <1115318643.17931.23.camel@bobcat.mine.nu> On Thu, 2005-05-05 at 20:36 +0200, Kyrre Ness Sjobak wrote: > I am not able to boot my computer using either of the two latest kernels > (2.6.11-1.1286_FC4 and .1284). All that happens is during "Starting > udev" i get the error "MAKEDEV mkdir: File exists". Yet udev gets the > green "OK". It then hangs at "Initiating hardware...". > > Kernel .1282, .1276, and .1275 are all OK - they boot, no problem. > (there is an error "mknod: failed to create /dev/"{"console", "zero", > "null"}" 17" > > Anyone with similar experiences? Yep. > This is a Dell Latitude c600/c500 laptop computer. IBM ThinkPad A30p here. From notting at redhat.com Thu May 5 18:44:34 2005 From: notting at redhat.com (Bill Nottingham) Date: Thu, 5 May 2005 14:44:34 -0400 Subject: Two latest kernels and udev In-Reply-To: <1115318643.17931.23.camel@bobcat.mine.nu> References: <1115318181.3343.14.camel@localhost.localdomain> <1115318643.17931.23.camel@bobcat.mine.nu> Message-ID: <20050505184434.GA29499@nostromo.devel.redhat.com> Ville Skytt? (ville.skytta at iki.fi) said: > On Thu, 2005-05-05 at 20:36 +0200, Kyrre Ness Sjobak wrote: > > I am not able to boot my computer using either of the two latest kernels > > (2.6.11-1.1286_FC4 and .1284). All that happens is during "Starting > > udev" i get the error "MAKEDEV mkdir: File exists". Yet udev gets the > > green "OK". It then hangs at "Initiating hardware...". > > > > Kernel .1282, .1276, and .1275 are all OK - they boot, no problem. > > (there is an error "mknod: failed to create /dev/"{"console", "zero", > > "null"}" 17" > > > > Anyone with similar experiences? > > Yep. Index: start_udev =================================================================== RCS file: /cvs/dist/rpms/udev/devel/start_udev,v retrieving revision 1.35 retrieving revision 1.36 diff -u -r1.35 -r1.36 --- start_udev 2 Mar 2005 16:19:37 -0000 1.35 +++ start_udev 5 May 2005 16:24:22 -0000 1.36 @@ -200,7 +200,7 @@ echo -n "$STRING " # mount the tmpfs on ${udev_root%/}, if not already done -LANG=C fgrep -q "none ${udev_root%/} " /proc/mounts || { +LANG=C awk "\$2 == \"${udev_root%/}\" && \$3 == \"tmpfs\" { exit 1 }" /proc/mounts && { if LANG=C fgrep -q "none ${udev_root%/}/pts " /proc/mounts; then PTSDIR=$(mktemp -d) mount --move $udev_root/pts "$PTSDIR" Bill From pjones at redhat.com Thu May 5 19:54:33 2005 From: pjones at redhat.com (Peter Jones) Date: Thu, 05 May 2005 15:54:33 -0400 Subject: What is wrong in this spec file? In-Reply-To: <1115238512.23655.6.camel@fctmp> References: <1115238512.23655.6.camel@fctmp> Message-ID: <1115322873.30401.18.camel@localhost.localdomain> On Wed, 2005-05-04 at 22:28 +0200, Rakotomandimby (R12y) Mihamina wrote: [lots of stuff] So, lots of people have already told you to use a BuildRoot, and they're correct. Mathew Miller also said that you'd be better off not using the "root" user to build. He's absolutely correct, so I figure that the instructions for this don't hit the list often enough, so here's a simple guide to doing so: 1) log in as a non-root user. I like "pjones" for that, but suit yourself ;) 2) create a .rpmmacros file. In it, put something like: %_topdir /home/pjones/build %_tmp %{_topdir}/tmp %_tmpdir %{_tmp} %_tmppath %{_tmp} 3) make some directories: cd /home/pjones mkdir build cd build mkdir RPMS SPECS SOURCES BUILD SRPMS tmp 4) test that everything's right: rpm -Uvh foo-1-1.src.rpm cd ~/build/SPECS/ rpmbuild -ba foo.spec It should start doing the obvious build stuff. When it's done, it'll write packages out to ~/build/RPMS/$ARCH/foo-1.1.$ARCH.rpm, just like it used to do in /usr/src/redhat . -- Peter From hp at redhat.com Thu May 5 20:29:42 2005 From: hp at redhat.com (Havoc Pennington) Date: Thu, 05 May 2005 16:29:42 -0400 Subject: Stateless Debian Project In-Reply-To: <42771C1F.4040206@hclcomnet.co.in> References: <42771C1F.4040206@hclcomnet.co.in> Message-ID: <1115324982.23440.68.camel@localhost.localdomain> Hi, It looks like we do plan to continue the stateless project (and related efforts) - lining up the people and the plan right now. It's true that it won't make FC4 though. Havoc From hp at redhat.com Thu May 5 20:43:37 2005 From: hp at redhat.com (Havoc Pennington) Date: Thu, 05 May 2005 16:43:37 -0400 Subject: Stateless Debian Project In-Reply-To: <1115324982.23440.68.camel@localhost.localdomain> References: <42771C1F.4040206@hclcomnet.co.in> <1115324982.23440.68.camel@localhost.localdomain> Message-ID: <1115325817.23440.72.camel@localhost.localdomain> On Thu, 2005-05-05 at 16:29 -0400, Havoc Pennington wrote: > Hi, > > It looks like we do plan to continue the stateless project (and related > efforts) - lining up the people and the plan right now. It's true that > it won't make FC4 though. > BTW, part of the delay was collecting customer feedback and better understanding how this project relates to the real world, other architecture components, and what it should be like in general. Havoc From ivazquez at ivazquez.net Thu May 5 20:45:45 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Thu, 05 May 2005 16:45:45 -0400 Subject: Missing update announcements In-Reply-To: <1115311053.20494.17.camel@finch.boston.redhat.com> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> Message-ID: <1115325945.8962.8.camel@ignacio.ignacio.lan> On Thu, 2005-05-05 at 12:37 -0400, John Dennis wrote: > On Thu, 2005-05-05 at 18:03 +0200, Bernd Bartmann wrote: > > Again several updates/erratas for FC3 appeared on the download mirrors > > but were never announced. > > It's worthwhile to point out that the process for getting a new rpm > signed and pushed is entirely manual done via email requests. A > developer sends an email asking for the package to be pushed. It may > take between 1 hour and 2 days for the request to be honored. He cannot > generate the announcement until the sign/push occurs because the MD5 > sums won't be correct until the package is signed. So it is entirely > possible for a package to get signed while the developer is gone and he > may not be able to generate the announcement until he returns. Because > of the manual nature of the process and the inherent delays it is > entirely possible for several days to transpire between a package making > an appearance and the announcement. Why not have a template that the packager fills out partially with %summary, %description, etc., then have a place where the signer puts the md5sums. That way the packager doesn't have to worry about when the packages are signed, and the announcement can go out as soon as the updates are ready. -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From ivazquez at ivazquez.net Thu May 5 20:51:15 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Thu, 05 May 2005 16:51:15 -0400 Subject: What is wrong in this spec file? In-Reply-To: <1115322873.30401.18.camel@localhost.localdomain> References: <1115238512.23655.6.camel@fctmp> <1115322873.30401.18.camel@localhost.localdomain> Message-ID: <1115326275.8962.13.camel@ignacio.ignacio.lan> On Thu, 2005-05-05 at 15:54 -0400, Peter Jones wrote: > On Wed, 2005-05-04 at 22:28 +0200, Rakotomandimby (R12y) Mihamina wrote: > [lots of stuff] > > So, lots of people have already told you to use a BuildRoot, and they're > correct. Mathew Miller also said that you'd be better off not using the > "root" user to build. He's absolutely correct, so I figure that the > instructions for this don't hit the list often enough, so here's a > simple guide to doing so: > > 1) ... > > 2) ... > > 3) ... Or just install the fedora-rpmdevtools package from Extras and use fedora-buildrpmtree. -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From kyrre at solution-forge.net Thu May 5 21:47:05 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Thu, 05 May 2005 23:47:05 +0200 Subject: Stateless Debian Project In-Reply-To: <1115325817.23440.72.camel@localhost.localdomain> References: <42771C1F.4040206@hclcomnet.co.in> <1115324982.23440.68.camel@localhost.localdomain> <1115325817.23440.72.camel@localhost.localdomain> Message-ID: <1115329624.4893.1.camel@localhost.localdomain> tor, 05.05.2005 kl. 22.43 skrev Havoc Pennington: > On Thu, 2005-05-05 at 16:29 -0400, Havoc Pennington wrote: > > Hi, > > > > It looks like we do plan to continue the stateless project (and related > > efforts) - lining up the people and the plan right now. It's true that > > it won't make FC4 though. > > > > BTW, part of the delay was collecting customer feedback and better > understanding how this project relates to the real world, other > architecture components, and what it should be like in general. > > Havoc > Just wondering - is this anyhow connected with Netscape Directory Server? "Anyone" who has configured an openLDAP-server knows that it takes a limb... Kyrre From gmartyn at verizon.net Fri May 6 01:10:44 2005 From: gmartyn at verizon.net (Greg Martyn) Date: Thu, 05 May 2005 21:10:44 -0400 Subject: initng Message-ID: <200505052110.44562.gmartyn@verizon.net> Hi all, I was just wondering if this dovetails with some of the early-login stuff that has been discussed on this list. It was written with gentoo in mind though.. "Jimmy Wennlund wrote initng, a replacement for the Sys-V style "init" application. It allows for better service dependency checking and will start services in a highly parallel fashion, dramatically speeding up the Linux boot process." http://www.osnews.com/story.php?news_id=10513 -Greg From gmartyn at verizon.net Fri May 6 01:20:24 2005 From: gmartyn at verizon.net (Greg Martyn) Date: Thu, 05 May 2005 21:20:24 -0400 Subject: initng In-Reply-To: <200505052110.44562.gmartyn@verizon.net> References: <200505052110.44562.gmartyn@verizon.net> Message-ID: <200505052120.24245.gmartyn@verizon.net> D'oh! someone else beat me to it.. From rodd at clarkson.id.au Fri May 6 01:48:15 2005 From: rodd at clarkson.id.au (Rodd Clarkson) Date: Fri, 06 May 2005 11:48:15 +1000 Subject: Two latest kernels and udev In-Reply-To: <1115318643.17931.23.camel@bobcat.mine.nu> References: <1115318181.3343.14.camel@localhost.localdomain> <1115318643.17931.23.camel@bobcat.mine.nu> Message-ID: <1115344095.3783.11.camel@jellyfish.redfishdemo.com> On Thu, 2005-05-05 at 21:44 +0300, Ville Skytt? wrote: > > Anyone with similar experiences? > > Yep. > > > This is a Dell Latitude c600/c500 laptop computer. > > IBM ThinkPad A30p here. There's been a couple of other threads about this. Do both these laptops have PIII chipsets? Rodd From naoki at valuecommerce.com Fri May 6 03:23:09 2005 From: naoki at valuecommerce.com (Naoki) Date: Fri, 06 May 2005 12:23:09 +0900 Subject: rawhide report: 20050505 changes In-Reply-To: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> References: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> Message-ID: <1115349789.16090.8.camel@dragon.sys.intra> Do we like uppercase letters in package names? I'm not trolling but I would like to know if there is a documented reason for this. It does cause occasional minor annoyances and goes against the de facto standard. Of course it does ensure your packages are at the top of the list ;) > Updated Packages: > > GFS-kernel-2.6.11.4-20050503.144108.FC4.5 > NetworkManager-0.4-10.cvs20050404 > HelixPlayer-1:1.0.4-4 From dmalcolm at redhat.com Fri May 6 03:35:59 2005 From: dmalcolm at redhat.com (David Malcolm) Date: Thu, 05 May 2005 23:35:59 -0400 Subject: rawhide report: 20050505 changes In-Reply-To: <1115349789.16090.8.camel@dragon.sys.intra> References: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> <1115349789.16090.8.camel@dragon.sys.intra> Message-ID: <1115350560.18184.70.camel@cassandra.boston.redhat.com> On Fri, 2005-05-06 at 12:23 +0900, Naoki wrote: > Do we like uppercase letters in package names? > > I'm not trolling but I would like to know if there is a documented > reason for this. It does cause occasional minor annoyances and goes > against the de facto standard. Of course it does ensure your packages > are at the top of the list ;) Numerals go even higher. 4Suite wins hands-down :-) From rmoore at engnuix.com Fri May 6 05:12:05 2005 From: rmoore at engnuix.com (Reuben Moore) Date: Thu, 05 May 2005 22:12:05 -0700 Subject: Stateless Debian Project In-Reply-To: <1115325817.23440.72.camel@localhost.localdomain> References: <42771C1F.4040206@hclcomnet.co.in> <1115324982.23440.68.camel@localhost.localdomain> <1115325817.23440.72.camel@localhost.localdomain> Message-ID: <427AFCA5.6070907@engnuix.com> Hi all, I'm glad to hear that the "Fedora" Stateless Linux project will continue. While I have not had the chance to test it out yet, I am very interested in it's potential and will most certainly be setting up the prototype system to start in with any help I can provide in its testing and or development. (-especially since looking at this lists threads regarding Fedora's involvement gave me the sense of Fedora/Red Hat abandonment.) Not to overdo my concern for this project, but the ability to "throw a computer out the window" and then be able to "recreate its software, configuration, and user data bit-for-bit identically on a new piece of hardware" among other aspects of the project benefits like thin clients has me stoked. I joined this mailing list exclusively because of the Stateless Linux project. I can see all the Linux servers at the ISP I administrate as (cached client) servers today. Not to mention shared root virtual server environment possibilities. I've setup LTSP to wireless laptops for a local golf course lounge room that has stateless ability so someone screwing with the desktop returned clean environment for the next persons use. I can see the stateless solution even better with a more powerful laptop for local cache snapshot improving on network bandwidth, speed increase, and other bennies of using the local machine for ALL the processing -and still be stateless or even thin. I'd like to see Fedora/Red Hat take the cake on these types of solutions and provide, as mentioned for "goals" in the StatelessLinux.pdf, having "out of the box" -designed in as part of an OS, go head to head with any up and coming competition. See: http://hardware.slashdot.org/article.pl?sid=05/04/19/0111218&from=rss . Regardless of the ups and downs of thin client fashion I do not believe these kind of solutions are going away. Conclusively, this email is forwarding to the local LUG in my area to try and bolster some incentive(revival) of our own (as of lately) Linux lackluster. Also, given the LUG community in my area has a lot of Debian users, some of relative links mentioned in these mailings regarding Stateless Linux accordingly. Fedora-devel archives for this mailing list reside here: https://www.redhat.com/archives/fedora-devel-list/ Debian links associated in this project here: http://sourceforge.net/projects/statelessdebian/ http://www.informatik.uni-koeln.de/fai/ Starting point for Fedora Stateless Linux project information here: http://fedora.redhat.com/projects/stateless/ Thanks for listening, -Reuben PS, lets hear some input on this RPLUG, I'd be happy to do a meeting / demo of such much like I did with LTSP. Havoc Pennington wrote: >On Thu, 2005-05-05 at 16:29 -0400, Havoc Pennington wrote: > > >>Hi, >> >>It looks like we do plan to continue the stateless project (and related >>efforts) - lining up the people and the plan right now. It's true that >>it won't make FC4 though. >> >> >> > >BTW, part of the delay was collecting customer feedback and better >understanding how this project relates to the real world, other >architecture components, and what it should be like in general. > >Havoc > > > > From mharris at www.linux.org.uk Fri May 6 05:50:50 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Fri, 06 May 2005 01:50:50 -0400 Subject: Missing update announcements In-Reply-To: <6c18a4f05050509584792d76b@mail.gmail.com> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> <6c18a4f05050509584792d76b@mail.gmail.com> Message-ID: <427B05BA.7060007@www.linux.org.uk> Bernd Bartmann wrote: > Every of the missing announcements is much longer than 2 days out. > Here are the dates when the files appeared on the download mirror that > I use: > > perl-5.8.5-11.FC3, released 2005-04-27, #156366 > logwatch-5.2.2-1.FC3.1, released 2005-04-26, #156367 > devhelp-0.9.2-2.3.2, released 2005-04-22, #156017 > epiphany-1.4.4-4.3.2, released 2005-04-22, #156016 > mozilla-1.7.7-1.3.1, released 2005-04-22, #156015 > logwatch-5.2.2-1.FC3, released 2005-04-21, #156014 > firefox-1.0.3-1.3.1, released 2005-04-19, #156013 > postfix-2.1.5-5, released 2005-03-17, #151588 > nfs-utils-1.0.6-52, released 2005-02-19, #149901 > > Up to now I have no script to monitor the missing announcement. I'm > still hoping that this problem is addressed in a way so that it cannot > happen again. I just take a look at my rsync mirror logs and compare > them to the announcements that appeared on the mailing list. If no > announcement comes out within two days after the rpms appear on the > download mirror I open up a bug in bugzilla. So far most rpm packagers > were helpful and sent out the announcement when they see the > bugreport. But sometimes even this fails :-( > Why is it not possible to automatically sent out the announcement > after the rpms are signed and the MD5 sums are known? > > Is it possible to trigger an email to the packager after the package > is signed? Something like a reminder "package is ready for > distribution, please sent out an announcement". Taking several steps back to look at the forest instead of being unable to see it for all the trees ... The more reliance a given "system" has on human beings to perform a set of steps, the more room said system has for human error. The current fedora-update mechanism is currently entirely human driven with almost zero automation, unlike the errata mechanisms used for Red Hat Linux in the past, and Red Hat Enterprise Linux currently. What the Fedora Project needs right now to "solve" this and other problems related to errata, is an automated errata tool system that handles as much of the process as can be automated, leaving the engineer to do developmental engineering and maintenance instead of doing manual human release engineering. There are different potential solutions, but the first thing that comes to mind, is a web based CGI solution, which allows an engineer to open an erratum ticket, file some initial basic details summarizing the problems being addressed, the nature of the issues (security fix, bug fix, enhancement), and then be able to follow each step of the errata release process one at a time through to final release of packages. The process should handle the creation of the advisory text, and references to CVE names and other URLs related to any security problems or bugs being fixed. There should be a field to indicate which Red Hat bug IDs are being fixed in the update. There should also be a way for the engineer to point the errata CGI to the generated rpms, and to request that they be signed by those who have RPM GPG signing priveledges. Once the rpms are signed, the next "state" in the process would be errata advisory text approval - how this is handled could be done different ways. Once the text is approved, then and only then would the system permit the files to be pushed out to mirrors, and once the files are pushed, the system would automatically send out the advisory text. This theoretical system should also have a mechanism in place to be able to handle embargoed security items, so that the errata process can be started early, and a date entered into the tool which prevents public release of the packages until the embargo date is reached. This automates prevention of accidental early disclosure of security issues rather than relying on humans to not make mistakes. Call me crazy, I don't know where I managed to get all these ideas from, but it all just came together in my mind somehow.[1] Now that there is at least one theoretical solution to the problem, if this solution were to be considered, we need the following: 1) Someone to design the detailed solution 2) Someone to implement the design and test it 3) Put the new system in place and start using it. In #1 and #2 above, "someone" could be either a Red Hat employee voluntarily taking on the task on their personal time, or it could be a Red Hat employee taking on the task under work time, or it could be a person in the community taking on the task. My personal thought, is that it wont happen by a Red Hat employee under company time, unless it is designated as a high priority task that needs to be addressed right away, which for better or worse, I don't think is the case right now. It could happen by a Red Hat employee on their own time, if someone is enthusiastic about it enough and has the personal time to spare, however I think if anyone had the time and was enthusiastic about it enough, that it would have been done by now. It's much easier to just keep using the existing manual system than to spend a few weeks manpower in one's personal time to do it. Doesn't really sound like a "fun" project to me that we're likely to see one scratch one's personal itch per se. My thought is that the only way we'll ever see a new fedora errata infrastructure come into existance, is if Red Hat decides it is high priority enough to allocate developer resources to solving the problem internally, or alternatively if someone in the community got highly motivated to design something and hack on it along with some input/advice from Red Hatters. Now that I've expressed my thoughts out loud, before anyone steps up to say "Ok Mike, great idea, now send us your patches and code to do this." (which tends to be a frequent common response when one suggests ideas about a solution to a problem), I'll be the first to admit I'm not personally interested in spending time coding the solution to this problem on or off company time. I suspect everyone feels the same or we'd have something by now perhaps. Nonetheless, I do put forth my opinions above on how this could be solved, in hopes that someone who does have a direct personal interest and the time to devote to the task, might find the motivation to take up the project and perhaps we'll end up with something to try out in the future. Until then, IMHO, there's a high likelyhood that Fedora updates will more or less continue to be inconsistently released in a fairly ad hoc manner. Remove humans from the process as much as possible, making computers do the work, and the problem mostly solves itself. TTYL From mharris at www.linux.org.uk Fri May 6 06:03:45 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Fri, 06 May 2005 02:03:45 -0400 Subject: rawhide report: 20050505 changes In-Reply-To: <1115349789.16090.8.camel@dragon.sys.intra> References: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> <1115349789.16090.8.camel@dragon.sys.intra> Message-ID: <427B08C1.2060608@www.linux.org.uk> Naoki wrote: > Do we like uppercase letters in package names? Depends on how you define who "we" are. I personally hate package names with uppercase in them, wether they're all uppercase, mixed case, or CamelTextNames. When we switched to X.Org X11, I managed to cleverly assimilate the mixed case by naming the packaging xorg-x11, instead of the hideous looking: X.Org-X11 or perhaps something worse. ;o) > I'm not trolling but I would like to know if there is a documented > reason for this. It does cause occasional minor annoyances and goes There is no standard whatsoever. It's more or less up to whoever is creating an rpm package to decide how to name it. In general, most rpm packagers tend to use the upstream tarball name when possible unless it strays greatly from common practice. Others may choose to name it something that looks aesthetically pleasing to them, or which matches the end-user visible product naming. For example, I've seen "Real Player" rpm packages named: realplayer-x.y-z.i386.rpm and I've seen: RealPlayer-x.y-z.i386.rpm In my personal opinion, and feel free to differ (I'm sure there are many who do, or everyone would be doing it the same way already), I find all lowercase packagenames easier to handle from an administration viewpoint, and for querying the rpm database for things, etc. It ultimately is up to whoever is the one making the packages though, and wether they feel strongly about it one way or the other, or can be influenced to do it one way or another. > against the de facto standard. Of course it does ensure your packages > are at the top of the list ;) Not really. It depends on what LC_COLLATE is among other things. Some locales seem to sort all capitalized names first, then all lowercase like ABCabc, while other locales seem to sort AaBbCc or aAbBcC. Thankfully we haven't seen rpm packages named things like: 0000-foo-1.0-1.rpm or __-==[R34lPl4y3r]==-___-10.1-3.i386.rpm I'm sure that horrid day will eventually arrive though. ;o) From gauravp at hclcomnet.co.in Fri May 6 07:05:52 2005 From: gauravp at hclcomnet.co.in (gaurav) Date: Fri, 06 May 2005 12:35:52 +0530 Subject: Stateless Debian Project In-Reply-To: <1115329624.4893.1.camel@localhost.localdomain> References: <42771C1F.4040206@hclcomnet.co.in><1115324982.23440.68.camel@loc alhost.localdomain><1115325817.23440.72.camel@localhost.localdomain> <1115329624.4893.1.camel@localhost.localdomain> Message-ID: <427B1750.8000100@hclcomnet.co.in> Kyrre Ness Sjobak wrote: >tor, 05.05.2005 kl. 22.43 skrev Havoc Pennington: > > >>On Thu, 2005-05-05 at 16:29 -0400, Havoc Pennington wrote: >> >> >>>Hi, >>> >>>It looks like we do plan to continue the stateless project (and related >>>efforts) - lining up the people and the plan right now. It's true that >>>it won't make FC4 though. >>> >>> cool....we can abstract some Distro specific stuff (rpm, deb,fyi or anaconda) from generic stuff (lpad, caching etc) ...to make it truly Linux wide effort ....some enhancement or ideas can flow easily for cross pollination ...maybe two projects (Stateless Deb and Stateless RH) can collaborate >>> >>> >>BTW, part of the delay was collecting customer feedback and better >>understanding how this project relates to the real world, other >>architecture components, and what it should be like in general. >> >>Havoc >> >> >> > >Just wondering - is this anyhow connected with Netscape Directory >Server? > >"Anyone" who has configured an openLDAP-server knows that it takes a >limb... > >Kyrre > > > From kyrre at solution-forge.net Fri May 6 08:41:53 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Fri, 06 May 2005 10:41:53 +0200 Subject: Two latest kernels and udev In-Reply-To: <1115344095.3783.11.camel@jellyfish.redfishdemo.com> References: <1115318181.3343.14.camel@localhost.localdomain> <1115318643.17931.23.camel@bobcat.mine.nu> <1115344095.3783.11.camel@jellyfish.redfishdemo.com> Message-ID: <1115368912.3343.0.camel@localhost.localdomain> fre, 06.05.2005 kl. 03.48 skrev Rodd Clarkson: > On Thu, 2005-05-05 at 21:44 +0300, Ville Skytt? wrote: > > > > Anyone with similar experiences? > > > > Yep. > > > > > This is a Dell Latitude c600/c500 laptop computer. > > > > IBM ThinkPad A30p here. > > There's been a couple of other threads about this. Do both these > laptops have PIII chipsets? > > > Rodd > > At least mine has (Dell Latitude c600/c500) From kyrre at solution-forge.net Fri May 6 09:43:02 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Fri, 06 May 2005 11:43:02 +0200 Subject: Scriptlet errors while installing eclipse Message-ID: <1115372581.3343.17.camel@localhost.localdomain> I did a "yum install eclipse", and rpm showed the errors in the attached text file. Thougt someone on this list probably are interested :) Kyrre Ness Sj?b?k -------------- next part -------------- Running Transaction Installing: jakarta-commons-logging ####################### [ 1/48] /var/tmp/rpm-tmp.82500: line 1: rebuild-gcj-db: command not found error: %post(jakarta-commons-logging-1.0.4-2jpp_2fc.i386) scriptlet failed, exit status 127 Installing: jakarta-commons-collections ####################### [ 2/48] /var/tmp/rpm-tmp.48817: line 1: rebuild-gcj-db: command not found error: %post(jakarta-commons-collections-3.1-1jpp_2fc.i386) scriptlet failed, exit status 127 Installing: jakarta-commons-beanutils ####################### [ 3/48] /var/tmp/rpm-tmp.13492: line 1: rebuild-gcj-db: command not found error: %post(jakarta-commons-beanutils-1.7.0-1jpp_2fc.i386) scriptlet failed, exit status 127 Installing: regexp ####################### [ 4/48] Installing: junit ####################### [ 5/48] Installing: bcel ####################### [ 6/48] Installing: jakarta-commons-digester ####################### [ 7/48] /var/tmp/rpm-tmp.30561: line 1: rebuild-gcj-db: command not found error: %post(jakarta-commons-digester-1.6-2jpp_2fc.i386) scriptlet failed, exit status 127 Installing: xml-commons ####################### [ 8/48] Installing: xml-commons-apis ####################### [ 9/48] /var/tmp/rpm-tmp.55682: line 1: rebuild-gcj-db: command not found error: %post(xml-commons-apis-1.0-0.b2.6jpp_11fc.i386) scriptlet failed, exit status 127 Installing: classpathx-jaf ####################### [10/48] Installing: jdepend ####################### [11/48] Installing: oro ####################### [12/48] Installing: gjdoc ####################### [13/48] Installing: jzlib ####################### [14/48] Installing: antlr ####################### [15/48] Installing: libgcj-devel ####################### [16/48] Installing: gcc-java ####################### [17/48] Installing: gnu-crypto-jce-jdk1.4 ####################### [18/48] Installing: gnu-crypto-sasl-jdk1.4 ####################### [19/48] Installing: gnu-crypto ####################### [20/48] Installing: classpathx-mail ####################### [21/48] Installing: java-1.4.2-gcj-compat ####################### [22/48] Installing: jsch ####################### [23/48] Installing: xalan-j2 ####################### [24/48] Installing: log4j ####################### [25/48] Installing: xml-commons-resolver ####################### [26/48] Installing: mx4j ####################### [27/48] Installing: jakarta-commons-modeler ####################### [28/48] Installing: eclipse-ecj ####################### [29/48] Installing: java-1.4.2-gcj-compat-devel ####################### [30/48] Installing: ant ####################### [31/48] Installing: ant-trax ####################### [32/48] Installing: ant-apache-bcel ####################### [33/48] Installing: ant-antlr ####################### [34/48] Installing: ant-swing ####################### [35/48] Installing: ant-junit ####################### [36/48] Installing: ant-jsch ####################### [37/48] Installing: ant-jdepend ####################### [38/48] Installing: ant-nodeps ####################### [39/48] Installing: ant-commons-logging ####################### [40/48] Installing: ant-jmf ####################### [41/48] Installing: ant-apache-oro ####################### [42/48] Installing: ant-apache-resolver ####################### [43/48] Installing: ant-apache-regexp ####################### [44/48] Installing: ant-apache-log4j ####################### [45/48] Installing: libswt3-gtk2 ####################### [46/48] Installing: eclipse-platform ####################### [47/48] Installing: jessie From fedora at wir-sind-cool.org Fri May 6 11:17:37 2005 From: fedora at wir-sind-cool.org (Michael Schwendt) Date: Fri, 6 May 2005 13:17:37 +0200 Subject: Missing update announcements In-Reply-To: <427B05BA.7060007@www.linux.org.uk> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> <6c18a4f05050509584792d76b@mail.gmail.com> <427B05BA.7060007@www.linux.org.uk> Message-ID: <20050506131737.0d6c9349.fedora@wir-sind-cool.org> On Fri, 06 May 2005 01:50:50 -0400, Mike A. Harris wrote: > The current fedora-update mechanism is currently entirely > human driven with almost zero automation, unlike the errata > mechanisms used for Red Hat Linux in the past, and Red Hat > Enterprise Linux currently. Here I would have expected a short description of how the RHL/RHEL mechanism works currently, what automation is used there and why it has been abandoned with the step from RHL to FC. From jspaleta at gmail.com Fri May 6 13:21:36 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Fri, 6 May 2005 09:21:36 -0400 Subject: Scriptlet errors while installing eclipse In-Reply-To: <1115372581.3343.17.camel@localhost.localdomain> References: <1115372581.3343.17.camel@localhost.localdomain> Message-ID: <604aa791050506062142e1c48e@mail.gmail.com> On 5/6/05, Kyrre Ness Sjobak wrote: > I did a "yum install eclipse", and rpm showed the errors in the attached > text file. > > Thougt someone on this list probably are interested :) Bug ticket numbers would be more interesting. seems both jakarta-commons-digester and xml-commons-apis either need to require java-1.4.2-gcj-compat or test for the existance of /usr/bin/rebuild-gcj-db if the scriptlet action is optional. And since jakarta-commons-digester is using rm in the scriptlet as well, you might want to suggest that jakarta-commons-digester require coreutils in the bug report you plan to file about this. -jef From jspaleta at gmail.com Fri May 6 14:55:48 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Fri, 6 May 2005 10:55:48 -0400 Subject: Missing update announcements In-Reply-To: <427B05BA.7060007@www.linux.org.uk> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> <6c18a4f05050509584792d76b@mail.gmail.com> <427B05BA.7060007@www.linux.org.uk> Message-ID: <604aa79105050607552dbf392b@mail.gmail.com> On 5/6/05, Mike A. Harris wrote: > Call me crazy, I don't know where I managed to get all these > ideas from, but it all just came together in my mind > somehow.[1] Now that there is at least one theoretical > solution to the problem, if this solution were to be considered, > we need the following: > > 1) Someone to design the detailed solution > > 2) Someone to implement the design and test it > > 3) Put the new system in place and start using it. > > In #1 and #2 above, "someone" could be either a Red Hat employee > voluntarily taking on the task on their personal time, or it > could be a Red Hat employee taking on the task under work time, > or it could be a person in the community taking on the task. While I think outside volunteers could take a crack at implementing and testing a prototype solution outside the RedHat fenceline, some requirements as to specific implementation details would need to be communicated from inside the RedHat fenceline before anyone started working on this. Especially if at the end of the day this service needs to be migrated back onto RedHat controlled infrastructure. If RedHat wants to avoid a php based solution for example, that should be communicated up-front when Fedora leadership communicates the task proposal. If however Fedora Core leadership is okay with a helper service that lives outside the RedHat fenceline on a community managed server then implementation requirements probably aren't as tight as long as Fedora leadership is confident in the ability of the community involved in maintaining the service. Because this is a service that has to be hosted, this task is a little more complicated than the boottime graph application project we have seen initiated in the fedora lists previously. -jef From carlos.efr at mail.telepac.pt Fri May 6 15:06:22 2005 From: carlos.efr at mail.telepac.pt (Carlos Rodrigues) Date: Fri, 06 May 2005 16:06:22 +0100 Subject: Wrond disk space in "df" Message-ID: <427B87EE.4060504@mail.telepac.pt> Hi! I have an external USB 2.0 250Gb hard-drive formatted with fat32 and df always shows 64Kb of used disk space, when there are already a couple of gigabytes there. I already submitted this as bug #155819 (https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=155819) but was wondering if anybody else is seeing this problem, or is it just me? -- Carlos Rodrigues From daly at rio.sci.ccny.cuny.edu Fri May 6 16:11:44 2005 From: daly at rio.sci.ccny.cuny.edu (Tim Daly) Date: Fri, 6 May 2005 12:11:44 -0400 Subject: rawhide report: 20050505 changes In-Reply-To: <427B08C1.2060608@www.linux.org.uk> (mharris@www.linux.org.uk) References: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> <1115349789.16090.8.camel@dragon.sys.intra> <427B08C1.2060608@www.linux.org.uk> Message-ID: <200505061611.j46GBiE10506@rio.sci.ccny.cuny.edu> Actually there is a minor reason to prefer lower case. ISO9660 standard filenames are lowercase 8.3 format, I believe. Thus writing an uppercase name requires one of the minor format extensions. Tim Daly From buildsys at redhat.com Fri May 6 15:48:18 2005 From: buildsys at redhat.com (Build System) Date: Fri, 6 May 2005 11:48:18 -0400 Subject: rawhide report: 20050506 changes Message-ID: <200505061548.j46FmIeI017281@porkchop.devel.redhat.com> Removed package dev86 Updated Packages: GFS-6.1-0.pre22.3 ----------------- * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. * Wed Dec 22 2004 Chris Feist - Added code to re-read sysctl.conf. * Mon Dec 20 2004 Chris Feist - Rebuild with new sources. GFS-kernel-2.6.11.5-20050505.133825.FC4.0 ----------------------------------------- anaconda-10.2.0.61-1 -------------------- * Thu May 05 2005 Jeremy Katz - 10.2.0.61-1 - and fix pkgorder for the gfs stuff * Thu May 05 2005 Jeremy Katz - 10.2.0.60-1 - Better handling of the langsupport group (clumens) - Don't install the gfs stuff for all kernel variants, that brings in kernel-smp on an everything install (#156849) - Don't grow a partition beyond the largest freespace on a drive - HFS+ support - Pull in more selinux policy files to try to get /home labeled right - Fix typo causing segfault (pnasrat) ant-0:1.6.2-3jpp_6fc -------------------- * Thu May 05 2005 Gary Benson 0:1.6.2-3jpp_6fc - Add dependencies for %post and %postun scriptlets (#156901). booty-0.52-1 ------------ * Wed May 04 2005 Jeremy Katz - 0.52-1 - add enablecdboot, enableofboot, and enablenetboot to yaboot.conf - try to detect mac os x and set up dual boot on pmac * Mon May 02 2005 Jeremy Katz - add handling for sx8 device names ccs-0.25-0.4 ------------ * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. cman-1.0-0.pre33.7 ------------------ * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. * Mon Dec 20 2004 Chris Feist - Rebuild with new sources. * Mon Dec 13 2004 Chris Feist - Rebuild with new sources. comps-extras-10.3-1 ------------------- * Thu May 05 2005 Bill Nottingham - 10.3-1 - updated icons () ethereal-0.10.11-1 ------------------ * Thu May 05 2005 Radek Vokal 0.10.11-1 - update to upstream, many security flaws fixed evolution-connector-2.2.2-3 --------------------------- * Wed May 04 2005 David Malcolm - 2.2.2-3 - updated noinst patch: libexchange is now a convenience library again; use -R syntax to express path to Evolution's private libraries rather than -Wl since libtool cannot properly intrepret the latter; regenerated resulting patch. * Mon May 02 2005 David Malcolm - 2.2.2-2 - disabling noinst patch as not yet applied * Mon Apr 11 2005 David Malcolm - 2.2.2-1 - 2.2.2 fence-1.27-3 ------------ * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. gimp-print-4.2.7-6 ------------------ * Thu May 05 2005 Bill Nottingham 4.2.7-6 - hardlink cups ppds gjdoc-0.7.4-5 ------------- * Thu May 05 2005 Andrew Overholt 0.7.4-5 - Add patch to handle single quotes in options (Julian Scheid). - Add patch to handle lack of whitespace before member names (Julian). gnome-desktop-2.10.0-1 ---------------------- * Thu Mar 17 2005 Ray Strode - 2.10.0-1 - Update to upstream version 2.10.0 * Wed Mar 02 2005 Mark McLoughlin 2.9.91-3 - Rebuild with gcc4 * Mon Feb 21 2005 Than Ngo 2.9.91-2 - gnome-about only in GNOME-menu gphoto2-2.1.5-8 --------------- * Mon May 02 2005 David Zeuthen 2.1.5-8 - Build and install hal device information files * Tue Mar 29 2005 Tim Waugh - Re-enable docs. * Mon Mar 28 2005 Matthias Clasen 2.1.5-7 - Rebuild against newer libexif gtk-doc-1.3-2 ------------- * Thu May 05 2005 Matthias Clasen 1.3-1 - accept ':' in ids gulm-1.0-0.pre28.6 ------------------ * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. hwdata-0.157-1 -------------- * Thu May 05 2005 Jeremy Katz - 0.157-1 - add 20" Apple Cinema Display jakarta-commons-beanutils-0:1.7.0-1jpp_3fc ------------------------------------------ * Thu May 05 2005 Gary Benson - 0:1.7.0-1jpp_3fc - Add dependencies for %post and %postun scriptlets (#156901). jakarta-commons-collections-0:3.1-1jpp_3fc ------------------------------------------ * Thu May 05 2005 Gary Benson - 0:3.1-1jpp_3fc - Add dependencies for %post and %postun scriptlets (#156901). jakarta-commons-digester-0:1.6-2jpp_3fc --------------------------------------- * Thu May 05 2005 Gary Benson - 0:1.6-2jpp_3fc - Add dependencies for %post and %postun scriptlets (#156901). jakarta-commons-el-0:1.0-2jpp_2fc --------------------------------- * Thu May 05 2005 Gary Benson - 0:1.0-2jpp_2fc - BC-compile. jakarta-commons-logging-0:1.0.4-2jpp_3fc ---------------------------------------- * Thu May 05 2005 Gary Benson - 0:1.0.4-2jpp_3fc - Add dependencies for %post and %postun scriptlets (#156901). jakarta-commons-modeler-0:1.1-3jpp_3fc -------------------------------------- * Thu May 05 2005 Gary Benson - 0:1.1-3jpp_3fc - Add dependencies for %post and %postun scriptlets (#156901). mc-1:4.6.1a-0.9 --------------- * Wed May 04 2005 Jindrich Novy 4.6.1a-0.9 - update from CVS - sync with .utf8 patch - fix broken charset conversion feature in the .utf8 patch, Andrew V. Samoilov (#154516) mozilla-37:1.7.7-3 ------------------ * Wed May 04 2005 Christopher Aillon 37:1.7.7-3 - Fix for some more cursor issues in textareas (149991, 150002, 152089) - Add upstream patch to fix bidi justification of pango - Add patch to fix launching of helper applications * Tue May 03 2005 Christopher Aillon - Add patch from Marcel Mol supporting launching with filenames containing whitespace * Wed Apr 27 2005 Warren Togami - remove JVM version probing (#116445) - correct confusing PANGO vars in startup script mx4j-1:2.1.0-1jpp_6fc --------------------- * Thu May 05 2005 Gary Benson 0:2.1.0-1jpp_6fc - Add dependencies for %post and %postun scriptlets (#156901). regexp-0:1.3-1jpp_5fc --------------------- * Thu May 05 2005 Gary Benson 0:1.3-1jpp_5fc - BC-compile. rgmanager-1.9.31-2 ------------------ * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. rhpl-0.162-2 ------------ * Thu May 05 2005 Jeremy Katz - 0.162-1 - look up monitor in MonitorsDB if we can't probe horiz/vert tomcat5-0:5.0.30-5jpp_4fc ------------------------- * Thu May 05 2005 Gary Benson 0:5.0.30-5jpp_4fc - Remove references to the never used commons-daemon. - Add a runtime dependency on puretls. * Thu May 05 2005 Gary Benson 0:5.0.30-5jpp_3fc - Add dependencies for %post and %postun scriptlets (#156901). udev-057-4 ---------- * Thu May 05 2005 Bill Nottingham - 057-4 - better check for mounted tmpfs on /dev (#156862) xerces-j2-0:2.6.2-4jpp_5fc -------------------------- * Thu May 05 2005 Gary Benson 0:2.6.2-4jpp_5fc - Add dependencies for %post and %postun scriptlets (#156901). xml-commons-0:1.0-0.b2.6jpp_12fc -------------------------------- * Thu May 05 2005 Gary Benson - 0:1.0-0.b2.6jpp_12fc - Add dependencies for %post and %postun scriptlets (#156901). yum-2.3.2-2 ----------- * Thu May 05 2005 Jeremy Katz - 2.3.2-2 - handle ppc64/sparc64 "correctly" From fedora at camperquake.de Fri May 6 15:59:59 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Fri, 6 May 2005 17:59:59 +0200 Subject: rawhide report: 20050505 changes In-Reply-To: <200505061611.j46GBiE10506@rio.sci.ccny.cuny.edu> References: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> <1115349789.16090.8.camel@dragon.sys.intra> <427B08C1.2060608@www.linux.org.uk> <200505061611.j46GBiE10506@rio.sci.ccny.cuny.edu> Message-ID: <20050506155959.GA27482@ryoko.camperquake.de> On Fri, May 06, 2005 at 12:11:44PM -0400, Tim Daly wrote: > Actually there is a minor reason to prefer lower case. > ISO9660 standard filenames are lowercase 8.3 format, I believe. > Thus writing an uppercase name requires one of the minor format extensions. Is there anyone still left who cares about plain ISO9660? From notting at redhat.com Fri May 6 16:15:41 2005 From: notting at redhat.com (Bill Nottingham) Date: Fri, 6 May 2005 12:15:41 -0400 Subject: Missing update announcements In-Reply-To: <20050506131737.0d6c9349.fedora@wir-sind-cool.org> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> <6c18a4f05050509584792d76b@mail.gmail.com> <427B05BA.7060007@www.linux.org.uk> <20050506131737.0d6c9349.fedora@wir-sind-cool.org> Message-ID: <20050506161541.GC12404@nostromo.devel.redhat.com> Michael Schwendt (fedora at wir-sind-cool.org) said: > > The current fedora-update mechanism is currently entirely > > human driven with almost zero automation, unlike the errata > > mechanisms used for Red Hat Linux in the past, and Red Hat > > Enterprise Linux currently. > > Here I would have expected a short description of how the RHL/RHEL > mechanism works currently, Developer opens a request, writes an errata description, and enters the package file names. There's some automated testing, then QA takes over. Docs beats the errata description mangled by the developer into something more sane. Eventually, QA asks for the packages to be signed, then they're pushed, mail sent, etc. > what automation is used there and why > it has been abandoned with the step from RHL to FC. It has heavy ties into RHN, among other things, which didn't work very well. Bill From sean.bruno at dsl-only.net Fri May 6 16:54:07 2005 From: sean.bruno at dsl-only.net (Sean Bruno) Date: Fri, 06 May 2005 09:54:07 -0700 Subject: Dell/Adaptec Raid Controller(aacraid) Message-ID: <1115398447.6722.29.camel@homebox> I have been screaming at Dell Tech Support over the last couple of weeks trying to get them to at least acknowledge that there is a re-occuring problem with their raid controller on the Poweredge 2650(they call the adaptec controller a 3D/i). We are running AS 3(R4) and Dell TS told us to not use the v1.1.5 of the aacraid driver, but to use their driver(v1.1.4 based, but really a completely different version). I have been trying to get them to contact Red Hat, or the upstream kernel maintainers for the module, but they(Dell) are being kind of "We don't make these any more, go away!" Here is the link to the Dell v1.1.4 driver and their "dkms" system (glorified make). http://support.dell.com/support/downloads/download.aspx?c=us&cs=555&l=en&s=biz&releaseid=R71817&SystemID=PWE_FOS_XEO_2650&category=0&os=LE30&osl=en&deviceid=1375&devlib=35&fileid=93190 How do I propose a merge between the Dell v1.1.4 and the current v1.1.5 of the aacraid module in the 2.4 kernel? Sean From pbrobinson at gmail.com Fri May 6 16:58:24 2005 From: pbrobinson at gmail.com (Peter Robinson) Date: Fri, 6 May 2005 17:58:24 +0100 Subject: Dell/Adaptec Raid Controller(aacraid) In-Reply-To: <1115398447.6722.29.camel@homebox> References: <1115398447.6722.29.camel@homebox> Message-ID: <5256d0b0505060958156808@mail.gmail.com> If your running RH AS this is an issue for those mailing lists or RedHat support. This is a list for Fedora development. Fedora doesn't even support the 2.4 kernel anymore. Peter On 5/6/05, Sean Bruno wrote: > I have been screaming at Dell Tech Support over the last couple of weeks > trying to get them to at least acknowledge that there is a re-occuring > problem with their raid controller on the Poweredge 2650(they call the > adaptec controller a 3D/i). > > We are running AS 3(R4) and Dell TS told us to not use the v1.1.5 of the > aacraid driver, but to use their driver(v1.1.4 based, but really a > completely different version). > > I have been trying to get them to contact Red Hat, or the upstream > kernel maintainers for the module, but they(Dell) are being kind of "We > don't make these any more, go away!" > > Here is the link to the Dell v1.1.4 driver and their "dkms" system > (glorified make). > > http://support.dell.com/support/downloads/download.aspx?c=us&cs=555&l=en&s=biz&releaseid=R71817&SystemID=PWE_FOS_XEO_2650&category=0&os=LE30&osl=en&deviceid=1375&devlib=35&fileid=93190 > > How do I propose a merge between the Dell v1.1.4 and the current v1.1.5 > of the aacraid module in the 2.4 kernel? > > Sean > > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list > From sean.bruno at dsl-only.net Fri May 6 17:03:58 2005 From: sean.bruno at dsl-only.net (Sean Bruno) Date: Fri, 06 May 2005 10:03:58 -0700 Subject: Dell/Adaptec Raid Controller(aacraid) In-Reply-To: <5256d0b0505060958156808@mail.gmail.com> References: <1115398447.6722.29.camel@homebox> <5256d0b0505060958156808@mail.gmail.com> Message-ID: <1115399038.6722.31.camel@homebox> Crap, my bad...hit the wrong list... Sean On Fri, 2005-05-06 at 17:58 +0100, Peter Robinson wrote: > If your running RH AS this is an issue for those mailing lists or > RedHat support. This is a list for Fedora development. Fedora doesn't > even support the 2.4 kernel anymore. > > Peter > > On 5/6/05, Sean Bruno wrote: > > I have been screaming at Dell Tech Support over the last couple of weeks > > trying to get them to at least acknowledge that there is a re-occuring > > problem with their raid controller on the Poweredge 2650(they call the > > adaptec controller a 3D/i). > > > > We are running AS 3(R4) and Dell TS told us to not use the v1.1.5 of the > > aacraid driver, but to use their driver(v1.1.4 based, but really a > > completely different version). > > > > I have been trying to get them to contact Red Hat, or the upstream > > kernel maintainers for the module, but they(Dell) are being kind of "We > > don't make these any more, go away!" > > > > Here is the link to the Dell v1.1.4 driver and their "dkms" system > > (glorified make). > > > > http://support.dell.com/support/downloads/download.aspx?c=us&cs=555&l=en&s=biz&releaseid=R71817&SystemID=PWE_FOS_XEO_2650&category=0&os=LE30&osl=en&deviceid=1375&devlib=35&fileid=93190 > > > > How do I propose a merge between the Dell v1.1.4 and the current v1.1.5 > > of the aacraid module in the 2.4 kernel? > > > > Sean > > > > -- > > fedora-devel-list mailing list > > fedora-devel-list at redhat.com > > http://www.redhat.com/mailman/listinfo/fedora-devel-list > > > From kyrre at solution-forge.net Fri May 6 18:04:42 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Fri, 06 May 2005 20:04:42 +0200 Subject: Scriptlet errors while installing eclipse In-Reply-To: <604aa791050506062142e1c48e@mail.gmail.com> References: <1115372581.3343.17.camel@localhost.localdomain> <604aa791050506062142e1c48e@mail.gmail.com> Message-ID: <1115402681.3343.20.camel@localhost.localdomain> fre, 06.05.2005 kl. 15.21 skrev Jeff Spaleta: > On 5/6/05, Kyrre Ness Sjobak wrote: > > I did a "yum install eclipse", and rpm showed the errors in the attached > > text file. > > > > Thougt someone on this list probably are interested :) > > Bug ticket numbers would be more interesting. > seems both jakarta-commons-digester and xml-commons-apis > either need to require java-1.4.2-gcj-compat or test for the existance of > /usr/bin/rebuild-gcj-db if the scriptlet action is optional. > > And since jakarta-commons-digester is using rm in the scriptlet as > well, you might want to suggest that jakarta-commons-digester require > coreutils in the bug report you plan to file about this. > > -jef since you seem to have much better understanding of what's wrong than me, why don't you pick it up? Anyway, i was kindof hoping the maintainer would see it fix it... From notting at redhat.com Fri May 6 18:20:38 2005 From: notting at redhat.com (Bill Nottingham) Date: Fri, 6 May 2005 14:20:38 -0400 Subject: Fedora Core 4 Test 3 notice - slip Message-ID: <20050506182038.GB32231@nostromo.devel.redhat.com> Due to some last-minute problems that required fixing, we're going to push out the release of Fedora Core 4 Test 3 one day. It's now scheduled for Tuesday, May 10, at 10AM EDT (1400 UTC). Bill From jspaleta at gmail.com Fri May 6 18:41:23 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Fri, 6 May 2005 14:41:23 -0400 Subject: Scriptlet errors while installing eclipse In-Reply-To: <1115402681.3343.20.camel@localhost.localdomain> References: <1115372581.3343.17.camel@localhost.localdomain> <604aa791050506062142e1c48e@mail.gmail.com> <1115402681.3343.20.camel@localhost.localdomain> Message-ID: <604aa791050506114110cfebe0@mail.gmail.com> On 5/6/05, Kyrre Ness Sjobak wrote: > since you seem to have much better understanding of what's wrong than > me, why don't you pick it up? teach a man to fish.... > > Anyway, i was kindof hoping the maintainer would see it fix it... The only way to be sure that any specific maintainer will see the problem is to bugzilla it. Mailinglists are good for discussion and identifying a problem, but if you want to be sure the maintainer of any particular package is notified you need to file a ticket in bugzilla. Just throwing a message on this list doesn't garuntee you the right person is going to see or remember this singular mailinglist post in the flood of messages. In this case, someone did exactly that and its already taken care of with today's builds according to the rawhide build report https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156901 -jef"thinks the homepage search box on the new bugzilla isn't working as expected because the search 'ALL rebuild-gcj-db' does not return the 156901 bug"spaleta From mls at suse.de Fri May 6 20:45:09 2005 From: mls at suse.de (Michael Schroeder) Date: Fri, 6 May 2005 22:45:09 +0200 Subject: rpm's zlib and deltarpms In-Reply-To: <20050504182644.GA3024@suse.de> References: <20050504182644.GA3024@suse.de> Message-ID: <20050506204509.GC4600@suse.de> On Wed, May 04, 2005 at 08:26:44PM +0200, Michael Schroeder wrote: > I just found out that the rsync patch applied to rpm's internal > zlib library makes it produce zlib-incompatible payloads. That is, > a decompress and recompress with normal zlib doesn't produce the > same output. Normally this doesn't do much harm, but with deltarpms > I need to reconstruct the exact compressed payload. So at the > moment deltarpms won't work with FC4 rpms. Status update: there's now a new version that automatically detects the rsyncable gzip compression. You can get it via ftp://ftp.suse.com/pub/projects/deltarpm/deltarpm-2.3.tar.bz2 Enjoy, Michael. -- Michael Schroeder mls at suse.de main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} From linuxnow at newtral.org Fri May 6 21:32:57 2005 From: linuxnow at newtral.org (Pau Aliagas) Date: Fri, 6 May 2005 23:32:57 +0200 (CEST) Subject: Dell/Adaptec Raid Controller(aacraid) In-Reply-To: <1115399038.6722.31.camel@homebox> References: <1115398447.6722.29.camel@homebox> <5256d0b0505060958156808@mail.gmail.com> <1115399038.6722.31.camel@homebox> Message-ID: On Fri, 6 May 2005, Sean Bruno wrote: >>> How do I propose a merge between the Dell v1.1.4 and the current v1.1.5 >>> of the aacraid module in the 2.4 kernel? I've been suffering the same problem for months, but I didn't want ot instal the "Dell" module. So I waited and... good news! > Crap, my bad...hit the wrong list... Latest fedora kernel works like a charm. I used to have daily crasehs but since kernel-2.6.11-1.14_FC3smp I've had more than 2 weeks uptime. I just couldn't believe it. Pau From wtogami at redhat.com Fri May 6 23:27:32 2005 From: wtogami at redhat.com (Warren Togami) Date: Fri, 06 May 2005 13:27:32 -1000 Subject: What is wrong in this spec file? In-Reply-To: <1115322873.30401.18.camel@localhost.localdomain> References: <1115238512.23655.6.camel@fctmp> <1115322873.30401.18.camel@localhost.localdomain> Message-ID: <427BFD64.7070601@redhat.com> Peter Jones wrote: > On Wed, 2005-05-04 at 22:28 +0200, Rakotomandimby (R12y) Mihamina wrote: > [lots of stuff] > > So, lots of people have already told you to use a BuildRoot, and they're > correct. Mathew Miller also said that you'd be better off not using the > "root" user to build. He's absolutely correct, so I figure that the > instructions for this don't hit the list often enough, so here's a > simple guide to doing so: > > 1) log in as a non-root user. I like "pjones" for that, but suit > yourself ;) > > 2) create a .rpmmacros file. In it, put something like: > > %_topdir /home/pjones/build > %_tmp %{_topdir}/tmp > %_tmpdir %{_tmp} > %_tmppath %{_tmp} > > 3) make some directories: > > cd /home/pjones > mkdir build > cd build > mkdir RPMS SPECS SOURCES BUILD SRPMS tmp > > 4) test that everything's right: > > rpm -Uvh foo-1-1.src.rpm > cd ~/build/SPECS/ > rpmbuild -ba foo.spec > > It should start doing the obvious build stuff. When it's done, it'll > write packages out to ~/build/RPMS/$ARCH/foo-1.1.$ARCH.rpm, just like it > used to do in /usr/src/redhat . > You can do this, or just install fedora-rpmdevtools from Extras and run "fedora-buildrpmtree" as a non-root user. Warren From dheerajgoswami at yahoo.com Sat May 7 00:39:45 2005 From: dheerajgoswami at yahoo.com (Dheeraj Goswami) Date: Sat, 7 May 2005 01:39:45 +0100 (BST) Subject: About Latest fedora Release? Message-ID: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> Hi Friends, I wanted to know where can I find the latest (stable) fedora release to install? Is the one on redhat site latest one or there is another place to pick-up from? TIA ~dheeraj --------------------------------- Yahoo! Messenger - want a free & easy way to contact your friends online? -------------- next part -------------- An HTML attachment was scrubbed... URL: From webmaster at margo.bijoux.nom.br Sat May 7 01:05:11 2005 From: webmaster at margo.bijoux.nom.br (Pedro Fernandes Macedo) Date: Fri, 06 May 2005 22:05:11 -0300 Subject: About Latest fedora Release? In-Reply-To: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> Message-ID: <427C1447.6080908@margo.bijoux.nom.br> Dheeraj Goswami wrote: > Hi Friends, > I wanted to know where can I find the latest (stable) fedora release > to install? Is the one on redhat site latest one or there is another > place to pick-up from? > This is the wrong list for the question. You should had used fedora-list . The latest stable release can always be found at http://fedora.redhat.com/download/ . For mirrors , check http://fedora.redhat.com/download/mirrors.html . -- Pedro Macedo From prasadvvv at lucent.com Sat May 7 01:07:47 2005 From: prasadvvv at lucent.com (Vendra, Hari Prasad V V P Ch S (Hari Prasad)) Date: Sat, 7 May 2005 06:37:47 +0530 Subject: PPP sessions Message-ID: <6733C768256DEC42A72BAFEFA9CF06D210033CBF@ii0015exch002u.iprc.lucent.com> Hi, I am using Fedora 3 Kernel 2.6.9-1.667. After opening 155 PPP sessions. My system automatically hanging and rebooting. I don't know what could be the reason. I am using the following command /opt/ppp-2.4.3/pppd/pppd /dev/pts/159 -detach asyncmap 0 user CLIENT1 noipdefault local noauth persist Could you please suggest me , how do i debug. Regards Hari From thesource at ldb-jab.org Sat May 7 01:26:25 2005 From: thesource at ldb-jab.org (Lawrence Bowie) Date: Fri, 06 May 2005 21:26:25 -0400 Subject: About Latest fedora Release? In-Reply-To: <427C1447.6080908@margo.bijoux.nom.br> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> <427C1447.6080908@margo.bijoux.nom.br> Message-ID: <427C1941.60806@ldb-jab.org> He just asked a simple question .. crap .. can you be accomodating .. LDB Pedro Fernandes Macedo wrote: > Dheeraj Goswami wrote: > >> Hi Friends, >> I wanted to know where can I find the latest (stable) fedora release >> to install? Is the one on redhat site latest one or there is another >> place to pick-up from? >> > > > > This is the wrong list for the question. You should had used > fedora-list . > The latest stable release can always be found at > http://fedora.redhat.com/download/ . For mirrors , check > http://fedora.redhat.com/download/mirrors.html . > > -- > Pedro Macedo > From skvidal at phy.duke.edu Sat May 7 01:28:17 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Fri, 06 May 2005 21:28:17 -0400 Subject: About Latest fedora Release? In-Reply-To: <427C1941.60806@ldb-jab.org> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> <427C1447.6080908@margo.bijoux.nom.br> <427C1941.60806@ldb-jab.org> Message-ID: <1115429297.15593.16.camel@cutter> On Fri, 2005-05-06 at 21:26 -0400, Lawrence Bowie wrote: > He just asked a simple question .. crap .. can you be accomodating .. > He wasn't rude. He just told him that he was offtopic and he nicely told him where he could get the answer to his question. -sv From sean.bruno at dsl-only.net Sat May 7 01:53:16 2005 From: sean.bruno at dsl-only.net (Sean Bruno) Date: Fri, 06 May 2005 18:53:16 -0700 Subject: ARTSD In-Reply-To: <1114968503.6971.11.camel@localhost.localdomain> References: <042920052009.431.427294910005280B000001AF2200748184020A04049A030C03A103079B@comcast.net> <200504292214.06093.ml-fedora@fathomssen.de> <000a01c54d0e$aebefc00$0201a8c0@Greeney> <200504301854.44620.rjune@bravegnuworld.com> <20050501052638.A18884@tiki-lounge.com> <1114968503.6971.11.camel@localhost.localdomain> Message-ID: <1115430796.6722.84.camel@homebox> On Sun, 2005-05-01 at 10:28 -0700, Per Bjornsson wrote: > On Sun, 2005-05-01 at 05:26 -0700, Toshio Kuratomi wrote: > > > AFAIK ALSA is Linux specific whereas gnome and kde are cross-platform. So > > the important thing isn't to have alsa do network audio. It's to have a > > cross platform API that addresses the present shortcomings (Alan posts > > about those nearly every time this comes up.) Then it can be implemented on > > Linux and other platforms in whatever form makes sense. > > Well, the solution that GNOME seems to have converged on is to use > Gstreamer and use whatever default output method Gstreamer is set up > for. On many desktop systems, using the Gstreamer alsasink output plugin > would be the best; in other cases you can either use e.g. esd or some > other sound server which can provide network transparency. At least some > KDE programs use Gstreamer as well, and apparently there is talk of > using Gstreamer pretty much as the default audio solution for KDE 4. So > it seems that Gstreamer can fulfill the role of being _the_ audio API > that desktop apps use. > > /Per > > -- > Per Bjornsson > Ph.D. Candidate, Department of Applied Physics, Stanford University > FYI, it looks like the issue with artsd going mental has departed. Thanks to whomever looked into it and repaired the problem! Sean From nman64 at n-man.com Sat May 7 03:27:55 2005 From: nman64 at n-man.com (Patrick Barnes) Date: Fri, 06 May 2005 22:27:55 -0500 Subject: About Latest fedora Release? In-Reply-To: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> Message-ID: <427C35BB.3040802@n-man.com> http://fedora.redhat.com/participate/communicate/ clearly states: fedora-devel-list - For developers, developers, developers. If you are interested in helping create Fedora Core releases, this is the list for you. http://www.redhat.com/mailman/listinfo/fedora-devel-list clearly states: THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION ONLY. I don't think it can be any more clear. I don't know how so many people miss this simple concept. This list is NOT for people to ask questions of the developers. I have to suggest changing any page that refers to this list to state in large/bold print the previous sentence. For the sake of not being completely rude, the answer is: Fedora Core 3, which is clearly listed on the main site, is the latest stable release. All future stable releases will be similarly clearly announced on the main site in a timely manner. There is never any need to check for a more recent version than is named there. The next stable release, Fedora Core 4, is scheduled for release on June 6th. -Patrick "The N-Man" Barnes nman64 at n-man.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: OpenPGP digital signature URL: From nman64 at n-man.com Sat May 7 03:28:16 2005 From: nman64 at n-man.com (Patrick Barnes) Date: Fri, 06 May 2005 22:28:16 -0500 Subject: PPP sessions In-Reply-To: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> Message-ID: <427C35D0.5060801@n-man.com> http://fedora.redhat.com/participate/communicate/ clearly states: fedora-devel-list - For developers, developers, developers. If you are interested in helping create Fedora Core releases, this is the list for you. http://www.redhat.com/mailman/listinfo/fedora-devel-list clearly states: THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION ONLY. I don't think it can be any more clear. I don't know how so many people miss this simple concept. This list is NOT for people to ask questions of the developers. I have to suggest changing any page that refers to this list to state in large/bold print the previous sentence. For the sake of not being completely rude, the answer is: First, try upgrading. You're using the original FC3 kernel, which has known issues. If you want any kind of support for your query, you need to try the latest kernel update for FC3. If that still gives you trouble, seek help in the appropriate list, and then, if necessary, file a report in Bugzilla. -Patrick "The N-Man" Barnes nman64 at n-man.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: OpenPGP digital signature URL: From thesource at ldb-jab.org Sat May 7 03:46:14 2005 From: thesource at ldb-jab.org (Lawrence Bowie) Date: Fri, 06 May 2005 23:46:14 -0400 Subject: About Latest fedora Release? In-Reply-To: <427C35BB.3040802@n-man.com> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> <427C35BB.3040802@n-man.com> Message-ID: <427C3A05.5030505@ldb-jab.org> Yes, thank you for that excerpt. But why be religious about it? He just asked a question .. Guess what? It will happen again and again .. do not answer it if it is not appropiate. LDB Patrick Barnes wrote: > http://fedora.redhat.com/participate/communicate/ clearly states: > fedora-devel-list - For developers, developers, developers. If you are > interested in helping create Fedora Core releases, this is the list > for you. > > http://www.redhat.com/mailman/listinfo/fedora-devel-list clearly states: > THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION > ONLY. > > I don't think it can be any more clear. I don't know how so many people > miss this simple concept. This list is NOT for people to ask questions > of the developers. I have to suggest changing any page that refers to > this list to state in large/bold print the previous sentence. > > For the sake of not being completely rude, the answer is: Fedora Core 3, > which is clearly listed on the main site, is the latest stable release. > All future stable releases will be similarly clearly announced on the > main site in a timely manner. There is never any need to check for a > more recent version than is named there. The next stable release, > Fedora Core 4, is scheduled for release on June 6th. > > -Patrick "The N-Man" Barnes > nman64 at n-man.com > > From ottohaliburton at comcast.net Sat May 7 09:05:43 2005 From: ottohaliburton at comcast.net (Otto Haliburton) Date: Sat, 7 May 2005 04:05:43 -0500 Subject: About Latest fedora Release? In-Reply-To: <1115429297.15593.16.camel@cutter> Message-ID: <000f01c552e3$f38fab60$4801a8c0@C515816A> > -----Original Message----- > From: fedora-devel-list-bounces at redhat.com [mailto:fedora-devel-list- > bounces at redhat.com] On Behalf Of seth vidal > Sent: Friday, May 06, 2005 8:28 PM > To: Development discussions related to Fedora Core > Subject: Re: About Latest fedora Release? > > On Fri, 2005-05-06 at 21:26 -0400, Lawrence Bowie wrote: > > He just asked a simple question .. crap .. can you be accomodating .. > > > > He wasn't rude. He just told him that he was offtopic and he nicely told > him where he could get the answer to his question. > > -sv > As has been said before it is easier to ignore the post and keep on pushing than to try to lecture someone on the place to make a post. The lecture has ended up with lectures to the lecture. It would have been easier to ignore the post or answer the question and go to another topic than it is to tell someone where to go to ask the question. This will happen all the time on all list, and developers are no exception. But the holier than thou attitude gets you nowhere. And he was rude!!!!! From wbeebe at gmail.com Sat May 7 11:34:54 2005 From: wbeebe at gmail.com (William Beebe) Date: Sat, 7 May 2005 07:34:54 -0400 Subject: About Latest fedora Release? In-Reply-To: <000f01c552e3$f38fab60$4801a8c0@C515816A> References: <1115429297.15593.16.camel@cutter> <000f01c552e3$f38fab60$4801a8c0@C515816A> Message-ID: Yes, it is easier to ignore than reply. However, you can reply and then say something (politely) that here's where to find out such information in the future. It's sad we've forgotten the admonition to go the extra mile. On 5/7/05, Otto Haliburton wrote: > > > > > -----Original Message----- > > From: fedora-devel-list-bounces at redhat.com [mailto:fedora-devel-list- > > bounces at redhat.com] On Behalf Of seth vidal > > Sent: Friday, May 06, 2005 8:28 PM > > To: Development discussions related to Fedora Core > > Subject: Re: About Latest fedora Release? > > > > On Fri, 2005-05-06 at 21:26 -0400, Lawrence Bowie wrote: > > > He just asked a simple question .. crap .. can you be accomodating .. > > > > > > > He wasn't rude. He just told him that he was offtopic and he nicely told > > him where he could get the answer to his question. > > > > -sv > > > As has been said before it is easier to ignore the post and keep on > pushing > than to try to lecture someone on the place to make a post. The lecture > has > ended up with lectures to the lecture. It would have been easier to ignore > the post or answer the question and go to another topic than it is to tell > someone where to go to ask the question. This will happen all the time on > all list, and developers are no exception. But the holier than thou > attitude gets you nowhere. And he was rude!!!!! > > > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From buildsys at redhat.com Sat May 7 12:06:17 2005 From: buildsys at redhat.com (Build System) Date: Sat, 7 May 2005 08:06:17 -0400 Subject: rawhide report: 20050507 changes Message-ID: <200505071206.j47C6Hid003879@porkchop.devel.redhat.com> Removed package jakarta-commons-daemon Updated Packages: GFS-6.1-0.pre22.4 ----------------- * Fri May 06 2005 Chris Feist - Cleaned up .spec file and %files section. * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. * Wed Dec 22 2004 Chris Feist - Added code to re-read sysctl.conf. GFS-kernel-2.6.11.5-20050505.133825.FC4.2 ----------------------------------------- * Fri May 06 2005 Chris Feist - Cleanup spec file and fix %files section. ccs-0.25-0.9 ------------ * Fri May 06 2005 Chris Feist - Cleanup .spec file to not include globs. cman-1.0-0.pre33.8 ------------------ cman-kernel-2.6.11.3-20050425.154843.FC4.11 ------------------------------------------- * Fri May 06 2005 Chris Feist - Cleaned up .spec file. dlm-1.0-0.pre21.4 ----------------- * Fri May 06 2005 Chris Feist - Cleaned up .spec file. * Mon Dec 20 2004 Chris Feist - various bug fixes (new upstream sources) * Fri Dec 10 2004 Chris Feist - various bug fixes (new upstream sources) dlm-kernel-2.6.11.3-20050425.154843.FC4.13 ------------------------------------------ * Fri May 06 2005 Chris Feist - Cleaned up .spec file. eclipse-1:3.1.0_fc-0.M6.15 -------------------------- * Thu May 05 2005 Andrew Overholt 3.1.0_fc-0.M6.15 - Rebuild with new gjdoc (rh#152049). * Wed May 04 2005 Ben Konrath - Re-enable jdt.ui/jdt.jar.so and require gcj 4.0.0-2 (rh#151296). eclipse-bugzilla-1:0.1.0_fc-15 ------------------------------ * Thu May 05 2005 Jeff Pound 0.1.0_fc-15 - Fix provider expansion when adding queries. - Add name parings for default urls. fence-1.27-10 ------------- * Fri May 06 2005 Chris Feist - Cleaned up spec file (bz #157031). gnbd-1.0-0.pre13.2 ------------------ * Fri May 06 2005 Chris Feist - Cleanup .spec file, don't glob /usr/share/man. * Mon Dec 20 2004 Chris Feist - Rebuild with new sources. * Mon Dec 13 2004 Chris Feist - Rebuild with new sources. gnbd-kernel-2.6.11.2-20050420.133124.FC4.15 ------------------------------------------- * Fri May 06 2005 Chris Feist - Cleanup .spec file. Don't include file globs anymore. gulm-1.0-0.pre28.7 ------------------ * Thu May 05 2005 Bill Nottingham - fix chkconfig line gulm-1.0-0.pre28.9 ------------------ * Fri May 06 2005 Chris Feist - fixed various problems with .spec file * Thu May 05 2005 Bill Nottingham - fix chkconfig line iddev-1.9-20 ------------ * Fri May 06 2005 Chris Feist - Cleaned up .spec file. kernel-2.6.11-1.1287_FC4 ------------------------ * Thu May 05 2005 David Woodhouse - Import audit fixes from upstream magma-1.0-0.pre21.7 ------------------- * Fri May 06 2005 Chris Feist - .spec file cleanup magma-plugins-1.0-0.pre16.12 ---------------------------- * Fri May 06 2005 Chris Feist - Cleaned up spec file (i386 i486 i586 i686 pentium3 pentium4 athlon, %files) mx4j-1:2.1.0-1jpp_7fc --------------------- * Fri May 06 2005 Gary Benson 0:2.1.0-1jpp_7fc - Fix ownership in /usr/share/java. rgmanager-1.9.31-3 ------------------ struts11-0:1.1-1jpp_5fc ----------------------- * Fri May 06 2005 Gary Benson - 0:1.1-1jpp_5fc - BC-compile. system-config-bind-4.0.0-14_FC4 ------------------------------- * Thu May 05 2005 Jason Vas Dias - 4.0.0-14 - fix out-of-zone data reporting - fix bug 156913: wrong file permissions for config files * Wed May 04 2005 Jason Vas Dias - 4.0.0-12 - fix bug 156884: handle named.conf with NO options clause . system-config-date-1.7.18-1 --------------------------- * Fri May 06 2005 Nils Philippsen 1.7.18 - make desktop file translatable (#156792) - avoid DeprecationWarnings - use DESTDIR consistently (#156782) system-config-nfs-1.3.9-1 ------------------------- * Fri May 06 2005 Nils Philippsen 1.3.9 - make desktop file rebuild consistently * Fri May 06 2005 Nils Philippsen 1.3.8 - make desktop file translatable (#156796) - use DESTDIR consistently (#156786) system-config-samba-1.2.31-1 ---------------------------- * Fri May 06 2005 Nils Philippsen - 1.2.31 - use DESTDIR consistently * Fri May 06 2005 Nils Philippsen - 1.2.30 - make desktop file translatable (#156798) system-config-services-0.8.24-1 ------------------------------- * Fri May 06 2005 Nils Philippsen - 0.8.24 - make "make update-po" pick up translatable strings in desktop file (#156801) * Fri May 06 2005 Nils Philippsen - 0.8.23 - pick up new translations system-config-users-1.2.37-1 ---------------------------- * Fri May 06 2005 Nils Philippsen - 1.2.37 - make About menu entry translate (#156793) * Fri May 06 2005 Nils Philippsen - 1.2.36 - use DESTDIR consistently * Wed May 04 2005 Nils Philippsen - make desktop file translatable (#156793) tomcat5-0:5.0.30-5jpp_5fc ------------------------- * Fri May 06 2005 Gary Benson 0:5.0.30-5jpp_5fc - Add ant-launcher.jar to common classloader. - Use absolute paths for rebuild-gcj-db. - Don't include jars from other rpms in /usr/share/tomcat5/bin. - Correct permissions and remove junk from /usr/share/tomcat5/bin. From roozbeh at farsiweb.info Sat May 7 12:23:39 2005 From: roozbeh at farsiweb.info (Roozbeh Pournader) Date: Sat, 07 May 2005 16:53:39 +0430 Subject: Fedora Core 4 Test 3 notice - slip In-Reply-To: <20050506182038.GB32231@nostromo.devel.redhat.com> References: <20050506182038.GB32231@nostromo.devel.redhat.com> Message-ID: <1115468619.5138.50.camel@tameshk.bamdad.org> On Fri, 2005-05-06 at 14:20 -0400, Bill Nottingham wrote: > Due to some last-minute problems that required fixing, we're going to > push out the release of Fedora Core 4 Test 3 one day. It's now scheduled > for Tuesday, May 10, at 10AM EDT (1400 UTC). It seems that the schedule page and the iCalendar file are not updated. Would someone please update them? roozbeh From kyrre at solution-forge.net Sat May 7 12:32:49 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sat, 07 May 2005 14:32:49 +0200 Subject: Firefox etc. default homepage Message-ID: <1115469169.3343.17.camel@localhost.localdomain> Today we use the release notes as standard homepage for all browsers, instead of whatever is upstream default. Why do we do that, while constantly repeating the mantra "upstream!, upstream!, upstream!"? In my personal opinion, the release notes belong in "help" or something like that (even a shortcut on the standard desktop would be more intuitive) - not as the standard homepage in every browser. This confuses people, and addmitedly, the release notes aren't that much usefull to the users as the standard homepage - especially in firefox, where the standard is: http://www.google.no/firefox?client=firefox-a&rls=org.mozilla:nb-NO:official which is simply a google search with some hints on using firefox etc. Kyrre Ness Sj?b?k From sundaram at redhat.com Sat May 7 12:36:32 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Sat, 07 May 2005 18:06:32 +0530 Subject: Firefox etc. default homepage In-Reply-To: <1115469169.3343.17.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> Message-ID: <427CB650.8020508@redhat.com> Kyrre Ness Sjobak wrote: >Today we use the release notes as standard homepage for all browsers, >instead of whatever is upstream default. Why do we do that, while >constantly repeating the mantra "upstream!, upstream!, upstream!"? > > Even after having the release notes setup as the default page, many users miss out important information present in it. If we are moving to the upstream default, then we need a better prominent help option available in the panel or menu by default. regards Rahul From jos at xos.nl Sat May 7 12:37:56 2005 From: jos at xos.nl (Jos Vos) Date: Sat, 7 May 2005 14:37:56 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115469169.3343.17.camel@localhost.localdomain>; from kyrre@solution-forge.net on Sat, May 07, 2005 at 02:32:49PM +0200 References: <1115469169.3343.17.camel@localhost.localdomain> Message-ID: <20050507143756.A29829@xos037.xos.nl> On Sat, May 07, 2005 at 02:32:49PM +0200, Kyrre Ness Sjobak wrote: > Today we use the release notes as standard homepage for all browsers, > instead of whatever is upstream default. Why do we do that, while > constantly repeating the mantra "upstream!, upstream!, upstream!"? I personally prefer a local (file://...) homepage for a browser as default i.s.o. a homepage requiring a working network, DNS, etc. so I consider it a good choice, but YMMV... -- -- Jos Vos -- X/OS Experts in Open Systems BV | Phone: +31 20 6938364 -- Amsterdam, The Netherlands | Fax: +31 20 6948204 From kyrre at solution-forge.net Sat May 7 13:01:53 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sat, 07 May 2005 15:01:53 +0200 Subject: Standard yum configuration Message-ID: <1115470913.3343.47.camel@localhost.localdomain> Some questions about the standard yum config: - Will the fedora/fedora-extras gpgkeys be added by default, or could the yum error message at least give some hints how to install them? Could it even ask "%s gpgkey not installed, would you like to install them now? [y/N]" like up2date? I think you must thrust RH anyway, as you allready has installed the whole OS packaged and signed by them... - Will fedora-extras.repo be included in /etc/yum.repos.d/ by default? Will it be enabled by default? - Will pup be installed by default? What will happen to up2date? Kyrre Ness Sj?b?k From kyrre at solution-forge.net Sat May 7 13:06:08 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sat, 07 May 2005 15:06:08 +0200 Subject: Firefox etc. default homepage In-Reply-To: <427CB650.8020508@redhat.com> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> Message-ID: <1115471168.3343.51.camel@localhost.localdomain> l?r, 07.05.2005 kl. 14.36 skrev Rahul Sundaram: > Kyrre Ness Sjobak wrote: > > >Today we use the release notes as standard homepage for all browsers, > >instead of whatever is upstream default. Why do we do that, while > >constantly repeating the mantra "upstream!, upstream!, upstream!"? > > > > > Even after having the release notes setup as the default page, many > users miss out important information present in it. If we are moving to > the upstream default, then we need a better prominent help option > available in the panel or menu by default. Definitively. Perhaps giving the release notes when people ask for them, together with other documentation - and not try to "push it down their throat"? Another reason to put it somewhere else, is to ensure it doesn't "get lost" when the user changes his/her homepage to something more usefull... Not every user is also an admin, and the stuff in the release-notes is primarily meant for admins. From jspaleta at gmail.com Sat May 7 13:19:31 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Sat, 7 May 2005 09:19:31 -0400 Subject: Standard yum configuration In-Reply-To: <1115470913.3343.47.camel@localhost.localdomain> References: <1115470913.3343.47.camel@localhost.localdomain> Message-ID: <604aa79105050706196a9bacee@mail.gmail.com> On 5/7/05, Kyrre Ness Sjobak wrote: > Some questions about the standard yum config: > - Will the fedora/fedora-extras gpgkeys be added by default, or could yum in rawhide can read a gpg key definition url from the .repo file. If the repo file includes the gpgkey url yum will attempt to import that key if its needed. man yum.conf look for gpgkey The only thing that is needed is to make sure the repo files included in fedora-release include an appropriate gpgkey url. -jef From ivazquez at ivazquez.net Sat May 7 13:31:30 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Sat, 07 May 2005 09:31:30 -0400 Subject: Standard yum configuration In-Reply-To: <604aa79105050706196a9bacee@mail.gmail.com> References: <1115470913.3343.47.camel@localhost.localdomain> <604aa79105050706196a9bacee@mail.gmail.com> Message-ID: <1115472690.7124.11.camel@ignacio.ignacio.lan> On Sat, 2005-05-07 at 09:19 -0400, Jeff Spaleta wrote: > On 5/7/05, Kyrre Ness Sjobak wrote: > > Some questions about the standard yum config: > > - Will the fedora/fedora-extras gpgkeys be added by default, or could > > yum in rawhide can read a gpg key definition url from the .repo file. Or even in FC3 updates. But I digress... -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From sundaram at redhat.com Sat May 7 13:34:18 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Sat, 07 May 2005 19:04:18 +0530 Subject: Firefox etc. default homepage In-Reply-To: <1115471168.3343.51.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> Message-ID: <427CC3DA.5040701@redhat.com> Hi >Definitively. Perhaps giving the release notes when people ask for them, >together with other documentation - and not try to "push it down their >throat"? > Users do not ask for information like this typically because they arent aware that its documented at all. A promiment display on the browser isnt equivalent to "push it down their throat" . We might replace those by a first login wizard which guides them through instead of the current setup but it still requires a glaring display to make sure users wont miss out. " Another reason to put it somewhere else, is to ensure it doesn't "get lost" when the user changes his/her homepage to something more usefull..." good point " Not every user is also an admin, and the stuff in the release-notes is primarily meant for admins." I have been working on a more end user friendly version of the release notes which can be used for non-root logins as the default page regards Rahul From arjanv at redhat.com Sat May 7 13:41:57 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Sat, 07 May 2005 15:41:57 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115469169.3343.17.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> Message-ID: <1115473317.6388.44.camel@laptopd505.fenrus.org> On Sat, 2005-05-07 at 14:32 +0200, Kyrre Ness Sjobak wrote: > Today we use the release notes as standard homepage for all browsers, > instead of whatever is upstream default. Why do we do that, while > constantly repeating the mantra "upstream!, upstream!, upstream!"? for me the "upstream" mantra applies far more to the code than to any actual settings of applications. If the user can and should change things, I don't see an issue with Fedora doing so as shipped as well. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From ottohaliburton at comcast.net Sat May 7 13:42:29 2005 From: ottohaliburton at comcast.net (Otto Haliburton) Date: Sat, 7 May 2005 08:42:29 -0500 Subject: About Latest fedora Release? In-Reply-To: Message-ID: <002301c5530a$9d635bc0$4801a8c0@C515816A> I hate to top post on this list cause in general it is a no no. Let's give an example that may end this thing. You go to a gastrointestinal surgeon and you tell him I have heartburn and he sends you to a doctor for the treatment. Do you want that surgeon to ever operate on you? I wouldn't what if his operation caused me to have heartburn then doesn't he know enough to solve the problem. It is the developers in many ways that are causing the problems that are generating the question and if they don't know the answer do you want them to develop stuff for the linux system. I think not!!!!!! -----Original Message----- From: fedora-devel-list-bounces at redhat.com [mailto:fedora-devel-list-bounces at redhat.com] On Behalf Of William Beebe Sent: Saturday, May 07, 2005 6:35 AM To: Development discussions related to Fedora Core Subject: Re: About Latest fedora Release? Yes, it is easier to ignore than reply. However, you can reply and then say something (politely) that here's where to find out such information in the future. It's sad we've forgotten the admonition to go the extra mile. On 5/7/05, Otto Haliburton wrote: > -----Original Message----- > From: fedora-devel-list-bounces at redhat.com [mailto:fedora-devel-list- > bounces at redhat.com] On Behalf Of seth vidal > Sent: Friday, May 06, 2005 8:28 PM > To: Development discussions related to Fedora Core > Subject: Re: About Latest fedora Release? > > On Fri, 2005-05-06 at 21:26 -0400, Lawrence Bowie wrote: > > He just asked a simple question .. crap .. can you be accomodating .. > > > > He wasn't rude. He just told him that he was offtopic and he nicely told > him where he could get the answer to his question. > > -sv > As has been said before it is easier to ignore the post and keep on pushing than to try to lecture someone on the place to make a post. The lecture has ended up with lectures to the lecture. It would have been easier to ignore the post or answer the question and go to another topic than it is to tell someone where to go to ask the question. This will happen all the time on all list, and developers are no exception. But the holier than thou attitude gets you nowhere. And he was rude!!!!! -- fedora-devel-list mailing list fedora-devel-list at redhat.com http://www.redhat.com/mailman/listinfo/fedora-devel-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From mbneto at gmail.com Sat May 7 14:25:22 2005 From: mbneto at gmail.com (mbneto) Date: Sat, 7 May 2005 10:25:22 -0400 Subject: %postun Error Message-ID: <5cf776b805050707254407d05d@mail.gmail.com> Hi, The latest updates complain about a postrun. Seems rpm related. i.e ... Cleanup : ant ####################### [36/38] (ant-1.6.2-3jpp_5fc.i386) scriptlet failed, exit status 255 It occurs with other packages as well. From j.w.r.degoede at hhs.nl Sat May 7 14:53:54 2005 From: j.w.r.degoede at hhs.nl (Hans de Goede) Date: Sat, 07 May 2005 16:53:54 +0200 Subject: %postun Error In-Reply-To: <5cf776b805050707254407d05d@mail.gmail.com> References: <5cf776b805050707254407d05d@mail.gmail.com> Message-ID: <427CD682.5070603@hhs.nl> mbneto wrote: > Hi, > > The latest updates complain about a postrun. Seems rpm related. > > i.e > ... > Cleanup : ant ####################### [36/38] > (ant-1.6.2-3jpp_5fc.i386) scriptlet failed, exit status 255 > > It occurs with other packages as well. > Hmm, I had this to when I did a piecemeal upgrade to rawhide, the fix for me was to install the latest selinux libs (can't remember the proper package name) or just make sure your system is fully uptodate with yum -y. Better make sure this is fixex soon, because each post-uninstall script which fails causes the old version of the package to stay in the rpmdb. After you've got this fixed you can rpm -qa |sort, look for duplictes and rpm -e --dbonly the dups, this might cause problems if files we're moved / removed, but not doing --justdb will nuke files from the newer version. Regards, Hans From nman64 at n-man.com Sat May 7 01:23:48 2005 From: nman64 at n-man.com (Patrick Barnes) Date: Fri, 06 May 2005 20:23:48 -0500 Subject: PPP sessions In-Reply-To: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> Message-ID: <427C18A4.20609@n-man.com> http://fedora.redhat.com/participate/communicate/ clearly states: fedora-devel-list - For developers, developers, developers. If you are interested in helping create Fedora Core releases, this is the list for you. http://www.redhat.com/mailman/listinfo/fedora-devel-list clearly states: THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION ONLY. I don't think it can be any more clear. I don't know how so many people miss this simple concept. This list is NOT for people to ask questions of the developers. I have to suggest changing any page that refers to this list to state in large/bold print the previous sentence. For the sake of not being completely rude, the answer is: First, try upgrading. You're using the original FC3 kernel, which has known issues. If you want any kind of support for your query, you need to try the latest kernel update for FC3. If that still gives you trouble, seek help in the appropriate list, and then, if necessary, file a report in Bugzilla. -Patrick "The N-Man" Barnes nman64 at n-man.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: OpenPGP digital signature URL: From nman64 at n-man.com Sat May 7 01:18:03 2005 From: nman64 at n-man.com (Patrick Barnes) Date: Fri, 06 May 2005 20:18:03 -0500 Subject: About Latest fedora Release? In-Reply-To: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> Message-ID: <427C174B.8050907@n-man.com> http://fedora.redhat.com/participate/communicate/ clearly states: fedora-devel-list - For developers, developers, developers. If you are interested in helping create Fedora Core releases, this is the list for you. http://www.redhat.com/mailman/listinfo/fedora-devel-list clearly states: THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION ONLY. I don't think it can be any more clear. I don't know how so many people miss this simple concept. This list is NOT for people to ask questions of the developers. I have to suggest changing any page that refers to this list to state in large/bold print the previous sentence. For the sake of not being completely rude, the answer is: Fedora Core 3, which is clearly listed on the main site, is the latest stable release. All future stable releases will be similarly clearly announced on the main site in a timely manner. There is never any need to check for a more recent version than is named there. The next stable release, Fedora Core 4, is scheduled for release on June 6th. -Patrick "The N-Man" Barnes nman64 at n-man.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: OpenPGP digital signature URL: From kyrre at solution-forge.net Sat May 7 15:25:52 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sat, 07 May 2005 17:25:52 +0200 Subject: Standard yum configuration In-Reply-To: <604aa79105050706196a9bacee@mail.gmail.com> References: <1115470913.3343.47.camel@localhost.localdomain> <604aa79105050706196a9bacee@mail.gmail.com> Message-ID: <1115479552.3343.53.camel@localhost.localdomain> l?r, 07.05.2005 kl. 15.19 skrev Jeff Spaleta: > On 5/7/05, Kyrre Ness Sjobak wrote: > > Some questions about the standard yum config: > > - Will the fedora/fedora-extras gpgkeys be added by default, or could > > yum in rawhide can read a gpg key definition url from the .repo file. > If the repo file includes the gpgkey url yum will attempt to import > that key if its needed. > man yum.conf look for gpgkey > > The only thing that is needed is to make sure the repo files included > in fedora-release include an appropriate gpgkey url. > > -jef https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=157144 From steve at rueb.com Sat May 7 15:37:17 2005 From: steve at rueb.com (Steve Bergman) Date: Sat, 07 May 2005 10:37:17 -0500 Subject: About Latest fedora Release? In-Reply-To: <427C3A05.5030505@ldb-jab.org> References: <20050507003945.84641.qmail@web30810.mail.mud.yahoo.com> <427C35BB.3040802@n-man.com> <427C3A05.5030505@ldb-jab.org> Message-ID: <427CE0AD.5090203@rueb.com> Lawrence Bowie wrote: > Yes, thank you for that excerpt. But why be religious about it? > He just asked a question .. Guess what? It will happen > again and again .. do not answer it if it is not appropiate. > Well, perhaps I should not contribute to the prolongation of this thread by posting to it, but hey. I don't think that the original answering post was rude. I thought it was helpful, if a bit terse. I'm usually one to call people on being rude to newbies on mailing lists, but it never occured to me that the reply was rude. The information as to what list is appropriate for end user questions will no doubt be helpful to a newcomer to Fedora, and the inclusion of that infomation enhanced the value of the post. I don't think that silence would have been an appropriate response. Add to that the fact that this is a developers' list (Note to self: Then why am I posting to this OT thread?), and the fact that it is very clearly labelled as such and I really do not see the problem here. Now on fedora-list, I do believe that it is desirable to go out of one's way to be extra polite. On the fedora-list, we are Fedora's ambassadors to the rest of the world. -Steve Bergman From kyrre at solution-forge.net Sat May 7 15:43:33 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sat, 07 May 2005 17:43:33 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115473317.6388.44.camel@laptopd505.fenrus.org> References: <1115469169.3343.17.camel@localhost.localdomain> <1115473317.6388.44.camel@laptopd505.fenrus.org> Message-ID: <1115480612.3343.56.camel@localhost.localdomain> l?r, 07.05.2005 kl. 15.41 skrev Arjan van de Ven: > On Sat, 2005-05-07 at 14:32 +0200, Kyrre Ness Sjobak wrote: > > Today we use the release notes as standard homepage for all browsers, > > instead of whatever is upstream default. Why do we do that, while > > constantly repeating the mantra "upstream!, upstream!, upstream!"? > > for me the "upstream" mantra applies far more to the code than to any > actual settings of applications. If the user can and should change > things, I don't see an issue with Fedora doing so as shipped as well. > But you still ship with "spatial" nautilus, and the last discussion about it (as far as i remember) contained a lot of "upstream is configured with spatial as default, so we must as well"... Kyrre Sj?b?k From sundaram at redhat.com Sat May 7 15:55:23 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Sat, 07 May 2005 21:25:23 +0530 Subject: Firefox etc. default homepage In-Reply-To: <1115480612.3343.56.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <1115473317.6388.44.camel@laptopd505.fenrus.org> <1115480612.3343.56.camel@localhost.localdomain> Message-ID: <427CE4EB.6060005@redhat.com> Hi >> >> >> >But you still ship with "spatial" nautilus, and the last discussion >about it (as far as i remember) contained a lot of "upstream is >configured with spatial as default, so we must as well"... > >Kyrre Sj?b?k > > > dont bring the spatial flamewar yet again please. if you dont like the setting just turn it off. its pretty easy to do that now. regards Rahul From otaylor at redhat.com Sat May 7 16:52:55 2005 From: otaylor at redhat.com (Owen Taylor) Date: Sat, 07 May 2005 12:52:55 -0400 Subject: Firefox etc. default homepage In-Reply-To: <1115469169.3343.17.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> Message-ID: <1115484775.29653.21.camel@localhost.localdomain> On Sat, 2005-05-07 at 14:32 +0200, Kyrre Ness Sjobak wrote: > Today we use the release notes as standard homepage for all browsers, > instead of whatever is upstream default. Why do we do that, while > constantly repeating the mantra "upstream!, upstream!, upstream!"? > > In my personal opinion, the release notes belong in "help" or something > like that (even a shortcut on the standard desktop would be more > intuitive) - not as the standard homepage in every browser. > > This confuses people, and addmitedly, the release notes aren't that much > usefull to the users as the standard homepage - especially in firefox, > where the standard is: > http://www.google.no/firefox?client=firefox-a&rls=org.mozilla:nb-NO:official > which is simply a google search with some hints on using firefox etc. If a site actually deploys fedora to tend users, they certainly shouldn't leave the release notes as the default page. But the target for fedora really is the enthusiast community; while we encourage people to use it in others ways, the prototypical Fedora user is someone who: - Installs their own machine (**) - cares about the software they just installed - Has a good chance of following further links and getting involved with Fedora This is someone who is going to get a benefit from seeing the release notes (*) I don't think there is any attempt to say that the release notes are what anybody will leave as their homepage, but It might be even cooler to have some sort of "fedora portal" (I use that word with great reluctance, but I don't know a better one) as the default homepage... but that would take a lot more active resources than the release notes. Regards, Owen (*) The obvious question then is why we don't configure the default desktop for a hacker... at least add a terminal launcher. There are multiple reasons here .. *my* personal take on it is that "configured for the hacker" often leads to sloppiness where the end-user configuration is just an untested veneer with a bunch of good looking knobs. Not so much of a concern for the default homepage, which will be changed anyways. (**) If we could identify the user who installed the machine and give just *them* the release notes, that might be cool, but it's not obvious how to make that identification. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From kyrre at solution-forge.net Sat May 7 17:01:22 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sat, 07 May 2005 19:01:22 +0200 Subject: Firefox etc. default homepage In-Reply-To: <427CC3DA.5040701@redhat.com> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> <427CC3DA.5040701@redhat.com> Message-ID: <1115485281.3343.86.camel@localhost.localdomain> l?r, 07.05.2005 kl. 15.34 skrev Rahul Sundaram: > Hi > > >Definitively. Perhaps giving the release notes when people ask for them, > >together with other documentation - and not try to "push it down their > >throat"? > > > Users do not ask for information like this typically because they arent > aware that its documented at all. A promiment display on the browser > isnt equivalent to "push it down their throat" . We might replace those > by a first login wizard which guides them through instead of the current > setup but it still requires a glaring display to make sure users wont > miss out. > > " > It is, at least to some degree, "pushed down their throat", as a user who clicks on "web browser" expects that, and nothing more/less - not a big manual about how the computer they happen to open the webbrowser on ticks. Just take my mother as an example: She is a pretty unexperienced computer user/nettizen, and about the only program she ever uses is Firefox (with the apropriate plugins) on an old, beat-up laptop running Windows ME. Which she is quite happy with - she can get info on whatever she wants, in a no-fuzz way. Sometimes she wants to use my computer (because she only wants to check something really quick and doesn't care to turn on hers, etc.), wich runs fedora. No problem, she finds the webbrowser icon, and is greeted with a page full of non-interesting (to her) crap, instead of the usual google search page. Thats when she cries for help, and tells me my computer is difficult to use. Same story with new users on the system i admin at school - i set them up with a home folder and an account on the LDAP server, and let them log on. First thing they go for is the webbrowser. Secound is thing they do is staring, frightened at the display. A login-something would certainly be better than the current situation. Doesn't KDE have something like this - kandalf? Would that be possible to use (and create something similar for Gnome if it doesn't already exist) as a basis, and giving such things as links to release notes, fedoraforum.org etc.? But that would still be giving the user information the user didn't ask for. What about a shortcut on the standard desktop+somewhere in the laucher menu to the docs provided by the docs project? And how many of the RHEL docs could be copy-pasted into fedora usage? > Another reason to put it somewhere else, is to ensure it doesn't "get > lost" when the user changes his/her homepage to something more usefull..." > > good point > > > > " > > Not every user is also an admin, and the stuff in the release-notes is > primarily meant for admins." > > > I have been working on a more end user friendly version of the release > notes which can be used for non-root logins as the default page No sensefull admin i know logs in grapically as root, and if he/she does, he/she does not start surfing the web - and therefore does not start firefox. Sice firefox is never started as root, the "full" release notes never get seen. Kyrre Ness Sj?b?k From alan at redhat.com Sat May 7 17:21:17 2005 From: alan at redhat.com (Alan Cox) Date: Sat, 7 May 2005 13:21:17 -0400 Subject: Dell/Adaptec Raid Controller(aacraid) In-Reply-To: <1115398447.6722.29.camel@homebox> References: <1115398447.6722.29.camel@homebox> Message-ID: <20050507172117.GC14887@devserv.devel.redhat.com> On Fri, May 06, 2005 at 09:54:07AM -0700, Sean Bruno wrote: > How do I propose a merge between the Dell v1.1.4 and the current v1.1.5 > of the aacraid module in the 2.4 kernel? You ask Mark Salyzn at Adaptec (see MAINTAINERS). From tadams-lists at myrealbox.com Sat May 7 18:01:31 2005 From: tadams-lists at myrealbox.com (Trever L. Adams) Date: Sat, 07 May 2005 12:01:31 -0600 Subject: FC4 and kernel 2.6.12 Message-ID: <1115488891.2813.2.camel@localhost.localdomain> I am well aware we are getting late in the game. However, I have heard that real firewalling for IPv6 is in the pre series for 2.6.12. Is there anyway we can get this into FC4? This is not some trivial thing. It is something that has been needed for some time. Trever -- "God is real, unless declared integer." -- Unknown From mattdm at mattdm.org Sat May 7 18:11:08 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Sat, 7 May 2005 14:11:08 -0400 Subject: FC4 and kernel 2.6.12 In-Reply-To: <1115488891.2813.2.camel@localhost.localdomain> References: <1115488891.2813.2.camel@localhost.localdomain> Message-ID: <20050507181108.GA18485@jadzia.bu.edu> On Sat, May 07, 2005 at 12:01:31PM -0600, Trever L. Adams wrote: > I am well aware we are getting late in the game. However, I have heard > that real firewalling for IPv6 is in the pre series for 2.6.12. Is there > anyway we can get this into FC4? This is not some trivial thing. It is > something that has been needed for some time. It's not likely to be backported, but I imagine that the 2.6.12 kernel will be put into FC4 (if FC3 is any indication, at least in an update if it's not ready for the initial release). -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 74 degrees Fahrenheit. From fedora at camperquake.de Sat May 7 18:29:40 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Sat, 7 May 2005 20:29:40 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115485281.3343.86.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> <427CC3DA.5040701@redhat.com> <1115485281.3343.86.camel@localhost.localdomain> Message-ID: <20050507202940.3b33b5e8@nausicaa.camperquake.de> Hi. Kyrre Ness Sjobak wrote: > It is, at least to some degree, "pushed down their throat", as a user > who clicks on "web browser" expects that, and nothing more/less - not a > big manual about how the computer they happen to open the webbrowser on > ticks. Well, if they would actually _read_ the instructions that come with the most complex machine a normal user can buy on this planet it would certainly prevent a lot of, let's say "unnecessary" questions, now wouldn't it? Hiding them somewhere does not do anyone any good. After he has read them he is free to hide it. -- "Das Swapfile waechst, die Platte rasselt, wer hat denn da das free() vermasselt?" From mpeters at mac.com Sat May 7 18:30:35 2005 From: mpeters at mac.com (Michael A. Peters) Date: Sat, 07 May 2005 11:30:35 -0700 Subject: Standard yum configuration In-Reply-To: <1115470913.3343.47.camel@localhost.localdomain> References: <1115470913.3343.47.camel@localhost.localdomain> Message-ID: <1115490635.18558.4.camel@laptop.mpeters.local> On Sat, 2005-05-07 at 15:01 +0200, Kyrre Ness Sjobak wrote: > Some questions about the standard yum config: > - Will the fedora/fedora-extras gpgkeys be added by default, or could > the yum error message at least give some hints how to install them? > Could it even ask "%s gpgkey not installed, would you like to install > them now? [y/N]" like up2date? I think you must thrust RH anyway, as you > allready has installed the whole OS packaged and signed by them... > - Will fedora-extras.repo be included in /etc/yum.repos.d/ by default? > Will it be enabled by default? > - Will pup be installed by default? What will happen to up2date? With my install of FC4T2 - extras was enabled by default, with a gpgkey= defined so that it would fetch the key when needed. With my http rawhide install (post fc4t2) that was not the case, but I expect that because it was rawhide and not test. I'm guessing that in fc4 both extras and core will be enabled in yum, and both will import gpg key as needed. That is at least what it looks like they are planning based on a fresh (not upgrade) fc4t2 install. From katzj at redhat.com Sat May 7 18:48:53 2005 From: katzj at redhat.com (Jeremy Katz) Date: Sat, 07 May 2005 14:48:53 -0400 Subject: FC4 and kernel 2.6.12 In-Reply-To: <1115488891.2813.2.camel@localhost.localdomain> References: <1115488891.2813.2.camel@localhost.localdomain> Message-ID: <1115491733.3302.85.camel@bree.local.net> On Sat, 2005-05-07 at 12:01 -0600, Trever L. Adams wrote: > I am well aware we are getting late in the game. However, I have heard > that real firewalling for IPv6 is in the pre series for 2.6.12. Is there > anyway we can get this into FC4? This is not some trivial thing. It is > something that has been needed for some time. The current kernel for the devel tree (targeting FC4) is 2.6.12-rc3 based. We'll probably continue to move forward to the final 2.6.12 if it comes out in time for the release and if not, release it as an update later Jeremy From jspaleta at gmail.com Sat May 7 18:53:45 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Sat, 7 May 2005 14:53:45 -0400 Subject: Firefox etc. default homepage In-Reply-To: <1115485281.3343.86.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> <427CC3DA.5040701@redhat.com> <1115485281.3343.86.camel@localhost.localdomain> Message-ID: <604aa79105050711532244112@mail.gmail.com> On 5/7/05, Kyrre Ness Sjobak wrote: > It is, at least to some degree, "pushed down their throat", as a user > who clicks on "web browser" expects that, and nothing more/less - Most users using a desktop today... use IE as a browser... im pretty sure most users 'expect' to see an MSN default page.. or a page for AOL or some other ISP homepage.. if you want to get technical about expectations for 'most' computer users. Extrapolating your mother's experience with firefox.. doesn't say anything about common expectation. What you expect.... is not what I expect... and somehow I doubt either of us can be held up as an example of what the typical desktop user is thinking. I don't expect to see a google page or any other specific page. I expect to see a default that I will end up needing changing.. based on the circumstances in which that particular machine happens to be used in.. that is what I expect. At work my browser homepage ends up being an internal page from the internal webserver. At home right now its, http://darthside.blogspot.com/. The reality is most people installing fedora ARE admins of their own machines and I don't see a Fedora specific default homepage as being a particular poorer choice than a firefox default. The default page could be better in appearence and provide a link to the standard default firefox page and perhaps a brief instruction on how to choose a new homepge. But as a concept i think its perfectly acceptable to replace the default homepage with something fedora specific instead of firefox specific. The devil's in the details of what the fedora specific default should be. But I don't think it should be the firefox default... it should be a reasonably useful fedora default. If an oem does a fedora install... great...and I expect that OEM to change some default settings as they see fit for thier clients too. For large network installs, the network admins can most likely going to change the default homepage to something site specific anyways. Arguing about keeping the upstream default browser homepage seems a bit pedantic. It seems much more sensible to be arguing about what the fedora specific homepage should be including instead. -jef From jspaleta at gmail.com Sat May 7 18:58:16 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Sat, 7 May 2005 14:58:16 -0400 Subject: Firefox etc. default homepage In-Reply-To: <427CC3DA.5040701@redhat.com> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> <427CC3DA.5040701@redhat.com> Message-ID: <604aa79105050711581a820de1@mail.gmail.com> On 5/7/05, Rahul Sundaram wrote: > I have been working on a more end user friendly version of the release > notes which can be used for non-root logins as the default page any rough cuts of this in cvs or some url? and are you taking an RFE's against the design? and where would I file them? -jef From sundaram at redhat.com Sat May 7 19:12:44 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Sun, 08 May 2005 00:42:44 +0530 Subject: Firefox etc. default homepage In-Reply-To: <604aa79105050711581a820de1@mail.gmail.com> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> <427CC3DA.5040701@redhat.com> <604aa79105050711581a820de1@mail.gmail.com> Message-ID: <427D132C.30000@redhat.com> Hi >any rough cuts of this in cvs or some url? and are you taking an RFE's >against the design? and where would I file them? > >-jef > I am just volunteering to do this on my own interest and will make something available publicly after a few weeks. If anyone does have ideas please just drop me an email. I will take into consideration every idea anyone sends me and I will respond back in detail regards Rahul From mpeters at mac.com Sat May 7 19:42:53 2005 From: mpeters at mac.com (Michael A. Peters) Date: Sat, 07 May 2005 12:42:53 -0700 Subject: Firefox etc. default homepage In-Reply-To: <1115480612.3343.56.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <1115473317.6388.44.camel@laptopd505.fenrus.org> <1115480612.3343.56.camel@localhost.localdomain> Message-ID: <1115494974.18558.6.camel@laptop.mpeters.local> On Sat, 2005-05-07 at 17:43 +0200, Kyrre Ness Sjobak wrote: > But you still ship with "spatial" nautilus, and the last discussion > about it (as far as i remember) contained a lot of "upstream is > configured with spatial as default, so we must as well"... Once I got use to spatial nautilus, it's actually what I prefer. From sean.bruno at dsl-only.net Sat May 7 19:48:33 2005 From: sean.bruno at dsl-only.net (Sean Bruno) Date: Sat, 07 May 2005 12:48:33 -0700 Subject: Dell/Adaptec Raid Controller(aacraid) In-Reply-To: <20050507172117.GC14887@devserv.devel.redhat.com> References: <1115398447.6722.29.camel@homebox> <20050507172117.GC14887@devserv.devel.redhat.com> Message-ID: <1115495313.6722.86.camel@homebox> On Sat, 2005-05-07 at 13:21 -0400, Alan Cox wrote: > On Fri, May 06, 2005 at 09:54:07AM -0700, Sean Bruno wrote: > > How do I propose a merge between the Dell v1.1.4 and the current v1.1.5 > > of the aacraid module in the 2.4 kernel? > > You ask Mark Salyzn at Adaptec (see MAINTAINERS). > Thank you. At least I can whine at someone now... :) Sean From mpeters at mac.com Sat May 7 19:57:57 2005 From: mpeters at mac.com (Michael A. Peters) Date: Sat, 07 May 2005 12:57:57 -0700 Subject: Firefox etc. default homepage In-Reply-To: <1115484775.29653.21.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <1115484775.29653.21.camel@localhost.localdomain> Message-ID: <1115495877.18558.20.camel@laptop.mpeters.local> On Sat, 2005-05-07 at 12:52 -0400, Owen Taylor wrote: > > It might be even cooler to have some sort of "fedora portal" (I use that > word with great reluctance, but I don't know a better one) as the > default homepage... but that would take a lot more active resources > than the release notes. Default homepage should be a local file, not a web file. It could have a link to external, but it should be local file. Local file will work regardless of whether or not networking is up, dns is working, etc. Opening a new window is also much faster if it is from local file. Perhaps a more informative page (but a local page) with links to documentation, including the release notes, would be a better default than just the release notes. A brief page with links to: release notes (local file) FAQ (local file) /usr/share/doc user list a search box to search the user list From steve at rueb.com Sat May 7 21:41:58 2005 From: steve at rueb.com (Steve Bergman) Date: Sat, 07 May 2005 16:41:58 -0500 Subject: Firefox etc. default homepage In-Reply-To: <20050507202940.3b33b5e8@nausicaa.camperquake.de> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> <427CC3DA.5040701@redhat.com> <1115485281.3343.86.camel@localhost.localdomain> <20050507202940.3b33b5e8@nausicaa.camperquake.de> Message-ID: <427D3626.4050408@rueb.com> Ralf Ertzinger wrote: >Hi. > >Kyrre Ness Sjobak wrote: > > > >>It is, at least to some degree, "pushed down their throat", as a user >>who clicks on "web browser" expects that, and nothing more/less - not a >>big manual about how the computer they happen to open the webbrowser on >>ticks. >> >> > >Well, if they would actually _read_ the instructions that come with >the most complex machine a normal user can buy on this planet it >would certainly prevent a lot of, let's say "unnecessary" questions, >now wouldn't it? > > Well, I don't really feel strongly about all this, but it seems to me that for end user problem solving purposes, http://www.google.com/firfox would be more useful, with perhaps a bookmark to the fedora-list archives and the release notes preinstalled. And perhaps remove the numerous and confusing bookmarks that direct the user to all the different and various ways to spend money with RedHat. ;-) -Steve From hp at redhat.com Sat May 7 21:57:49 2005 From: hp at redhat.com (Havoc Pennington) Date: Sat, 07 May 2005 17:57:49 -0400 Subject: Firefox etc. default homepage In-Reply-To: <1115480612.3343.56.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <1115473317.6388.44.camel@laptopd505.fenrus.org> <1115480612.3343.56.camel@localhost.localdomain> Message-ID: <1115503069.3162.131.camel@localhost.localdomain> On Sat, 2005-05-07 at 17:43 +0200, Kyrre Ness Sjobak wrote: > l?r, 07.05.2005 kl. 15.41 skrev Arjan van de Ven: > > On Sat, 2005-05-07 at 14:32 +0200, Kyrre Ness Sjobak wrote: > > > Today we use the release notes as standard homepage for all browsers, > > > instead of whatever is upstream default. Why do we do that, while > > > constantly repeating the mantra "upstream!, upstream!, upstream!"? > > > > for me the "upstream" mantra applies far more to the code than to any > > actual settings of applications. If the user can and should change > > things, I don't see an issue with Fedora doing so as shipped as well. > > > But you still ship with "spatial" nautilus, and the last discussion > about it (as far as i remember) contained a lot of "upstream is > configured with spatial as default, so we must as well"... > It's probably relevant in this case that the upstream maintainers in question are also the Red Hat maintainers ;-) Havoc From zaitcev at redhat.com Sun May 8 02:55:21 2005 From: zaitcev at redhat.com (Pete Zaitcev) Date: Sat, 7 May 2005 19:55:21 -0700 Subject: FC4 and kernel 2.6.12 In-Reply-To: References: Message-ID: <20050507195521.1d80ee58.zaitcev@redhat.com> On Sat, 07 May 2005 12:01:31 -0600, "Trever L. Adams" wrote: > I have heard > that real firewalling for IPv6 is in the pre series for 2.6.12. [...] Sounds like rumor-mongering for me. > [] This is not some trivial thing. It is > something that has been needed for some time. My firewall works just fine. The thing is that I am, of course, ignorant about IPv6. But this "I heard ... real firewall" kind of thing is really very silly. Why don't you write specifically what functionality is missing? -- Pete From tadams-lists at myrealbox.com Sun May 8 04:22:45 2005 From: tadams-lists at myrealbox.com (Trever L. Adams) Date: Sat, 07 May 2005 22:22:45 -0600 Subject: FC4 and kernel 2.6.12 In-Reply-To: <20050507195521.1d80ee58.zaitcev@redhat.com> References: <20050507195521.1d80ee58.zaitcev@redhat.com> Message-ID: <1115526165.2813.6.camel@localhost.localdomain> Sure, you can block SYN packets and such like that. However, if it lakes state, there are still plenty of ways to cause problems. Supposedly, I believe I read it on LKM, 2.6.12 includes the netfilter/conntrack stuff that has been in the works for while. Yes, I probably should be more clear. However, in the past I have mentioned the need for this specifically and was told to wait until it was completed, well, it is, and it is being folded in. So, rumor mongering no! Assuming too much, probably. Trever On Sat, 2005-05-07 at 19:55 -0700, Pete Zaitcev wrote: > On Sat, 07 May 2005 12:01:31 -0600, "Trever L. Adams" wrote: > > > I have heard > > that real firewalling for IPv6 is in the pre series for 2.6.12. [...] > > Sounds like rumor-mongering for me. > > > [] This is not some trivial thing. It is > > something that has been needed for some time. > > My firewall works just fine. > > The thing is that I am, of course, ignorant about IPv6. But this > "I heard ... real firewall" kind of thing is really very silly. > Why don't you write specifically what functionality is missing? > > -- Pete > -- "Life is a comedy for those who think and a tragedy for those who feel." -- Unknown From dragoran at feuerpokemon.de Sun May 8 07:15:47 2005 From: dragoran at feuerpokemon.de (dragoran) Date: Sun, 08 May 2005 09:15:47 +0200 Subject: initng In-Reply-To: <200505052110.44562.gmartyn@verizon.net> References: <200505052110.44562.gmartyn@verizon.net> Message-ID: <427DBCA3.1090205@feuerpokemon.de> Greg Martyn wrote: >Hi all, > >I was just wondering if this dovetails with some of the early-login stuff that >has been discussed on this list. It was written with gentoo in mind though.. > >"Jimmy Wennlund wrote initng, a replacement for the Sys-V style "init" >application. It allows for better service dependency checking and will start >services in a highly parallel fashion, dramatically speeding up the Linux >boot process." >http://www.osnews.com/story.php?news_id=10513 > >-Greg > > > have anyone got it working under FC? it says thet it is unable to open /var/initng/in or /var/initng/out and hangs at 90%. From dragoran at feuerpokemon.de Sun May 8 10:04:06 2005 From: dragoran at feuerpokemon.de (dragoran) Date: Sun, 08 May 2005 12:04:06 +0200 Subject: initng In-Reply-To: <427DBCA3.1090205@feuerpokemon.de> References: <200505052110.44562.gmartyn@verizon.net> <427DBCA3.1090205@feuerpokemon.de> Message-ID: <427DE416.2080706@feuerpokemon.de> dragoran wrote: > Greg Martyn wrote: > >> Hi all, >> >> I was just wondering if this dovetails with some of the early-login >> stuff that has been discussed on this list. It was written with >> gentoo in mind though.. >> >> "Jimmy Wennlund wrote initng, a replacement for the Sys-V style >> "init" application. It allows for better service dependency checking >> and will start services in a highly parallel fashion, dramatically >> speeding up the Linux boot process." >> http://www.osnews.com/story.php?news_id=10513 >> >> -Greg >> >> >> > have anyone got it working under FC? > it says thet it is unable to open /var/initng/in or /var/initng/out > and hangs at 90%. > fixed this error by createing the fifos but it still hangs on 90% (system/coldplug) tryed: ng-update -d system/coldplug got this error: /sbin/ng-update: line 6: /sbin/functions.sh: No such file or directory /sbin/ng-update: line 138: einfo: command not found From kyrre at solution-forge.net Sun May 8 09:37:01 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 08 May 2005 11:37:01 +0200 Subject: Standard yum configuration In-Reply-To: <1115490635.18558.4.camel@laptop.mpeters.local> References: <1115470913.3343.47.camel@localhost.localdomain> <1115490635.18558.4.camel@laptop.mpeters.local> Message-ID: <1115545021.3343.1.camel@localhost.localdomain> l?r, 07.05.2005 kl. 20.30 skrev Michael A. Peters: > On Sat, 2005-05-07 at 15:01 +0200, Kyrre Ness Sjobak wrote: > > Some questions about the standard yum config: > > - Will the fedora/fedora-extras gpgkeys be added by default, or could > > the yum error message at least give some hints how to install them? > > Could it even ask "%s gpgkey not installed, would you like to install > > them now? [y/N]" like up2date? I think you must thrust RH anyway, as you > > allready has installed the whole OS packaged and signed by them... > > - Will fedora-extras.repo be included in /etc/yum.repos.d/ by default? > > Will it be enabled by default? > > - Will pup be installed by default? What will happen to up2date? > > With my install of FC4T2 - extras was enabled by default, with a gpgkey= > defined so that it would fetch the key when needed. > > With my http rawhide install (post fc4t2) that was not the case, but I > expect that because it was rawhide and not test. > OK that was the same kind of install i did - ftp post fc4t2. I assumed it was the same as an "updated" fc4t2. > I'm guessing that in fc4 both extras and core will be enabled in yum, > and both will import gpg key as needed. That is at least what it looks > like they are planning based on a fresh (not upgrade) fc4t2 install. But what about pup? From dwmw2 at infradead.org Sun May 8 10:13:06 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Sun, 08 May 2005 11:13:06 +0100 Subject: FC4 and kernel 2.6.12 In-Reply-To: <20050507195521.1d80ee58.zaitcev@redhat.com> References: <20050507195521.1d80ee58.zaitcev@redhat.com> Message-ID: <1115547188.29495.172.camel@localhost.localdomain> On Sat, 2005-05-07 at 19:55 -0700, Pete Zaitcev wrote: > The thing is that I am, of course, ignorant about IPv6. But this > "I heard ... real firewall" kind of thing is really very silly. > Why don't you write specifically what functionality is missing? Connection tracking. A brief look at the commit logs doesn't seem to show it being merged into 2.6.12-rc4 though. -- dwmw2 From kyrre at solution-forge.net Sun May 8 10:53:26 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 08 May 2005 12:53:26 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115469169.3343.17.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> Message-ID: <1115549606.3343.9.camel@localhost.localdomain> l?r, 07.05.2005 kl. 14.32 skrev Kyrre Ness Sjobak: > Today we use the release notes as standard homepage for all browsers, > instead of whatever is upstream default. Why do we do that, while > constantly repeating the mantra "upstream!, upstream!, upstream!"? > > In my personal opinion, the release notes belong in "help" or something > like that (even a shortcut on the standard desktop would be more > intuitive) - not as the standard homepage in every browser. > > This confuses people, and addmitedly, the release notes aren't that much > usefull to the users as the standard homepage - especially in firefox, > where the standard is: > http://www.google.no/firefox?client=firefox-a&rls=org.mozilla:nb-NO:official > which is simply a google search with some hints on using firefox etc. > > Kyrre Ness Sj?b?k Just to sum up: - Homepage should be a local file, as we cannot thrust Internet/DNS/etc. to be working, as the user migth see and error message "cannot connect to host blah blah". - Release notes should be easily accessable - Release notes should probably not be the #1 documentation shown - Documentation/release notes should be easily accessable from somewhere else, as users tend to change their homepage - Don't give users info they haven't asked for Any big disagreements? What should be done (should anything be done)? Kyrre From arjanv at redhat.com Sun May 8 11:05:31 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Sun, 08 May 2005 13:05:31 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115549606.3343.9.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <1115549606.3343.9.camel@localhost.localdomain> Message-ID: <1115550331.5998.17.camel@laptopd505.fenrus.org> On Sun, 2005-05-08 at 12:53 +0200, Kyrre Ness Sjobak wrote: > - Homepage should be a local file, as we cannot thrust Internet/DNS/etc. > to be working, as the user migth see and error message "cannot connect > to host blah blah". > - Release notes should be easily accessable > - Release notes should probably not be the #1 documentation shown > - Documentation/release notes should be easily accessable from somewhere > else, as users tend to change their homepage > - Don't give users info they haven't asked for > > Any big disagreements? What should be done (should anything be done)? if there was a very nice "welcome to fedora, click on any of these links ..." where the links are about things like the release notes, getting started with fedora desktop, setting up internet access with fedora and maybe 2 or 3 more, it'd be mightily cool :) To keep it comprehensive I suspect 5 or 6 is the hard limit on the number of these links. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From mike at netlyncs.com Sun May 8 11:35:18 2005 From: mike at netlyncs.com (Mike Chambers) Date: Sun, 08 May 2005 06:35:18 -0500 Subject: Firefox etc. default homepage In-Reply-To: <1115550331.5998.17.camel@laptopd505.fenrus.org> References: <1115469169.3343.17.camel@localhost.localdomain> <1115549606.3343.9.camel@localhost.localdomain> <1115550331.5998.17.camel@laptopd505.fenrus.org> Message-ID: <1115552118.23379.5.camel@scrappy.netlyncs.com> On Sun, 2005-05-08 at 13:05 +0200, Arjan van de Ven wrote: > where the links are about things like the release notes, getting started > with fedora desktop, setting up internet access with fedora and maybe 2 > or 3 more, it'd be mightily cool :) > To keep it comprehensive I suspect 5 or 6 is the hard limit on the > number of these links. I would agree with Arjan, in that the home page being a Fedora Core home page that is local. It could consist of things Arjan mentioned above, or maybe even link or two to Fedora Extras/Fedora Home/RedHat, etc.. -- Mike Chambers Madisonville, KY "I didn't lose my mind...I sold it on eBay!" -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From buildsys at redhat.com Sun May 8 12:07:01 2005 From: buildsys at redhat.com (Build System) Date: Sun, 8 May 2005 08:07:01 -0400 Subject: rawhide report: 20050508 changes Message-ID: <200505081207.j48C71ux008491@porkchop.devel.redhat.com> Updated Packages: From toshio at tiki-lounge.com Sun May 8 12:58:39 2005 From: toshio at tiki-lounge.com (Toshio Kuratomi) Date: Sun, 8 May 2005 05:58:39 -0700 Subject: Firefox etc. default homepage In-Reply-To: <1115552118.23379.5.camel@scrappy.netlyncs.com>; from mike@netlyncs.com on Sun, May 08, 2005 at 06:35:18AM -0500 References: <1115469169.3343.17.camel@localhost.localdomain> <1115549606.3343.9.camel@localhost.localdomain> <1115550331.5998.17.camel@laptopd505.fenrus.org> <1115552118.23379.5.camel@scrappy.netlyncs.com> Message-ID: <20050508055839.A22324@tiki-lounge.com> On Sun, May 08, 2005 at 06:35:18AM -0500, Mike Chambers wrote: > On Sun, 2005-05-08 at 13:05 +0200, Arjan van de Ven wrote: > > > where the links are about things like the release notes, getting started > > with fedora desktop, setting up internet access with fedora and maybe 2 > > or 3 more, it'd be mightily cool :) > > To keep it comprehensive I suspect 5 or 6 is the hard limit on the > > number of these links. > > I would agree with Arjan, in that the home page being a Fedora Core home > page that is local. It could consist of things Arjan mentioned above, > or maybe even link or two to Fedora Extras/Fedora Home/RedHat, etc.. And fedoraforum.org as it's become a blessed channel for end-user information. -Toshio From tadams-lists at myrealbox.com Sun May 8 16:17:19 2005 From: tadams-lists at myrealbox.com (Trever L. Adams) Date: Sun, 08 May 2005 10:17:19 -0600 Subject: FC4 and kernel 2.6.12 In-Reply-To: <1115547188.29495.172.camel@localhost.localdomain> References: <20050507195521.1d80ee58.zaitcev@redhat.com> <1115547188.29495.172.camel@localhost.localdomain> Message-ID: <1115569039.2815.1.camel@localhost.localdomain> Since Linus has moved to git (no short logs), I haven't followed very closely. But what I had read was it was supposed to already have been merged (the person was unsure of rc2 or rc3). Trever On Sun, 2005-05-08 at 11:13 +0100, David Woodhouse wrote: > On Sat, 2005-05-07 at 19:55 -0700, Pete Zaitcev wrote: > > The thing is that I am, of course, ignorant about IPv6. But this > > "I heard ... real firewall" kind of thing is really very silly. > > Why don't you write specifically what functionality is missing? > > Connection tracking. > > A brief look at the commit logs doesn't seem to show it being merged > into 2.6.12-rc4 though. > > -- > dwmw2 > -- "The best we can hope for concerning the people at large is that they be properly armed." -- Alexander Hamilton, The Federalist Papers at 184-188 From kwade at redhat.com Sun May 8 19:07:52 2005 From: kwade at redhat.com (Karsten Wade) Date: Sun, 08 May 2005 12:07:52 -0700 Subject: Firefox etc. default homepage In-Reply-To: <427D132C.30000@redhat.com> References: <1115469169.3343.17.camel@localhost.localdomain> <427CB650.8020508@redhat.com> <1115471168.3343.51.camel@localhost.localdomain> <427CC3DA.5040701@redhat.com> <604aa79105050711581a820de1@mail.gmail.com> <427D132C.30000@redhat.com> Message-ID: <1115579272.11170.3.camel@erato.phig.org> On Sun, 2005-05-08 at 00:42 +0530, Rahul Sundaram wrote: > Hi > > >any rough cuts of this in cvs or some url? and are you taking an RFE's > >against the design? and where would I file them? > > > >-jef > > > > I am just volunteering to do this on my own interest and will make > something available publicly after a few weeks. If anyone does have > ideas please just drop me an email. I will take into consideration every > idea anyone sends me and I will respond back in detail Rahul has discussed this with me as a parallel to the standard relnotes. Regardless of what we consider the target niche for Fedora is, there are _plenty_ of newbies who are using Fedora, thank ye gods. Our current relnotes are oriented for the admin, and that is a very important role they fill. Rahul's idea fills a similar role for the "typical end- user". As *both* an admin and a end-user, I would use both relnotes for their different purposes. I think Rahul's idea would either make a dandy default homepage, or a prominent top link on a link-aggregator default homepage. - Karsten -- Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/ gpg fingerprint: 2680 DBFD D968 3141 0115 5F1B D992 0E06 AD0E 0C41 Red Hat SELinux Guide http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/selinux-guide/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From seandarcy2 at gmail.com Sun May 8 19:33:31 2005 From: seandarcy2 at gmail.com (sean) Date: Sun, 08 May 2005 15:33:31 -0400 Subject: 2.6.11-1.1287_FC4 kernel panic on intel 915 Message-ID: I've installed kernel-smp-2.6.11-1.1287_FC4 on a Gigabyte GA-8i915G Pro ( with has ICH6 ) with a pentium 4 2.8g HT chip. It has a 300g maxtor SATA drive ( only drive ). It boots, but then kernel panic mounting root. It looks alright until: SCSI subsystem initialized Loading libata.ko module Loading ata_piix.ko module Loading dm-mod.ko module ata_piix: Unknown symbol ata_scsi_ioctl ata_piix: Unknown symbol ata_std_bios_param ata_piix: Unknown symbol ata_tf_read ata_piix: Unknown symbol ata_tf_write ata_piix: Unknown symbol ata_bmdma_start ata_piix: Unknown symbol ata_bmdma_setup ata_piix: Unknown symbol ata_bmdma_stop [...........] ata_piix: Unknown symbol ata_port_start insmod: error inserting '/lib/ata_piix.ko': -1 Unknown symbol in module and then all sorts of bad stuff happens. I did go get 2.6.12-rc4 today. Built it with ata_piix compiled in ( not as a module ). That worked. Booted. Mounted root. However, same kernel panic if ahci.ko was also compiled in. I also got a kernel panic when ata_piix was a module and I removed ahci.ko from modules. sean From lists at sapience.com Sun May 8 20:37:42 2005 From: lists at sapience.com (Mail Lists) Date: Sun, 8 May 2005 16:37:42 -0400 Subject: kernel-smp-2.6.11-1.1287_FC4 oops Message-ID: <20050508203741.GA5334@sapience.com> Filed here: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156664 Kernel-smp 2.6.11-1.1287 Crashes on boot: (non-smp is fine) Last message on screen before crash: Mounting root filesystem kjournald starting. Commit interval 5 seconds EXT3-fs: mounted filesystems with ordered data mode Switching to new root. unmounting old /proc unmounting old /sys This has happened for all kernels after 1268 starting with 1275. At this point it is hung - no further ourput on console. Alt-Sysrq-P shows: Hand transcribed, digital photo available with bugzilla entry. Pid: 1, comm: init EIP: 0060:[c03088f5>] CPU: 0 EPI is at _spin_unlow_irqrestore+0x1b/0x30 EFLAGS 00000296 Not tainted ... .. ... force_sig_info+0x5b/0xa7 ... do_page_fault+0x359/0x6a7 ... schedule+0x405/0xc5e ... schedule+0x431/0xc5e ... scheduler_tick+0x23b/0x414 ... do_page_fault+0x0/0x6a7 ... error_code+0x4f/0x54 Machine is 1 GiB mem, HT 3.6 GHz with SATA disk (intel ICH6R/ICH6RW) - dell precision 370 (intel based). Following is /proc/cpuinfo (lspci is in bugzilla) Let me know if there's anything further I can provide to help. g/ ------------------------------------------------------------ # cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 4 model name : Intel(R) Pentium(R) 4 CPU 3.60GHz stepping : 1 cpu MHz : 3591.578 cache size : 1024 KB physical id : 0 siblings : 2 core id : 0 cpu cores : 1 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 3 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmovpat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm pni monitor ds_cpl tm2 cid cx16 xtpr bogomips : 7127.04 -- fedora-test-list mailing list fedora-test-list at redhat.com To unsubscribe: http://www.redhat.com/mailman/listinfo/fedora-test-list From kyrre at solution-forge.net Sun May 8 20:49:10 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 08 May 2005 22:49:10 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115550331.5998.17.camel@laptopd505.fenrus.org> References: <1115469169.3343.17.camel@localhost.localdomain> <1115549606.3343.9.camel@localhost.localdomain> <1115550331.5998.17.camel@laptopd505.fenrus.org> Message-ID: <1115585350.3350.5.camel@localhost.localdomain> s?n, 08.05.2005 kl. 13.05 skrev Arjan van de Ven: > On Sun, 2005-05-08 at 12:53 +0200, Kyrre Ness Sjobak wrote: > > > - Homepage should be a local file, as we cannot thrust Internet/DNS/etc. > > to be working, as the user migth see and error message "cannot connect > > to host blah blah". > > - Release notes should be easily accessable > > - Release notes should probably not be the #1 documentation shown > > - Documentation/release notes should be easily accessable from somewhere > > else, as users tend to change their homepage > > - Don't give users info they haven't asked for > > > > Any big disagreements? What should be done (should anything be done)? > > if there was a very nice "welcome to fedora, click on any of these > links ..." > > where the links are about things like the release notes, getting started > with fedora desktop, setting up internet access with fedora and maybe 2 > or 3 more, it'd be mightily cool :) > To keep it comprehensive I suspect 5 or 6 is the hard limit on the > number of these links. Or simply "fitting it nicely in within a 800x600 display"? Anybody with web-designer-skills? I could even be localized... Giving some hints about "how to use this thing" would be great :) Giving some hints to what the release-notes actually is, would probably also be a good idea... Setting up internet is a no-job in the case of ethernet. Kyrre Ness Sj?b?k From stuart at elsn.org Mon May 9 10:23:50 2005 From: stuart at elsn.org (Stuart Ellis) Date: Mon, 09 May 2005 11:23:50 +0100 Subject: Firefox etc. default homepage In-Reply-To: <1115585350.3350.5.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <1115549606.3343.9.camel@localhost.localdomain> <1115550331.5998.17.camel@laptopd505.fenrus.org> <1115585350.3350.5.camel@localhost.localdomain> Message-ID: <1115634230.3293.233668872@webmail.messagingengine.com> On Sun, 08 May 2005 22:49:10 +0200, "Kyrre Ness Sjobak" said: > Giving some hints about "how to use this thing" would be great :) I've actually been building a document to tackle this (fedora-installation-guide/next-steps.xml in Docs CVS, but it will probably become a separate document), so there will be material that the home page could link to in there. E.g. there's a "Help on the Web" section that links to the forums. > Setting up internet is a no-job in the case of ethernet. True. I think that the post-install connectivity issues are mainly modems, proxy servers and wireless. -- Stuart Ellis stuart at elsn.org Fedora Documentation Project: http://fedora.redhat.com/projects/docs/ GPG key ID: 7098ABEA GPG key fingerprint: 68B0 E291 FB19 C845 E60E 9569 292E E365 7098 ABEA From ph18 at cornell.edu Mon May 9 13:43:37 2005 From: ph18 at cornell.edu (Paul A. Houle) Date: Mon, 09 May 2005 09:43:37 -0400 Subject: Dell/Adaptec Raid Controller(aacraid) In-Reply-To: <5256d0b0505060958156808@mail.gmail.com> References: <1115398447.6722.29.camel@homebox> <5256d0b0505060958156808@mail.gmail.com> Message-ID: On Fri, 6 May 2005 17:58:24 +0100, Peter Robinson wrote: > If your running RH AS this is an issue for those mailing lists or > RedHat support. This is a list for Fedora development. Fedora doesn't > even support the 2.4 kernel anymore. > > Peter > Well, after buying RedHat through a major computer vendor, I was quite suprised to find that Red Hat support wouldn't talk to me -- rather I'd have to get a difficult heisenbug fixed by a bunch of support people who didn't have a clue and who returned phone calls and emails about 30% of the time. Given those kinds of experiences with a product that costs > $1000, it's just a fact that some people are going to try every unconventional channel to get support possible. We're installing RHEL 4 AS on a production server that just got racked. With the experiences we've had with RHEL 3 reliability and support, we wouldn't even consider RHEL 4 if we didn't already have a site license. From russell at coker.com.au Mon May 9 15:42:52 2005 From: russell at coker.com.au (Russell Coker) Date: Tue, 10 May 2005 01:42:52 +1000 Subject: /usr/libexec Message-ID: <200505100142.55528.russell@coker.com.au> http://lists.debian.org/debian-devel/2005/05/msg00401.html I started a discussion on debian-devel about /usr/libexec with the hope of getting more compatability between different distributions (different file locations are a pain for SE Linux developers). The conclusion of the discussion at the above URL is that /usr/libexec (as used in Fedora) is against the FHS and that we should get the FHS changed if we want other distributions to follow our example in this regard. What's the procedure for submitting a suggestion for a FHS change? -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From rrelyea at redhat.com Mon May 9 16:09:12 2005 From: rrelyea at redhat.com (Bob Relyea) Date: Mon, 09 May 2005 09:09:12 -0700 Subject: Firefox etc. default homepage In-Reply-To: <1115473317.6388.44.camel@laptopd505.fenrus.org> References: <1115469169.3343.17.camel@localhost.localdomain> <1115473317.6388.44.camel@laptopd505.fenrus.org> Message-ID: <427F8B28.5000502@redhat.com> Arjan van de Ven wrote: >On Sat, 2005-05-07 at 14:32 +0200, Kyrre Ness Sjobak wrote: > > >>Today we use the release notes as standard homepage for all browsers, >>instead of whatever is upstream default. Why do we do that, while >>constantly repeating the mantra "upstream!, upstream!, upstream!"? >> >> > >for me the "upstream" mantra applies far more to the code than to any >actual settings of applications. If the user can and should change >things, I don't see an issue with Fedora doing so as shipped as well. > > > Not only that, but certain settings are designed by the upstream to be changed before any redistribution. Fedora as a package ships icons that people are required to change because of trademark issues, for instance. With Firefox, the home page default is one of the most elastic settings of a distribution. From the days before Firefox was open source (e.i. when it was called Netscape), changing the home page and the throbber icon were the 2 most common changes made by other vendors when they redistributed the code. It's designed to be changed by the redistributor, and in fact almost always is (I know of few people who redistribute Firefox or Mozilla that don't change the home page). So the "upstream!" argument doesn't hold (BTW the Firefox default is not Google, either, so you mother would still be lost because she started up in the 'download the latest version of Firefox' page). That doesn't mean the argument that the current default is silly, ineffective isn't valid, and I'm sure that topic can keep this thread going for quite some time...;). bob -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 3340 bytes Desc: S/MIME Cryptographic Signature URL: From kyrre at solution-forge.net Mon May 9 16:21:36 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Mon, 09 May 2005 18:21:36 +0200 Subject: Firefox etc. default homepage In-Reply-To: <1115634230.3293.233668872@webmail.messagingengine.com> References: <1115469169.3343.17.camel@localhost.localdomain> <1115549606.3343.9.camel@localhost.localdomain> <1115550331.5998.17.camel@laptopd505.fenrus.org> <1115585350.3350.5.camel@localhost.localdomain> <1115634230.3293.233668872@webmail.messagingengine.com> Message-ID: <1115655696.3344.2.camel@localhost.localdomain> man, 09.05.2005 kl. 12.23 skrev Stuart Ellis: > On Sun, 08 May 2005 22:49:10 +0200, "Kyrre Ness Sjobak" > said: > > > Giving some hints about "how to use this thing" would be great :) > > I've actually been building a document to tackle this > (fedora-installation-guide/next-steps.xml in Docs CVS, but it will > probably become a separate document), so there will be material that the > home page could link to in there. E.g. there's a "Help on the Web" > section that links to the forums. > > > Setting up internet is a no-job in the case of ethernet. > > True. I think that the post-install connectivity issues are mainly > modems, proxy servers and wireless. Would that include wireless cards for which drivers are not included in fedora? RPMS @ extras for those extra kernel modules? NetworkManager? What about fedorafaq.org - it is a highly usefull resource, and at least somewhat semi-official (as they get to use the fedora-announce-list)? Kyrre From sundaram at redhat.com Mon May 9 16:28:09 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Mon, 09 May 2005 21:58:09 +0530 Subject: Firefox etc. default homepage In-Reply-To: <1115655696.3344.2.camel@localhost.localdomain> References: <1115469169.3343.17.camel@localhost.localdomain> <1115549606.3343.9.camel@localhost.localdomain> <1115550331.5998.17.camel@laptopd505.fenrus.org> <1115585350.3350.5.camel@localhost.localdomain> <1115634230.3293.233668872@webmail.messagingengine.com> <1115655696.3344.2.camel@localhost.localdomain> Message-ID: <427F8F99.9010704@redhat.com> Hi >Would that include wireless cards for which drivers are not included in >fedora? RPMS @ extras for those extra kernel modules? NetworkManager? > >What about fedorafaq.org - it is a highly usefull resource, and at least >somewhat semi-official (as they get to use the fedora-announce-list)? > >Kyrre > I believe you should move this discussion to fedora-docs at this point. Fedorafaq.org cannot be linked for legal reasons. Read fedora-docs archive for details regards Rahul From buildsys at redhat.com Mon May 9 17:43:21 2005 From: buildsys at redhat.com (Build System) Date: Mon, 9 May 2005 13:43:21 -0400 Subject: rawhide report: 20050509 changes Message-ID: <200505091743.j49HhL6S023462@porkchop.devel.redhat.com> Updated Packages: From nalin at redhat.com Mon May 9 20:41:33 2005 From: nalin at redhat.com (Nalin Dahyabhai) Date: Mon, 9 May 2005 16:41:33 -0400 Subject: /usr/libexec In-Reply-To: <200505100142.55528.russell@coker.com.au> References: <200505100142.55528.russell@coker.com.au> Message-ID: <20050509204133.GB30278@redhat.com> On Tue, May 10, 2005 at 01:42:52AM +1000, Russell Coker wrote: > http://lists.debian.org/debian-devel/2005/05/msg00401.html > > I started a discussion on debian-devel about /usr/libexec with the hope of > getting more compatability between different distributions (different file > locations are a pain for SE Linux developers). > > The conclusion of the discussion at the above URL is that /usr/libexec (as > used in Fedora) is against the FHS and that we should get the FHS changed if > we want other distributions to follow our example in this regard. > > What's the procedure for submitting a suggestion for a FHS change? Quoting http://www.pathname.com/fhs/: The mailing list [1] is available to discuss interpretations of the FHS document and of possible future changes and additions. All proposals should be submitted as bugs on bugs.freestandards.org [2] using the "FHS" component). To be regarded seriously, proposals should include a unified patch to the sgml source (create the attachment after you've opened the bug, please do not cut-and-paste patches into comments). Whether the fact that various packages put files under /usr/libexec reflects a policy we're following for Fedora or is just a mass oversight is a different question. HTH, Nalin [1] http://lists.sourceforge.net/lists/listinfo/freestandards-fhs-discuss [2] http://bugs.freestandards.org/ From kaboom at oobleck.net Mon May 9 20:58:34 2005 From: kaboom at oobleck.net (Chris Ricker) Date: Mon, 9 May 2005 16:58:34 -0400 (EDT) Subject: /usr/libexec In-Reply-To: <20050509204133.GB30278@redhat.com> References: <200505100142.55528.russell@coker.com.au> <20050509204133.GB30278@redhat.com> Message-ID: On Mon, 9 May 2005, Nalin Dahyabhai wrote: > On Tue, May 10, 2005 at 01:42:52AM +1000, Russell Coker wrote: > > http://lists.debian.org/debian-devel/2005/05/msg00401.html > > > > I started a discussion on debian-devel about /usr/libexec with the hope of > > getting more compatability between different distributions (different file > > locations are a pain for SE Linux developers). > > > > The conclusion of the discussion at the above URL is that /usr/libexec (as > > used in Fedora) is against the FHS and that we should get the FHS changed if > > we want other distributions to follow our example in this regard. > > > > What's the procedure for submitting a suggestion for a FHS change? > > Quoting http://www.pathname.com/fhs/: > The mailing list [1] is available to discuss interpretations of the > FHS document and of possible future changes and additions. All > proposals should be submitted as bugs on bugs.freestandards.org [2] > using the "FHS" component). To be regarded seriously, proposals > should include a unified patch to the sgml source (create the > attachment after you've opened the bug, please do not cut-and-paste > patches into comments). > > Whether the fact that various packages put files under /usr/libexec > reflects a policy we're following for Fedora or is just a mass oversight > is a different question. FWIW, what little can be found about libexec in FHS discussion archives at sourceforge suggests that libexec has been deliberately excluded in the past on the grounds of not really serving a purpose.... later, chris From mandreiana.lists at gmail.com Mon May 9 22:11:55 2005 From: mandreiana.lists at gmail.com (Marius Andreiana) Date: Tue, 10 May 2005 01:11:55 +0300 Subject: esd won't allow gnome logout Message-ID: <1115676715.8779.6.camel@marte.biciclete.ro> Sometimes logging out of GNOME doesn't work (ctrl+alt+del has no effect). Regular users will hard reboot (and hopefully select linux again when grub asks what OS to boot). I've ran ps ax and started killing processes to find out which won't die, as gnome-session-save --kill was running. esd was running with parameters /usr/bin/esd -terminate -nobeeps -as 2 -spawnfd 13 and it will die only on killall -9. I don't know what starts it, I haven't enabled it in sound preferences. As a temporary solution I renamed /usr/bin/esd so it won't run again. Ever :D -- Marius Andreiana From hpa at zytor.com Tue May 10 05:02:30 2005 From: hpa at zytor.com (H. Peter Anvin) Date: Mon, 09 May 2005 22:02:30 -0700 Subject: /usr/libexec In-Reply-To: References: <200505100142.55528.russell@coker.com.au> <20050509204133.GB30278@redhat.com> Message-ID: <42804066.90508@zytor.com> Chris Ricker wrote: > > FWIW, what little can be found about libexec in FHS discussion archives at > sourceforge suggests that libexec has been deliberately excluded in the > past on the grounds of not really serving a purpose.... > It was included in an old version of FHS at the urging of the BSD folks. It seems to have added no value, so I guess it was removed; it definitely broke the lib/lib64 bit too. -hpa From mharris at www.linux.org.uk Tue May 10 11:19:02 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Tue, 10 May 2005 07:19:02 -0400 Subject: Missing update announcements In-Reply-To: <604aa79105050607552dbf392b@mail.gmail.com> References: <6c18a4f05050509036a77b73e@mail.gmail.com> <1115311053.20494.17.camel@finch.boston.redhat.com> <6c18a4f05050509584792d76b@mail.gmail.com> <427B05BA.7060007@www.linux.org.uk> <604aa79105050607552dbf392b@mail.gmail.com> Message-ID: <428098A6.1050406@www.linux.org.uk> Jeff Spaleta wrote: > On 5/6/05, Mike A. Harris wrote: > >>Call me crazy, I don't know where I managed to get all these >>ideas from, but it all just came together in my mind >>somehow.[1] Now that there is at least one theoretical >>solution to the problem, if this solution were to be considered, >>we need the following: >> >>1) Someone to design the detailed solution >> >>2) Someone to implement the design and test it >> >>3) Put the new system in place and start using it. >> >>In #1 and #2 above, "someone" could be either a Red Hat employee >>voluntarily taking on the task on their personal time, or it >>could be a Red Hat employee taking on the task under work time, >>or it could be a person in the community taking on the task. > > > While I think outside volunteers could take a crack at implementing > and testing a prototype solution outside the RedHat fenceline, some > requirements as to specific implementation details would need to be > communicated from inside the RedHat fenceline before anyone started > working on this. Indeed. I accounted for that in my original wording, without being detailed or specific. > Especially if at the end of the day this service > needs to be migrated back onto RedHat controlled infrastructure. That's pretty much an absolute must IMHO, for security reasons, as well as infrastructural reasons, availability, etc. > If RedHat wants to avoid a php based solution for example, that should > be communicated up-front when Fedora leadership communicates the task > proposal. While I can't speak for Fedora leadership, I personally think the choice of implementation language doesn't matter much, as long as it isn't something obscure. Red Hat has a strong internal preference for python, but if we're not implementing it, I can't see us rejecting something written in perl or php if it is well written code, as both are commonly used for such purposes. My own personal preference if I were to be touching the code at all (which I probably wont, but hey..) would be perl. ;o) (I'm currently in a perl fetish mode) > If however Fedora Core leadership is okay with a helper > service that lives outside the RedHat fenceline on a community managed > server then implementation requirements probably aren't as tight as > long as Fedora leadership is confident in the ability of the community > involved in maintaining the service. I'm making an assumption here, which is not authoritative, but I believe any solution would have to sit on Red Hat internal machines, within the corporate firewall in order to satisfy the SEC, legal, and whatever other 3 letter agencies out there - at least for Fedora Core most likely. For Fedora Extras, the same system could be used however, and it might be acceptable and useful to have that on external machines perhaps? That way the code can be easily tested, updated, etc. without legal mumbo jumbo and other things getting in the way during development. Just some random thoughts anyway. From mharris at www.linux.org.uk Tue May 10 11:22:28 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Tue, 10 May 2005 07:22:28 -0400 Subject: rawhide report: 20050505 changes In-Reply-To: <200505061611.j46GBiE10506@rio.sci.ccny.cuny.edu> References: <200505051213.j45CDYBo000353@porkchop.devel.redhat.com> <1115349789.16090.8.camel@dragon.sys.intra> <427B08C1.2060608@www.linux.org.uk> <200505061611.j46GBiE10506@rio.sci.ccny.cuny.edu> Message-ID: <42809974.4060407@www.linux.org.uk> Tim Daly wrote: > Actually there is a minor reason to prefer lower case. > ISO9660 standard filenames are lowercase 8.3 format, I believe. > Thus writing an uppercase name requires one of the minor format extensions. Which is a rather moot point, as Joliet is used on Win32, RockRidge on Linux, HFS on Mac, and are pretty much universal standards, widely accepted and implemented now. I can't see any "important" usage of CD's with rpm packages on them requiring 8.3 support. I'm not even sure if our current ISO images even support 8.3 via TRANS.TBL anymore or not. While I of course agree lowercase filenames are superior for various reasons, this one isn't compelling. ;o) From buildsys at redhat.com Tue May 10 12:12:35 2005 From: buildsys at redhat.com (Build System) Date: Tue, 10 May 2005 08:12:35 -0400 Subject: rawhide report: 20050510 changes Message-ID: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> Updated Packages: GConf2-2.10.0-4 --------------- * Mon May 09 2005 Mark McLoughlin 2.10.0-4 - Update to upstream evoldap.schema which uses GNOME's OID base rather than Red Hat's OID. GFS-6.1-0.pre22.6 ----------------- GFS-kernel-2.6.11.5-20050505.133825.FC4.6 ----------------------------------------- audit-0.7.4-1 ------------- * Sun May 08 2005 Steve Grubb 0.7.4-1 - Make sure ausearch ts & te obey DST. - Code cleanups to make file system watches work correctly ccs-0.25-0.17 ------------- * Sun May 08 2005 Florian La Roche - mv /etc/init.d -> /etc/rc.d/init.d - fix Requires * Fri May 06 2005 Chris Feist - Cleanup .spec file to not include globs. chkconfig-1.3.20-1 ------------------ * Thu May 05 2005 Bill Nottingham 1.3.20-1 - fix deletion of orphaned slave links (#131496, ) cman-1.0-0.pre33.10 ------------------- * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. * Mon Dec 20 2004 Chris Feist - Rebuild with new sources. * Mon Dec 13 2004 Chris Feist - Rebuild with new sources. cman-kernel-2.6.11.3-20050425.154843.FC4.14 ------------------------------------------- control-center-1:2.10.1-5 ------------------------- * Thu May 05 2005 Ray Strode - 1:2.10.1-5 - Don't pop up accessibility dialogs under currently focused window device-mapper-multipath-0.4.4-2.3 --------------------------------- * Fri May 06 2005 Bill Nottingham - 0.4.4-2.3 - Fix last fix. * Thu May 05 2005 Alasdair Kergon - 0.4.4-2.2 - Fix last fix. dlm-1.0-0.pre21.8 ----------------- * Sun May 08 2005 Florian La Roche - fix -devel requires dlm-kernel-2.6.11.3-20050425.154843.FC4.16 ------------------------------------------ elfutils-0.107-2 ---------------- * Sun May 08 2005 Roland McGrath - 0.107-2 - fix strip -f byte-swapping bug * Sun May 08 2005 Roland McGrath - 0.107-1 - update to 0.107 - readelf: improve DWARF output format - elflint: -d option to support checking separate debuginfo files - strip: fix ET_REL debuginfo files (#156341) evolution-2.2.2-5 ----------------- * Thu May 05 2005 David Malcolm - 2.2.2-5 - added evolution-2.2.2-fix-new-mail-notify.patch to CVS * Thu May 05 2005 David Malcolm - 2.2.2-4 - Removed explicit mozilla_build_version; instead use pkg-config to determine the path to the NSS/NSPR headers. - Use a macro to express requirement on pilot-link (was 1:0.11.4, now 0.12; patches depend on this) - Re-enabled the new-mail-notify plugin (my patch to handle differing DBus versions is in the upstream tarball; but configure.in disables the plugin for dbus versions > 0.23; patched configure.in to allow arbitrary DBus versions, and run autoconf at the start of the build) (#156328) * Sat Apr 30 2005 David Malcolm - 2.2.2-3 - updated mozilla_build_version to 1.7.7 fence-1.27-16 ------------- festival-1.95-3 --------------- * Thu Apr 28 2005 - 1.95-3 - require info packages so the post does not fail - remove /usr/bin/VCLocalRule from buildroot since it is an extranious file that does not need to be installed * Wed Apr 27 2005 Miloslav Trmac - 1.95-2 - Fix build with gcc 4 (#156132) - Require /sbin/install-info for scriptlets (#155698) - Don't ship /usr/bin/VCLocalRules (#75645) firefox-0:1.0.3-5 ----------------- * Mon May 09 2005 Christopher Aillon 0:1.0.3-5 - Correctly position the IM candidate window for most locales Note: it is still incorrectly positioned for zh_TW after this fix - Add temporary workaround to not create files in the user's $HOME (#149664) gcc-4.0.0-4 ----------- * Thu May 05 2005 Jakub Jelinek 4.0.0-4 - update from CVS - PRs libgcj/21372, target/21284 - fix fold ICE (James A. Morrison, PR tree-optimization/21085) - fix weak decl merging (PR c++/20961) - optimize i386/x86-64 movmem sequences (PR target/21329) * Wed May 04 2005 Jakub Jelinek 4.0.0-3 - update from CVS - PRs middle-end/19985, bootstrap/20633, c++/15875, c++/19542, c/15698, fortran/16861, java/21022, libfortran/17992, libfortran/19568, libfortran/19595, libfortran/20005, libfortran/20092, libfortran/20131, libfortran/20661, libfortran/20744, libgcj/21136, libstdc++/21209, libstdc++/21286, rtl-optimization/21144, target/16888, target/21098, treelang/21345 - fix loop unswitching (PR rtl-optimization/21330) - fix i386 ASHIFT to MULT address canonicalization (PR target/21297) - propagate tail call info bit through builtin expanders (PR middle-end/21265) - fix VEC_SELECT in combine (PR rtl-optimization/21239) - fix i386 V4SFmode vector initialization - fix Fortran FOREACH (PR fortran/15080) gimp-2:2.2.7-1 -------------- * Mon May 09 2005 Nils Philippsen - version 2.2.7, fixes bug in SSE2 assembly for Lighten Only layer mode (#145771) and various other bugs - on x86 and x86_64, use -msse and -msse2 to accomodate newer compilers * Wed Apr 27 2005 Jeremy Katz - 2:2.2.6-2 - silence %post * Mon Apr 11 2005 Nils Philippsen - version 2.2.6 gimp-print-4.2.7-7 ------------------ * Thu May 05 2005 Bill Nottingham 4.2.7-7 - rebuild gnbd-1.0-0.pre14.4 ------------------ gnbd-kernel-2.6.11.2-20050420.133124.FC4.19 ------------------------------------------- gnome-desktop-2.10.0-3 ---------------------- * Thu May 05 2005 Ray Strode - 2.10.0-3 - Revert previous patch--it was wrong(bug 155659). * Wed May 04 2005 Ray Strode - 2.10.0-2 - run gettext initialization routines on startup (bug 155659). gnupg-1.4.1-3 ------------- * Thu May 05 2005 Nalin Dahyabhai 1.4.1-3 - fix the execstack problem correctly this time (arjanv) gstreamer-plugins-0.8.8-6 ------------------------- * Mon May 09 2005 John (J5) Palmieri - 0.8.8-6 - Added patch to fix segfault (Bug #156622) gtk2-2.6.7-3 ------------ * Sun May 08 2005 Matthias Clasen - remove debug spew gulm-1.0-0.pre29.2 ------------------ * Sun May 08 2005 Florian La Roche - mv /etc/init.d -> /etc/rc.d/init.d - fix requires: * Fri May 06 2005 Chris Feist - fixed various problems with .spec file httpd-2.0.54-8 -------------- * Mon May 09 2005 Joe Orton 2.0.54-8 - drop old "powered by Red Hat" logos iddev-1.9-21 ------------ iiimf-1:12.2-3 -------------- * Mon May 09 2005 Akira TAGOH - 1:12.2-3 - Iiimx.Xresource: added a X resource file for iiimx, including the conversionOnKeys resource to allow shift+space. (#138309) * Mon May 09 2005 Jens Petersen - 1:12.2-2 - add gimlet-desktop-dir-149226.patch to install desktop file in the right place (Mark McLoughlin, 149226) * Tue May 03 2005 Jens Petersen - 1:12.2-1 - update to final upstream 12.2 release - the following patches are now upstream: iiimgcf-multilib-r2584-155960.patch, iiimsf-shift-space-for-korean-r2592-156171.patch, htt_xbe-correct-error-message-r2594-156169.patch, xiiimp-multilib-r2585-155963.patch - no longer need to rebootstrap iiimgcf kernel-2.6.11-1.1288_FC4 ------------------------ * Fri May 06 2005 Dave Jones - Add PCI ID for new sundance driver. (#156859) libexif-0.6.12-3 ---------------- * Fri May 06 2005 Matthias Clasen - Prevent infinite recursion (#156365) mkinitrd-4.2.12-1 ----------------- * Wed May 04 2005 Peter Jones - 4.2.12-1 - Don't copy "console=" arguments from /proc/cmdline to init netpbm-10.27-2 -------------- * Mon May 09 2005 Jindrich Novy 10.27-2 - fix invalid strcmp condition in bmptopnm, typo in pnmtojpeg (David Constanzo, #157106, #157118) - proper read longs and shorts in libpm.c (David Constanzo, #157110) - fix segfault in bmptopnm caused by freeing an uninitialized pointer parted-1.6.22-2 --------------- * Thu May 05 2005 Chris Lumens 1.6.22-2 - Added upstream patch to display certain Apple_Free partitions (#154479). perl-PDL-2.4.1-10 ----------------- * Sat Apr 30 2005 Jose Pedro Oliveira - 2.4.1-10 - Bring up to date with current Fedora.Extras perl spec template. (#156482) - disable SMP flags so it actually builds perl-Text-Kakasi-1.05-13 ------------------------ * Sat Apr 30 2005 Jose Pedro Oliveira - 1.05-13 - Added missing build requirement: kakasi-dict. (#156479) - Bring up to date with current Fedora.Extras perl spec template. perl-XML-NamespaceSupport-1.08-7 -------------------------------- redhat-artwork-0.122-8 ---------------------- * Fri May 06 2005 John (J5) Palmieri 0.122-8 - OpenOffice icon names change to not reflect versions anymore Change symlinks to reflect this * Thu May 05 2005 John (J5) Palmieri 0.122-7 - Add patch that changes the ythickness of the bluecurve-menu-item style in the bluecurve theme so the menu applet does not expand the default panel size which causes launchers icons in the panel to be scaled and become blured (Bug #146980) sendmail-8.13.4-2 ----------------- * Fri May 06 2005 Thomas Woerner 8.13.4-2 - using new certificates directory /etc/pki/tls/certs setarch-1.7-3 ------------- * Mon May 09 2005 Bill Nottingham 1.7-3 - don't strip binary, fixes debuginfo generation struts11-0:1.1-1jpp_6fc ----------------------- * Mon May 09 2005 Gary Benson - 0:1.1-1jpp_6fc - Work around a classloader issue (#157205). system-config-nfs-1.3.10-1 -------------------------- * Mon May 09 2005 Nils Philippsen 1.3.10 - pick up updated translations system-config-services-0.8.25-1 ------------------------------- * Mon May 09 2005 Nils Philippsen - 0.8.25 - pick up updated translations system-config-users-1.2.38-1 ---------------------------- * Mon May 09 2005 Nils Philippsen - 1.2.38 - pick up updated translations * Fri May 06 2005 Nils Philippsen - make desktop file rebuild consistently tcp_wrappers-7.6-39 ------------------- * Fri May 06 2005 Thomas Woerner 7.6-39 - fixed sig patch (#141110). Thanks to Nikita Shulga for the patch tetex-3.0-4 ----------- * Wed May 04 2005 Jindrich Novy 3.0-4 - add envlab package (#149212) - fix usage of uninitialized variable in ttf2afm.c (#149212) * Mon Apr 25 2005 Jindrich Novy - add Requires: fonts-japanese >= 0.20050222-2 to xdvi subpackage when Japanese is enabled - switch from /usr/share/fonts/ja to /usr/share/fonts/japanese (#155880) * Mon Apr 25 2005 MATSUURA Takanori - use htt as input method for pxdvi - update pxdvi to 22.84.9 j1.22 - fix pxdvi installation - update to ptex-src-3.1.8.1, dvips-jpatch-p1.7, mensexk-2.6a - delete japanese.map, now included in dvipsk-jpatch-p1.7 udev-057-5 ---------- * Thu May 05 2005 Bill Nottingham - 057-5 - rebuild up2date-4.4.18-1 ---------------- * Mon May 09 2005 Adrian Likins 4.4.18 - fix problem with missing import xscreensaver-1:4.21-3 --------------------- * Mon May 09 2005 Ray Strode 1:4.21-3 - Use @libexecdir@/xscreensaver instead of @HACKDIR@ in default configuration file so that the path gets expanded fully (bug 156906). From fedora at camperquake.de Tue May 10 12:19:32 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Tue, 10 May 2005 14:19:32 +0200 Subject: rawhide report: 20050510 changes In-Reply-To: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> Message-ID: <20050510121932.GA20078@ryoko.camperquake.de> On Tue, May 10, 2005 at 08:12:35AM -0400, Build System wrote: > - on x86 and x86_64, use -msse and -msse2 to accomodate newer compilers Not that I want to revive the whole .i386.rpm vs. i686.rpm discussion, but... how does gcc handle these flags? Does it build alternative code for processors that do not have these features? Or does the binary not run on anything below PII or compatible? From jakub at redhat.com Tue May 10 12:26:53 2005 From: jakub at redhat.com (Jakub Jelinek) Date: Tue, 10 May 2005 08:26:53 -0400 Subject: rawhide report: 20050510 changes In-Reply-To: <20050510121932.GA20078@ryoko.camperquake.de> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> Message-ID: <20050510122653.GC17420@devserv.devel.redhat.com> On Tue, May 10, 2005 at 02:19:32PM +0200, Ralf Ertzinger wrote: > On Tue, May 10, 2005 at 08:12:35AM -0400, Build System wrote: > > > - on x86 and x86_64, use -msse and -msse2 to accomodate newer compilers > > Not that I want to revive the whole .i386.rpm vs. i686.rpm discussion, > but... how does gcc handle these flags? Does it build alternative > code for processors that do not have these features? Or does the binary > not run on anything below PII or compatible? The way it is done in current rawhide's GIMP is definitely wrong. %ifarch %ix86 x86_64 CFLAGS="%optflags -msse -msse2" CXXFLAGS="%optflags -msse -msse2" \ %endif %configure \ a) on x86-64 this is unnecessary, all x86-64's have -msse2 by default b) -msse2 implies -msse c) if you use -msse2 in CFLAGS for all files, you can't run the latest GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, and GIMP should make sure that no routine from those sources will be ever called on pre-SSE2 chips Jakub From russell at coker.com.au Tue May 10 08:53:36 2005 From: russell at coker.com.au (Russell Coker) Date: Tue, 10 May 2005 18:53:36 +1000 Subject: /usr/libexec In-Reply-To: <42804066.90508@zytor.com> References: <200505100142.55528.russell@coker.com.au> <42804066.90508@zytor.com> Message-ID: <200505101853.40814.russell@coker.com.au> On Tuesday 10 May 2005 15:02, "H. Peter Anvin" wrote: > Chris Ricker wrote: > > FWIW, what little can be found about libexec in FHS discussion archives > > at sourceforge suggests that libexec has been deliberately excluded in > > the past on the grounds of not really serving a purpose.... > > It was included in an old version of FHS at the urging of the BSD folks. > It seems to have added no value, so I guess it was removed; it > definitely broke the lib/lib64 bit too. Why do you believe that it broke the lib/lib64 bit? AFAIK no-one has ever tested Postfix with parts of it running as 32bit and parts running as 64bit. Running it in such a manner seems likely to expose the user to previously undiscovered bugs while not providing any benefit that I can determine. If you wanted Postfix to load shared objects of different word sizes in different sub-processes then you would have a challenging task to determine which Postfix program loads which shared objects. Please give me an example of a program which has sub-processes that can run with different word sizes. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From hpa at zytor.com Tue May 10 14:48:05 2005 From: hpa at zytor.com (H. Peter Anvin) Date: Tue, 10 May 2005 07:48:05 -0700 Subject: /usr/libexec In-Reply-To: <200505101853.40814.russell@coker.com.au> References: <200505100142.55528.russell@coker.com.au> <42804066.90508@zytor.com> <200505101853.40814.russell@coker.com.au> Message-ID: <4280C9A5.6080405@zytor.com> Russell Coker wrote: > On Tuesday 10 May 2005 15:02, "H. Peter Anvin" wrote: > >>Chris Ricker wrote: >> >>>FWIW, what little can be found about libexec in FHS discussion archives >>>at sourceforge suggests that libexec has been deliberately excluded in >>>the past on the grounds of not really serving a purpose.... >> >>It was included in an old version of FHS at the urging of the BSD folks. >> It seems to have added no value, so I guess it was removed; it >>definitely broke the lib/lib64 bit too. > > > Why do you believe that it broke the lib/lib64 bit? > > AFAIK no-one has ever tested Postfix with parts of it running as 32bit and > parts running as 64bit. Running it in such a manner seems likely to expose > the user to previously undiscovered bugs while not providing any benefit that > I can determine. > > If you wanted Postfix to load shared objects of different word sizes in > different sub-processes then you would have a challenging task to determine > which Postfix program loads which shared objects. > > Please give me an example of a program which has sub-processes that can run > with different word sizes. > Who the **** is talking about Postfix? That being said, my point was that there doesn't seem to be any 32/64 separation for libexec, *and* it never made sense in the first place (it was only added for bug-compatibility with BSD), so that's presumably why it was removed. -hpa From nphilipp at redhat.com Tue May 10 14:53:55 2005 From: nphilipp at redhat.com (Nils Philippsen) Date: Tue, 10 May 2005 16:53:55 +0200 Subject: rawhide report: 20050510 changes In-Reply-To: <20050510122653.GC17420@devserv.devel.redhat.com> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> Message-ID: <1115736835.3662.27.camel@wombat.tiptoe.de> On Tue, 2005-05-10 at 08:26 -0400, Jakub Jelinek wrote: > On Tue, May 10, 2005 at 02:19:32PM +0200, Ralf Ertzinger wrote: > > On Tue, May 10, 2005 at 08:12:35AM -0400, Build System wrote: > > > > > - on x86 and x86_64, use -msse and -msse2 to accomodate newer compilers > > > > Not that I want to revive the whole .i386.rpm vs. i686.rpm discussion, > > but... how does gcc handle these flags? Does it build alternative > > code for processors that do not have these features? Or does the binary > > not run on anything below PII or compatible? > > The way it is done in current rawhide's GIMP is definitely wrong. > %ifarch %ix86 x86_64 > CFLAGS="%optflags -msse -msse2" CXXFLAGS="%optflags -msse -msse2" \ > %endif > %configure \ > > a) on x86-64 this is unnecessary, all x86-64's have -msse2 by default > b) -msse2 implies -msse ACK > c) if you use -msse2 in CFLAGS for all files, you can't run the latest > GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. > -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, > and GIMP should make sure that no routine from those sources will be > ever called on pre-SSE2 chips According to the gcc man page, you need to use "-mfpmath=sse" if you want to get SSE/SSE2 instructions generated from FP code. As I understand the manpage, "-msse" and the like only _enable_ its use in inline assembler code. GIMP does check whether MMX/SSE etc. are available when starting and only uses the optimized code paths if possible. Nils -- Nils Philippsen / Red Hat / nphilipp at redhat.com "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- B. Franklin, 1759 PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011 From jakub at redhat.com Tue May 10 15:02:10 2005 From: jakub at redhat.com (Jakub Jelinek) Date: Tue, 10 May 2005 11:02:10 -0400 Subject: rawhide report: 20050510 changes In-Reply-To: <1115736835.3662.27.camel@wombat.tiptoe.de> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> <1115736835.3662.27.camel@wombat.tiptoe.de> Message-ID: <20050510150209.GD17420@devserv.devel.redhat.com> On Tue, May 10, 2005 at 04:53:55PM +0200, Nils Philippsen wrote: > > c) if you use -msse2 in CFLAGS for all files, you can't run the latest > > GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. > > -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, > > and GIMP should make sure that no routine from those sources will be > > ever called on pre-SSE2 chips > > According to the gcc man page, you need to use "-mfpmath=sse" if you > want to get SSE/SSE2 instructions generated from FP code. As I No. -msse resp. -msse2 says that SSE resp. SSE2 instructions are available and can be used. So, with -msse2 you can use SSE2 builtins, you can use those regs in __asm statement clobbers/regs and GCC if it decides it is worthwhile can use the SSE2 registers/instructions (say with autovectorization, or with high register preasure to move data around, etc.). -mfpmath=sse uses the SSE2 unit for float/double arithmetics. If the only place that you want to enable -msse2 for are the clobber lists of __asm statements, then perhaps best would be to conditionalize them: __asm ("something" : ... : ... : #ifdef __SSE2__ "xmm0", "xmm1" #endif ); etc., because without -msse2 (resp. -msse (__SSE__ macro)), the registers are not known to the compiler, so there is no point to tell the compiler about them. Jakub From nphilipp at redhat.com Tue May 10 15:08:34 2005 From: nphilipp at redhat.com (Nils Philippsen) Date: Tue, 10 May 2005 17:08:34 +0200 Subject: rawhide report: 20050510 changes In-Reply-To: <20050510150209.GD17420@devserv.devel.redhat.com> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> <1115736835.3662.27.camel@wombat.tiptoe.de> <20050510150209.GD17420@devserv.devel.redhat.com> Message-ID: <1115737715.3662.31.camel@wombat.tiptoe.de> On Tue, 2005-05-10 at 11:02 -0400, Jakub Jelinek wrote: > On Tue, May 10, 2005 at 04:53:55PM +0200, Nils Philippsen wrote: > > > c) if you use -msse2 in CFLAGS for all files, you can't run the latest > > > GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. > > > -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, > > > and GIMP should make sure that no routine from those sources will be > > > ever called on pre-SSE2 chips > > > > According to the gcc man page, you need to use "-mfpmath=sse" if you > > want to get SSE/SSE2 instructions generated from FP code. As I > > No. -msse resp. -msse2 says that SSE resp. SSE2 instructions are available > and can be used. So, with -msse2 you can use SSE2 builtins, you can > use those regs in __asm statement clobbers/regs and GCC if it decides > it is worthwhile can use the SSE2 registers/instructions (say with > autovectorization, or with high register preasure to move data around, > etc.). > -mfpmath=sse uses the SSE2 unit for float/double arithmetics. > > If the only place that you want to enable -msse2 for are the clobber > lists of __asm statements, then perhaps best would be to conditionalize > them: > __asm ("something" : ... : ... : > #ifdef __SSE2__ > "xmm0", "xmm1" > #endif > ); > etc., because without -msse2 (resp. -msse (__SSE__ macro)), the registers > are not known to the compiler, so there is no point to tell the compiler > about them. Ahh. If I understand you correctly, given that the various asm optimized functions are in separate source files, it should be ensured that -mmmx/sse/sse2/3dnow are used only for the relevant source files? Nils -- Nils Philippsen / Red Hat / nphilipp at redhat.com "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- B. Franklin, 1759 PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011 From jakub at redhat.com Tue May 10 15:11:50 2005 From: jakub at redhat.com (Jakub Jelinek) Date: Tue, 10 May 2005 11:11:50 -0400 Subject: rawhide report: 20050510 changes In-Reply-To: <1115737715.3662.31.camel@wombat.tiptoe.de> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> <1115736835.3662.27.camel@wombat.tiptoe.de> <20050510150209.GD17420@devserv.devel.redhat.com> <1115737715.3662.31.camel@wombat.tiptoe.de> Message-ID: <20050510151150.GE17420@devserv.devel.redhat.com> On Tue, May 10, 2005 at 05:08:34PM +0200, Nils Philippsen wrote: > > If the only place that you want to enable -msse2 for are the clobber > > lists of __asm statements, then perhaps best would be to conditionalize > > them: > > __asm ("something" : ... : ... : > > #ifdef __SSE2__ > > "xmm0", "xmm1" > > #endif > > ); > > etc., because without -msse2 (resp. -msse (__SSE__ macro)), the registers > > are not known to the compiler, so there is no point to tell the compiler > > about them. > > Ahh. If I understand you correctly, given that the various asm optimized > functions are in separate source files, it should be ensured that > -mmmx/sse/sse2/3dnow are used only for the relevant source files? Yes (and assuming they don't also contain code that is called unconditionally). Or don't pass -mmmx/-msse/-msse2/-m3dnow etc. at all, and hide all the SSE2 etc. stuff in __asm and conditionally don't expose the clobbers to GCC. Jakub From nphilipp at redhat.com Tue May 10 15:17:20 2005 From: nphilipp at redhat.com (Nils Philippsen) Date: Tue, 10 May 2005 17:17:20 +0200 Subject: rawhide report: 20050510 changes In-Reply-To: <20050510151150.GE17420@devserv.devel.redhat.com> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> <1115736835.3662.27.camel@wombat.tiptoe.de> <20050510150209.GD17420@devserv.devel.redhat.com> <1115737715.3662.31.camel@wombat.tiptoe.de> <20050510151150.GE17420@devserv.devel.redhat.com> Message-ID: <1115738240.3662.36.camel@wombat.tiptoe.de> On Tue, 2005-05-10 at 11:11 -0400, Jakub Jelinek wrote: > On Tue, May 10, 2005 at 05:08:34PM +0200, Nils Philippsen wrote: > > > If the only place that you want to enable -msse2 for are the clobber > > > lists of __asm statements, then perhaps best would be to conditionalize > > > them: > > > __asm ("something" : ... : ... : > > > #ifdef __SSE2__ > > > "xmm0", "xmm1" > > > #endif > > > ); > > > etc., because without -msse2 (resp. -msse (__SSE__ macro)), the registers > > > are not known to the compiler, so there is no point to tell the compiler > > > about them. > > > > Ahh. If I understand you correctly, given that the various asm optimized > > functions are in separate source files, it should be ensured that > > -mmmx/sse/sse2/3dnow are used only for the relevant source files? > > Yes (and assuming they don't also contain code that is called > unconditionally). I expect that to be the case (the code is in files called gimp- composite-{generic,mmx,altivec,sse,sse2,...}). > Or don't pass -mmmx/-msse/-msse2/-m3dnow etc. at all, and hide all the > SSE2 etc. stuff in __asm and conditionally don't expose the clobbers > to GCC. The idea is to use the most optimized code path possible from one binary. Besides, I don't really think we want to have i686,pentiumX,athlon variants of GIMP packages ;-). Nils -- Nils Philippsen / Red Hat / nphilipp at redhat.com "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- B. Franklin, 1759 PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011 From otaylor at redhat.com Tue May 10 15:30:06 2005 From: otaylor at redhat.com (Owen Taylor) Date: Tue, 10 May 2005 11:30:06 -0400 Subject: rawhide report: 20050510 changes In-Reply-To: <20050510122653.GC17420@devserv.devel.redhat.com> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> Message-ID: <1115739006.30474.94.camel@localhost.localdomain> On Tue, 2005-05-10 at 08:26 -0400, Jakub Jelinek wrote: > a) on x86-64 this is unnecessary, all x86-64's have -msse2 by default > b) -msse2 implies -msse > c) if you use -msse2 in CFLAGS for all files, you can't run the latest > GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. > -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, > and GIMP should make sure that no routine from those sources will be > ever called on pre-SSE2 chips This sounds like a big pain ... shouldn't there be a way to say "use sse2/sse only for builtins" ? Without that, there is no way to use builtins in single functions; you need to have one file for sse2, one for sse, one for mmx, and do the detection of the current processor somewhere else entirely. Owen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From russell at coker.com.au Tue May 10 15:33:10 2005 From: russell at coker.com.au (Russell Coker) Date: Wed, 11 May 2005 01:33:10 +1000 Subject: /usr/libexec In-Reply-To: <4280C9A5.6080405@zytor.com> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> Message-ID: <200505110133.14911.russell@coker.com.au> On Wednesday 11 May 2005 00:48, "H. Peter Anvin" wrote: > > >It was included in an old version of FHS at the urging of the BSD folks. > > > ?It seems to have added no value, so I guess it was removed; it > > >definitely broke the lib/lib64 bit too. > > > > Why do you believe that it broke the lib/lib64 bit? > > > > AFAIK no-one has ever tested Postfix with parts of it running as 32bit > > and parts running as 64bit. Running it in such a manner seems likely to > > expose the user to previously undiscovered bugs while not providing any > > benefit that I can determine. > > > > If you wanted Postfix to load shared objects of different word sizes in > > different sub-processes then you would have a challenging task to > > determine which Postfix program loads which shared objects. > > > > Please give me an example of a program which has sub-processes that can > > run with different word sizes. > > Who the **** is talking about Postfix? Almost everyone involved in this discussion. This entire discussion started on the SE Linux list concerning the details of Postfix policy. I then started a discussion on the debian-devel list about trying to get some convergence between the different distributions for file locations for programs such as Postfix. After the FHS issue was raised there I started a discussion here. You apparently didn't visit the URL I cited for the discussion. If you think that there's a better example of an application that uses libexec which we can use for this discussion then please let me know. > That being said, my point was that there doesn't seem to be any 32/64 > separation for libexec, *and* it never made sense in the first place (it > was only added for bug-compatibility with BSD), so that's presumably why > it was removed. If we have /usr/lib for 32bit SOs and /usr/lib64 for 64bit SOs, then it seems clear to me that programs which are part of Postfix which may be either 32bit or 64bit depending on which package is installed belong to neither category. Therefore another place such as /usr/libexec seems appropriate. The Gentoo people want /usr/lib/postfix for 32bit compiles and /usr/lib64/postfix on 64bit compiles. I believe that approach is totally wrong and that /usr/libexec/postfix (as used in Fedora) is the better option. If there's general agreement with that then we can move of requesting that the FHS be changed to make the Red Hat practice be a standard in this regard. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From hpa at zytor.com Tue May 10 16:01:37 2005 From: hpa at zytor.com (H. Peter Anvin) Date: Tue, 10 May 2005 09:01:37 -0700 Subject: /usr/libexec In-Reply-To: <200505110133.14911.russell@coker.com.au> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> <200505110133.14911.russell@coker.com.au> Message-ID: <4280DAE1.2070107@zytor.com> Russell Coker wrote: > >>That being said, my point was that there doesn't seem to be any 32/64 >>separation for libexec, *and* it never made sense in the first place (it >>was only added for bug-compatibility with BSD), so that's presumably why >>it was removed. > > If we have /usr/lib for 32bit SOs and /usr/lib64 for 64bit SOs, then it seems > clear to me that programs which are part of Postfix which may be either 32bit > or 64bit depending on which package is installed belong to neither category. > Therefore another place such as /usr/libexec seems appropriate. > > The Gentoo people want /usr/lib/postfix for 32bit compiles > and /usr/lib64/postfix on 64bit compiles. I believe that approach is totally > wrong and that /usr/libexec/postfix (as used in Fedora) is the better option. > If there's general agreement with that then we can move of requesting that > the FHS be changed to make the Red Hat practice be a standard in this regard. > I think you're completely wrong. Look at it this way: may not be sterilized for its internal interfaces, in respect of being cross-archictecture clean. If you put it in libexec, then for exactly the reasons you mention you *HAVE* to handle mixed-mode. Thus, although enforcing the separation may be *redundant*, it definitely not *harmful*, and might be beneficial. For stuff like shell scripts, that are inherently cross-platform, that's *exactly* what the share hierarchy is defined to be. Either which way, libexec is useless. -hpa From notting at redhat.com Tue May 10 16:02:27 2005 From: notting at redhat.com (Bill Nottingham) Date: Tue, 10 May 2005 12:02:27 -0400 Subject: /usr/libexec In-Reply-To: <200505110133.14911.russell@coker.com.au> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> <200505110133.14911.russell@coker.com.au> Message-ID: <20050510160227.GA3820@nostromo.devel.redhat.com> Russell Coker (russell at coker.com.au) said: > The Gentoo people want /usr/lib/postfix for 32bit compiles > and /usr/lib64/postfix on 64bit compiles. I believe that approach is totally > wrong and that /usr/libexec/postfix (as used in Fedora) is the better option. > If there's general agreement with that then we can move of requesting that > the FHS be changed to make the Red Hat practice be a standard in this regard. If the support programs need to be the same wordsize as the app/libraries in question, it should be in /usr/{lib,lib64}/. If they don't care, it can be in /usr/libexec. For example, xscreensaver has screensaver hacks in /usr/libexec/xscreensaver - these don't care about the wordsize at all (in fact, they can be shell/python/etc). So, /usr/libexec seems appropriate here. As a counter example of something broken, there is /usr/libexec/sudo_noexec.so, which really should be in /usr/{lib,lib64}/sudo. Bill From mls at suse.de Tue May 10 16:12:45 2005 From: mls at suse.de (Michael Schroeder) Date: Tue, 10 May 2005 18:12:45 +0200 Subject: /usr/libexec In-Reply-To: <20050510160227.GA3820@nostromo.devel.redhat.com> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> <200505110133.14911.russell@coker.com.au> <20050510160227.GA3820@nostromo.devel.redhat.com> Message-ID: <20050510161245.GA20358@suse.de> On Tue, May 10, 2005 at 12:02:27PM -0400, Bill Nottingham wrote: > If the support programs need to be the same wordsize as the app/libraries > in question, it should be in /usr/{lib,lib64}/. > > If they don't care, it can be in /usr/libexec. > > For example, xscreensaver has screensaver hacks in > /usr/libexec/xscreensaver - these don't care about the wordsize at all > (in fact, they can be shell/python/etc). So, /usr/libexec seems > appropriate here. Why not /usr/share/xscreensaver if they are arch independent? Cheers, Michael. -- Michael Schroeder mls at suse.de main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} From notting at redhat.com Tue May 10 16:28:28 2005 From: notting at redhat.com (Bill Nottingham) Date: Tue, 10 May 2005 12:28:28 -0400 Subject: /usr/libexec In-Reply-To: <20050510161245.GA20358@suse.de> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> <200505110133.14911.russell@coker.com.au> <20050510160227.GA3820@nostromo.devel.redhat.com> <20050510161245.GA20358@suse.de> Message-ID: <20050510162828.GA32140@nostromo.devel.redhat.com> Michael Schroeder (mls at suse.de) said: > On Tue, May 10, 2005 at 12:02:27PM -0400, Bill Nottingham wrote: > > If the support programs need to be the same wordsize as the app/libraries > > in question, it should be in /usr/{lib,lib64}/. > > > > If they don't care, it can be in /usr/libexec. > > > > For example, xscreensaver has screensaver hacks in > > /usr/libexec/xscreensaver - these don't care about the wordsize at all > > (in fact, they can be shell/python/etc). So, /usr/libexec seems > > appropriate here. > > Why not /usr/share/xscreensaver if they are arch independent? /usr/share is for files that can run on any architecture. These are files that can be archtecture-specific, but don't have to be. Bill From perbj at stanford.edu Tue May 10 16:49:14 2005 From: perbj at stanford.edu (Per Bjornsson) Date: Tue, 10 May 2005 09:49:14 -0700 Subject: /usr/libexec In-Reply-To: <20050510162828.GA32140@nostromo.devel.redhat.com> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> <200505110133.14911.russell@coker.com.au> <20050510160227.GA3820@nostromo.devel.redhat.com> <20050510161245.GA20358@suse.de> <20050510162828.GA32140@nostromo.devel.redhat.com> Message-ID: <1115743754.4775.22.camel@localhost.localdomain> On Tue, 2005-05-10 at 12:28 -0400, Bill Nottingham wrote: > Michael Schroeder (mls at suse.de) said: > > Why not /usr/share/xscreensaver if they are arch independent? > > /usr/share is for files that can run on any architecture. These > are files that can be archtecture-specific, but don't have to be. Yes, but would it actually hurt (well, apart from the work of changing the packaging) to have them live in /usr/lib/xscreensaver instead of /usr/libexec/xscreensaver? It seems that the former is what would be prescribed by the FHS. Since this came up in the SELinux context, there appears to be some real value in following the FHS since it would apparently significantly simplify setting up the policy in a distribution-independent fashion. /Per -- Per Bjornsson From russell at coker.com.au Tue May 10 16:54:36 2005 From: russell at coker.com.au (Russell Coker) Date: Wed, 11 May 2005 02:54:36 +1000 Subject: /usr/libexec In-Reply-To: <4280DAE1.2070107@zytor.com> References: <200505100142.55528.russell@coker.com.au> <200505110133.14911.russell@coker.com.au> <4280DAE1.2070107@zytor.com> Message-ID: <200505110254.39321.russell@coker.com.au> On Wednesday 11 May 2005 02:01, "H. Peter Anvin" wrote: > > If we have /usr/lib for 32bit SOs and /usr/lib64 for 64bit SOs, then it > > seems clear to me that programs which are part of Postfix which may be > > either 32bit or 64bit depending on which package is installed belong to > > neither category. Therefore another place such as /usr/libexec seems > > appropriate. > > > > The Gentoo people want /usr/lib/postfix for 32bit compiles > > and /usr/lib64/postfix on 64bit compiles. I believe that approach is > > totally wrong and that /usr/libexec/postfix (as used in Fedora) is the > > better option. If there's general agreement with that then we can move of > > requesting that the FHS be changed to make the Red Hat practice be a > > standard in this regard. > > I think you're completely wrong. > > Look at it this way: may not be sterilized for its > internal interfaces, in respect of being cross-archictecture clean. If That's what I expect. > you put it in libexec, then for exactly the reasons you mention you > *HAVE* to handle mixed-mode. If you have separate directories for 32bit and 64bit then you would have matching application configuration as to which directory to look in. Therefore you don't need the internal application interfaces to be cross-architecture clean unless you bork your config files. If you have only one directory for such programs (/usr/libexec) then you only have one version of the application installed at any time and therefore you can't have cross-architecture issues. I don't understand what the problem is. I think that Bill's idea makes sense. Use /usr/lib and /usr/lib64 for anything that needs separate versions for word sizes and use /usr/libexec for anything that doesn't. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From mitr at volny.cz Tue May 10 16:54:53 2005 From: mitr at volny.cz (Miloslav Trmac) Date: Tue, 10 May 2005 18:54:53 +0200 Subject: /usr/libexec In-Reply-To: <1115743754.4775.22.camel@localhost.localdomain> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> <200505110133.14911.russell@coker.com.au> <20050510160227.GA3820@nostromo.devel.redhat.com> <20050510161245.GA20358@suse.de> <20050510162828.GA32140@nostromo.devel.redhat.com> <1115743754.4775.22.camel@localhost.localdomain> Message-ID: <20050510165448.GA7113@chrys.ms.mff.cuni.cz> On Tue, May 10, 2005 at 09:49:14AM -0700, Per Bjornsson wrote: > On Tue, 2005-05-10 at 12:28 -0400, Bill Nottingham wrote: > > Michael Schroeder (mls at suse.de) said: > > > Why not /usr/share/xscreensaver if they are arch independent? > > > > /usr/share is for files that can run on any architecture. These > > are files that can be archtecture-specific, but don't have to be. > > Yes, but would it actually hurt (well, apart from the work of changing > the packaging) to have them live in /usr/lib/xscreensaver instead > of /usr/libexec/xscreensaver? Yes, it would (a little). Assume you write an xscreensaver hack in Python, naturally resulting in a noarch package. Now the package will have to contain both /usr/lib/xscreensaver/pythonhack and /usr/lib64/xscreensaver/pythonhack. Mirek From dax at gurulabs.com Tue May 10 20:35:45 2005 From: dax at gurulabs.com (Dax Kelson) Date: Tue, 10 May 2005 14:35:45 -0600 Subject: Converting your FC4test2 isos to FC4test3 isos Message-ID: <1115757345.10816.10.camel@mentorng.gurulabs.com> I noticed that test3 has been released, and I used this trick to convert my FC4test2 isos into FC4test3. This way I'm not downloading the bits that haven't changed. cd /path/to/your/FC4test2/isos mv FC4-test2-i386-disc1.iso FC4-test3-i386-disc1.iso mv FC4-test2-i386-disc2.iso FC4-test3-i386-disc2.iso mv FC4-test2-i386-disc3.iso FC4-test3-i386-disc3.iso mv FC4-test2-i386-disc4.iso FC4-test3-i386-disc4.iso rsync -v --stats --progress rsync:///fedora/linux/core/test/3.92/i386/iso/FC4-test3-i386-disc[1234].iso . Watching the data transfer speed, I notice that pretty often it jumps up much higher than the cap of my internet connection. This tells me that that my trick was worthwhile. Be sure to grab the SHA1SUM file and check your ISOs once the rsync completes. Dax Kelson Guru Labs From hpa at zytor.com Tue May 10 21:13:23 2005 From: hpa at zytor.com (H. Peter Anvin) Date: Tue, 10 May 2005 14:13:23 -0700 Subject: /usr/libexec In-Reply-To: <200505110254.39321.russell@coker.com.au> References: <200505100142.55528.russell@coker.com.au> <200505110133.14911.russell@coker.com.au> <4280DAE1.2070107@zytor.com> <200505110254.39321.russell@coker.com.au> Message-ID: <428123F3.8080602@zytor.com> Russell Coker wrote: > > I think that Bill's idea makes sense. Use /usr/lib and /usr/lib64 for > anything that needs separate versions for word sizes and use /usr/libexec for > anything that doesn't. > So how's that, again, different from /usr/share? -hpa From mattdm at mattdm.org Tue May 10 21:17:26 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Tue, 10 May 2005 17:17:26 -0400 Subject: /usr/libexec In-Reply-To: <428123F3.8080602@zytor.com> References: <200505100142.55528.russell@coker.com.au> <200505110133.14911.russell@coker.com.au> <4280DAE1.2070107@zytor.com> <200505110254.39321.russell@coker.com.au> <428123F3.8080602@zytor.com> Message-ID: <20050510211726.GA6685@jadzia.bu.edu> On Tue, May 10, 2005 at 02:13:23PM -0700, H. Peter Anvin wrote: > >I think that Bill's idea makes sense. Use /usr/lib and /usr/lib64 for > >anything that needs separate versions for word sizes and use /usr/libexec > >for anything that doesn't. > So how's that, again, different from /usr/share? /usr/share is for data files; executable code shouldn't be there. -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 82 degrees Fahrenheit. From rhally at mindspring.com Wed May 11 06:40:54 2005 From: rhally at mindspring.com (Richard Hally) Date: Wed, 11 May 2005 02:40:54 -0400 Subject: yum very slow Message-ID: <4281A8F6.1070503@mindspring.com> Yum seems to be very slow in resolving dependencies. I'm sitting here watching my rawhide test box take over an hour to do the dep resolution for todays updates from rawhide. This is on a box that has a 2.5ghz P4 and nothing else running on the machine. this has been occurring for at least a few weeks(or longer). What should I do to get useful information concerning this problem? Thanks Richard Hally From skvidal at phy.duke.edu Wed May 11 07:15:16 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Wed, 11 May 2005 03:15:16 -0400 Subject: yum very slow In-Reply-To: <4281A8F6.1070503@mindspring.com> References: <4281A8F6.1070503@mindspring.com> Message-ID: <1115795716.10234.130.camel@cutter> On Wed, 2005-05-11 at 02:40 -0400, Richard Hally wrote: > Yum seems to be very slow in resolving dependencies. I'm sitting here > watching my rawhide test box take over an hour to do the dep resolution > for todays updates from rawhide. This is on a box that has a 2.5ghz P4 > and nothing else running on the machine. this has been occurring for at > least a few weeks(or longer). What should I do to get useful information > concerning this problem? > are you sure it's the dep resolution that is slow or is it slowing down when it does the header download DURING the dep resolution. I'm making that distinction b/c one is a function of code and processor and the other is a function of mirror availability and bandwidth. -sv From rhally at mindspring.com Wed May 11 08:41:43 2005 From: rhally at mindspring.com (Richard Hally) Date: Wed, 11 May 2005 04:41:43 -0400 Subject: yum very slow In-Reply-To: <1115795716.10234.130.camel@cutter> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> Message-ID: <4281C547.5090506@mindspring.com> seth vidal wrote: >On Wed, 2005-05-11 at 02:40 -0400, Richard Hally wrote: > > >>Yum seems to be very slow in resolving dependencies. I'm sitting here >>watching my rawhide test box take over an hour to do the dep resolution >>for todays updates from rawhide. This is on a box that has a 2.5ghz P4 >>and nothing else running on the machine. this has been occurring for at >>least a few weeks(or longer). What should I do to get useful information >>concerning this problem? >> >> >> > >are you sure it's the dep resolution that is slow or is it slowing down >when it does the header download DURING the dep resolution. I'm making >that distinction b/c one is a function of code and processor and the >other is a function of mirror availability and bandwidth. > >-sv > > > > here is what it says: Excluding Packages in global exclude list Finished (wait an hour + ) Resolving Dependencies --> Populating transaction set with selected packages.. During the wait, there is continuous disk activity. and no network activity. This has been going on for a while (it didn't just start today) I did a yum clean all to see if it would help but it did not. From twoerner at redhat.com Wed May 11 09:31:22 2005 From: twoerner at redhat.com (Thomas Woerner) Date: Wed, 11 May 2005 11:31:22 +0200 Subject: /usr/libexec In-Reply-To: <4280DAE1.2070107@zytor.com> References: <200505100142.55528.russell@coker.com.au> <200505101853.40814.russell@coker.com.au> <4280C9A5.6080405@zytor.com> <200505110133.14911.russell@coker.com.au> <4280DAE1.2070107@zytor.com> Message-ID: <4281D0EA.8050105@redhat.com> H. Peter Anvin wrote: > Russell Coker wrote: > >> >>> That being said, my point was that there doesn't seem to be any 32/64 >>> separation for libexec, *and* it never made sense in the first place (it >>> was only added for bug-compatibility with BSD), so that's presumably why >>> it was removed. >> >> >> If we have /usr/lib for 32bit SOs and /usr/lib64 for 64bit SOs, then >> it seems clear to me that programs which are part of Postfix which may >> be either 32bit or 64bit depending on which package is installed >> belong to neither category. Therefore another place such as >> /usr/libexec seems appropriate. >> >> The Gentoo people want /usr/lib/postfix for 32bit compiles and >> /usr/lib64/postfix on 64bit compiles. I believe that approach is >> totally wrong and that /usr/libexec/postfix (as used in Fedora) is the >> better option. If there's general agreement with that then we can >> move of requesting that the FHS be changed to make the Red Hat >> practice be a standard in this regard. >> > > I think you're completely wrong. > > Look at it this way: may not be sterilized for its > internal interfaces, in respect of being cross-archictecture clean. If > you put it in libexec, then for exactly the reasons you mention you > *HAVE* to handle mixed-mode. > > Thus, although enforcing the separation may be *redundant*, it > definitely not *harmful*, and might be beneficial. > > For stuff like shell scripts, that are inherently cross-platform, that's > *exactly* what the share hierarchy is defined to be. > > Either which way, libexec is useless. > > -hpa > The binaries of postfix in /usr/libexec are the same arch as the binaries in /usr/bin, because the binary with the highest architecture wins if you are installing packages with different (and allowed) architectures. Therefore there is no mixed mode within the postfix package as long as you are not putting other binaries in there by hand. Thomas -- Thomas Woerner, Software Engineer Phone: +49-711-96437-0 Red Hat GmbH Fax : +49-711-96437-111 Hauptstaetterstr. 58 Email: twoerner at redhat.com D-70178 Stuttgart Web : http://www.redhat.de/ From yuan.bbbush at gmail.com Wed May 11 10:37:14 2005 From: yuan.bbbush at gmail.com (Yuan Yijun) Date: Wed, 11 May 2005 18:37:14 +0800 Subject: yum very slow In-Reply-To: <4281C547.5090506@mindspring.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> Message-ID: <9792751e050511033744b2cb29@mail.gmail.com> I always set vorbose message level to 10 instead of default value 3 :) 2005/5/11, Richard Hally : > seth vidal wrote: > > >On Wed, 2005-05-11 at 02:40 -0400, Richard Hally wrote: > > > > > >>Yum seems to be very slow in resolving dependencies. I'm sitting here > >>watching my rawhide test box take over an hour to do the dep resolution > >>for todays updates from rawhide. This is on a box that has a 2.5ghz P4 > >>and nothing else running on the machine. this has been occurring for at > >>least a few weeks(or longer). What should I do to get useful information > >>concerning this problem? > >> > >> > >> > > > >are you sure it's the dep resolution that is slow or is it slowing down > >when it does the header download DURING the dep resolution. I'm making > >that distinction b/c one is a function of code and processor and the > >other is a function of mirror availability and bandwidth. > > > >-sv > > > > > > > > > here is what it says: > Excluding Packages in global exclude list > Finished > (wait an hour + ) > Resolving Dependencies > --> Populating transaction set with selected packages.. > > During the wait, there is continuous disk activity. and no network > activity. > This has been going on for a while (it didn't just start today) > I did a yum clean all to see if it would help but it did not. > > > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list > From giallu at gmail.com Wed May 11 10:46:57 2005 From: giallu at gmail.com (Gianluca Sforna) Date: Wed, 11 May 2005 12:46:57 +0200 Subject: Thunderbird maildir manipulation error Message-ID: Hi, I am going to submit a bug for thunderbird and, since I have not installed anything newer than FC3, I would like to ask your help for testing this in FC4test and/or rawhide. The bug is related to maildir manipulation (using Dovecot IMAP server); here are the steps to reproduce: 1. create a "test" folder 2. create a "test2" subfolder inside "test" 3. try to move test2 to the top level With the installed thunderbird-1.0.2-1.3.2 I got this error: The current command did not succeed. The mail server responded: Mailbox exists.. just to be sure this problem was not related to the server, I tried with evolution (evolution-2.0.4-4) and it works as expected, that is, the folder is successfully moved. I will report this against the FC3 version of thunderbird unless someone can confirm the same error appears in more recent versions. Cheers Gianluca From arnaud.abelard at univ-nantes.fr Wed May 11 11:36:44 2005 From: arnaud.abelard at univ-nantes.fr (=?ISO-8859-1?Q?Arnaud_Ab=E9lard?=) Date: Wed, 11 May 2005 13:36:44 +0200 Subject: Thunderbird maildir manipulation error In-Reply-To: References: Message-ID: <4281EE4C.1030004@univ-nantes.fr> Gianluca Sforna wrote: > Hi, > I am going to submit a bug for thunderbird and, since I have not > installed anything newer than FC3, I would like to ask your help for > testing this in FC4test and/or rawhide. > The bug is related to maildir manipulation (using Dovecot IMAP > server); here are the steps to reproduce: > > 1. create a "test" folder > 2. create a "test2" subfolder inside "test" > 3. try to move test2 to the top level > > With the installed thunderbird-1.0.2-1.3.2 I got this error: > > The current command did not succeed. The mail server responded: Mailbox exists.. Sounds like a dovecot bug to me. Doesn't occur on my FC3 with thunderbird 1.0.2 and courier-imap. AA. -- Arnaud Ab?lard Administrateur Syst?mes et R?seaux Facult? de Sciences et Techniques Universit? de Nantes From jules at tdcadsl.dk Wed May 11 11:52:47 2005 From: jules at tdcadsl.dk (Jules Colding) Date: Wed, 11 May 2005 13:52:47 +0200 Subject: Freezes with 2.6.11-1.14_FC3smp Message-ID: <1115812367.5694.12.camel@omc-3.omesc.com> Hi, I have had two total freezes (couldn't even ping it) with this kernel today. Nothing in the logs, it just stopped dead. The only common about the two freezes was that I was browsing with Firefox when it happened. System is up to date. I have now booted 2.6.10-1.770_FC3smp and the problem seems to have vanished. I'll send further information if anyone wants it. Best regards, jules From buildsys at redhat.com Wed May 11 11:59:02 2005 From: buildsys at redhat.com (Build System) Date: Wed, 11 May 2005 07:59:02 -0400 Subject: rawhide report: 20050511 changes Message-ID: <200505111159.j4BBx2Pg027184@porkchop.devel.redhat.com> Updated Packages: ORBit2-2.12.1-3 --------------- * Tue May 10 2005 Mark McLoughlin 2.12.1-3 - Add patch to set the size of the IO thread stack to 256k in order to mitigate the apparent 10M jump in GNOME processes memory usage (#157297) bash-3.0-31 ----------- * Tue May 10 2005 Tim Waugh 3.0-31 - Small fix for multibyteifs patch to prevent segfault (bug #157260). * Wed Apr 20 2005 Tim Waugh - Fixed AFS support for output redirection, so that the correct errors are reported for other filesystems (bug #155373). cpuspeed-1:1.2.1-1.21 --------------------- * Mon May 09 2005 Dave Jones - Fix debuginfo generation. e2fsprogs-1.37-4 ---------------- * Tue May 10 2005 Jeremy Katz - 1.37-4 - added libblkid.so to devel package elfutils-0.108-1 ---------------- * Mon May 09 2005 Roland McGrath - 0.108-1 - update to 0.108 - merge strip fixes - sort records in dwarf_getsrclines, fix dwarf_getsrc_die searching - update elf.h from glibc elinks-0.10.3-2 --------------- * Tue May 10 2005 Miloslav Trmac - 0.10.3-2 - Fix checking for numeric command prefix (#152953, patch by Jonas Fonseca) - Fix invalid C causing assertion errors on ppc and ia64 (#156647) gtk2-2.6.7-4 ------------ * Tue May 10 2005 Matthias Clasen - remove the openssl prereq again, as it did not fix Florians problem. kernel-2.6.11-1.1290_FC4 ------------------------ * Mon May 09 2005 Dave Jones - Rebase to 2.6.12-rc4 | Xen builds are temporarily disabled again. libwnck-2.10.0-3 ---------------- * Tue May 10 2005 Ray Strode 2.10.0-3 - fix some rendering issues with the last patch * Tue May 10 2005 Ray Strode 2.10.0-2 - add noticable glowing effect to task bar to counteract focus stealing prevention (bug #157285) lsof-4.74-7 ----------- * Tue May 10 2005 Karel Zak 4.74-7 - fix debuginfo magma-plugins-1.0-0.pre18.3 --------------------------- mod_auth_kerb-5.0-6 ------------------- * Tue May 10 2005 Joe Orton 5.0-6 - update to 5.0rc6 - don't force CC=gcc4 mozilla-37:1.7.7-4 ------------------ * Fri May 06 2005 Christopher Aillon 37:1.7.7-4 - Own all the files we install, or otherwise create. (#74160) php-5.0.4-10 ------------ * Fri May 06 2005 Joe Orton 5.0.4-10 - disable RPATHs in shared extensions (#156974) postgresql-8.0.3-1 ------------------ * Tue May 10 2005 Tom Lane 8.0.3-1 - Update to PostgreSQL 8.0.3 (includes security and data-loss fixes; see bz#156727, CAN-2005-1409, CAN-2005-1410) - Update to jdbc driver build 311 - Recreate postgres user after superseding an rh-postgresql install (bug #151911) - Ensure postgresql server is restarted if running during an upgrade * Thu Apr 14 2005 Florian La Roche 8.0.2-2 - rebuild for postgresql-tcl * Tue Apr 12 2005 Tom Lane 8.0.2-1 - Update to PostgreSQL 8.0.2. procinfo-18-16 -------------- * Tue May 10 2005 Karel Zak 18-16 - fixed debuginfo procps-3.2.5-6 -------------- * Tue May 10 2005 Karel Zak 3.2.5-6 - fix permissions in the spec install section * Tue May 10 2005 Karel Zak 3.2.5-5 - fix debuginfo statserial-1.1-38 ----------------- * Tue May 10 2005 Tim Waugh 1.1-38 - Don't strip the binary. struts11-0:1.1-1jpp_7fc ----------------------- * Tue May 10 2005 Gary Benson - 0:1.1-1jpp_7fc - Embed struts-legacy classes in main jarfile. sysreport-1.4.1-2 ----------------- * Tue May 10 2005 Than Ngo 1.4.1-2 - fix typo bug * Tue May 10 2005 Than Ngo 1.4.1-1 - collect /etc/stinit.def, #154996 - additional info for Xorg/XFree86, #156305 - fix output problem #154120 - add -S sparse files, #150584, #156858 - don't overwrite output files, #154329 - collect /etc/ld.so.conf.d, #154328 - gather bind stats correctly #143632 - cosmetic fixes system-config-bind-4.0.0-16_FC4 ------------------------------- * Sun May 08 2005 Jason Vas Dias - 4.0.0-16_FC4 - fix bug 157207: allow build to succeed if bind package is not installed * Thu May 05 2005 Jason Vas Dias - 4.0.0-16 - fix out-of-zone data reporting - out-of-zone string comparison should be case-insensitive - fix bug 156913: wrong file permissions for config files * Wed May 04 2005 Jason Vas Dias - 4.0.0-12 - fix bug 156884: handle named.conf with NO options clause . system-config-printer-0.6.130-1 ------------------------------- * Tue May 10 2005 Tim Waugh 0.6.130-1 - 0.6.130: - Another driver name change added to updateconf.py (bug #156605). - Fixed Apply button sensitivity after queue is deleted. - Avoid the need for cluttering up the Info line (bug #117943). tomcat5-0:5.0.30-5jpp_6fc ------------------------- * Tue May 10 2005 Gary Benson 0:5.0.30-5jpp_6fc - Precompile webapps (#157205). vlock-1.3-19 ------------ * Tue May 10 2005 Karel Zak 1.3-19 - fix debuginfo From jspaleta at gmail.com Wed May 11 13:12:25 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Wed, 11 May 2005 09:12:25 -0400 Subject: yum very slow In-Reply-To: <4281C547.5090506@mindspring.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> Message-ID: <604aa791050511061263ff1b56@mail.gmail.com> On 5/11/05, Richard Hally wrote: > Excluding Packages in global exclude list > Finished > (wait an hour + ) > Resolving Dependencies > --> Populating transaction set with selected packages.. > > During the wait, there is continuous disk activity. and no network > activity. > This has been going on for a while (it didn't just start today) > I did a yum clean all to see if it would help but it did not. that's fascinatingly different than what I'm seeing. How much memory do you have.. and how much memory is the yum process using during that (wait an hour+) I'm definitely not experiencing this on my rawhide box. -jef From wbeebe at gmail.com Wed May 11 16:08:42 2005 From: wbeebe at gmail.com (William Beebe) Date: Wed, 11 May 2005 12:08:42 -0400 Subject: Problems with buring FC4 test ISOs Message-ID: I've been having problems burning ISOs for FC4 test 2 and test 3. If I download ISOs to Windows XP and then used Nero 6 to burn the CDs/DVDs, they failed the integrity check when booted up. If, on the other hand, I boot into Linux (SuSE 9.1 in this instance) and run K3b as root (kdesu k3b) I can burn the exact same ISO images to blank media using the exact same hardware and the new CD/DVD will successfully pass its integrity check. As stated I downloaded the DVD ISO on WinXP using Firefox. When the first burn failed I rebooted into SuSE 9.1, mounted the Windows file system to get to the ISO and burned a second successful copy. I'm currently installing FC4 test 3 using the good DVD. So far so good. DVD installation is so superior to the CD installation shuffle. It's worth the time to download or buy the DVD. Well, it just finished successfully installing. That's the fastest I've ever dropped a Linux on any box, regardless of distribution, and it's as fast if not faster than Windows. FC4t3 seems a lot snappier when booting and logging in as well. Gnome seems cleaner, but I'm still going to switch over to KDE once I've got other items reconfigured. Overall this is a very good beginning for Fedcora Core 4. -------------- next part -------------- An HTML attachment was scrubbed... URL: From florin at andrei.myip.org Wed May 11 17:04:12 2005 From: florin at andrei.myip.org (Florin Andrei) Date: Wed, 11 May 2005 10:04:12 -0700 Subject: the SSH worm thing Message-ID: <1115831052.6718.3.camel@stantz.corp.sgi.com> http://www.schneier.com/blog/archives/2005/05/the_potential_f.html I can't test it right now, but i wonder - what's the default setting on FC4, hash the hosts or not? -- Florin Andrei http://florin.myip.org/ From alan at redhat.com Wed May 11 17:34:00 2005 From: alan at redhat.com (Alan Cox) Date: Wed, 11 May 2005 13:34:00 -0400 Subject: the SSH worm thing In-Reply-To: <1115831052.6718.3.camel@stantz.corp.sgi.com> References: <1115831052.6718.3.camel@stantz.corp.sgi.com> Message-ID: <20050511173400.GA29873@devserv.devel.redhat.com> On Wed, May 11, 2005 at 10:04:12AM -0700, Florin Andrei wrote: > http://www.schneier.com/blog/archives/2005/05/the_potential_f.html > > I can't test it right now, but i wonder - what's the default setting on > FC4, hash the hosts or not? I'm not convinced it helps very much. I'll just read every .history file on your machine and hash the hostnames I find in that against the database. I'd also try cvs based attacks by using the keys that work and appear to be for cvs stuff to automate pushing updated autoconf scripts into every cvs I can 'fix'. There are just far too many other ways to identify an ssh host entry/key and to then use that the same way the analysed user has. Alan From tmraz at redhat.com Wed May 11 17:47:45 2005 From: tmraz at redhat.com (Tomas Mraz) Date: Wed, 11 May 2005 19:47:45 +0200 Subject: the SSH worm thing In-Reply-To: <20050511173400.GA29873@devserv.devel.redhat.com> References: <1115831052.6718.3.camel@stantz.corp.sgi.com> <20050511173400.GA29873@devserv.devel.redhat.com> Message-ID: <1115833666.5882.16.camel@perun.redhat.usu> On Wed, 2005-05-11 at 13:34 -0400, Alan Cox wrote: > On Wed, May 11, 2005 at 10:04:12AM -0700, Florin Andrei wrote: > > http://www.schneier.com/blog/archives/2005/05/the_potential_f.html > > > > I can't test it right now, but i wonder - what's the default setting on > > FC4, hash the hosts or not? No, it's currently left as default which means no hashing of known hosts. > I'm not convinced it helps very much. I'll just read every .history file on > your machine and hash the hostnames I find in that against the database. I'd > also try cvs based attacks by using the keys that work and appear to be for > cvs stuff to automate pushing updated autoconf scripts into every cvs I can > 'fix'. > > There are just far too many other ways to identify an ssh host entry/key and > to then use that the same way the analysed user has. Also if the attacker could read the known_hosts file he could also change the user's environment so it instead of ssh calls a malicious script/binary which would log user's credentials and only then called the real ssh binary. -- Tomas Mraz From jcastro at instant.com.br Wed May 11 18:28:52 2005 From: jcastro at instant.com.br (Juan Carlos Castro y Castro) Date: Wed, 11 May 2005 15:28:52 -0300 Subject: I'm experimenting with Kernel Preemption Message-ID: <42824EE4.9050308@instant.com.br> (I posted this in fedora-list and Rahul Sundaram suggested I post it here instead) Not on Fedora's kernel sources, but with 2.6.11ac7. I based my config on /boot/config-2.6.11-1.14_FC3 and dealt with the additional options with "make oldconfig". Then I browsed the configuration with "make menuconfig", just for fun. I saw kernel preemption was turned off, so I turned on. Afterwards, I notice the system is noticeably faster. Bootup is faster. Shutdown is faster. The Red Hat manu on GNOME pops up WAY faster. OpenOffice.org loading is faster. I suspect other things are faster too, but I'd have to time them. So my question is: why isn't preemption enabled in the FC3 packaged kernel? Does it conflict with something I haven't encountered yet? maybe some esoteric hardware combination? My hardware data is below. Another thing: what crucial patch, if any, am I missing by using 2.6.11ac7 instead of the FC3 packaged kernel? [jcastro at ws06 ~]$ cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 6 model : 8 model name : AMD Athlon(tm) XP 2400+ stepping : 1 cpu MHz : 1991.906 cache size : 256 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse pni syscall mmxext 3dnowext 3dnow bogomips : 3948.54 [jcastro at ws06 ~]$ /sbin/lspci 00:00.0 Host bridge: VIA Technologies, Inc. VT8378 [KM400/A] Chipset Host Bridge00:01.0 PCI bridge: VIA Technologies, Inc. VT8237 PCI Bridge 00:10.0 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 80) 00:10.1 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 80) 00:10.2 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 80) 00:10.3 USB Controller: VIA Technologies, Inc. USB 2.0 (rev 82) 00:11.0 ISA bridge: VIA Technologies, Inc. VT8235 ISA Bridge 00:11.1 IDE interface: VIA Technologies, Inc. VT82C586A/B/VT82C686/A/B/VT823x/A/C PIPC Bus Master IDE (rev 06) 00:11.5 Multimedia audio controller: VIA Technologies, Inc. VT8233/A/8235/8237 AC97 Audio Controller (rev 50) 00:12.0 Ethernet controller: VIA Technologies, Inc. VT6102 [Rhine-II] (rev 74) 01:00.0 VGA compatible controller: VIA Technologies, Inc. VT8378 [S3 UniChrome] Integrated Video (rev 01) From florin at andrei.myip.org Wed May 11 18:45:28 2005 From: florin at andrei.myip.org (Florin Andrei) Date: Wed, 11 May 2005 11:45:28 -0700 Subject: the SSH worm thing In-Reply-To: <1115833666.5882.16.camel@perun.redhat.usu> References: <1115831052.6718.3.camel@stantz.corp.sgi.com> <20050511173400.GA29873@devserv.devel.redhat.com> <1115833666.5882.16.camel@perun.redhat.usu> Message-ID: <1115837128.6718.17.camel@stantz.corp.sgi.com> On Wed, 2005-05-11 at 19:47 +0200, Tomas Mraz wrote: > On Wed, 2005-05-11 at 13:34 -0400, Alan Cox wrote: > > On Wed, May 11, 2005 at 10:04:12AM -0700, Florin Andrei wrote: > > > http://www.schneier.com/blog/archives/2005/05/the_potential_f.html > > I'm not convinced it helps very much. I'll just read every .history file on > > your machine and hash the hostnames I find in that against the database. Right, but the entries in the .history files are typically short-lived, while the ones in known_hosts are, more or less, forever. I just verified my .bash_history file and it has 10 different addresses/hostnames that i ssh'ed to. known_hosts has 40. That's approx. halfway through to the next order of magnitude. And i'm using ssh quite a lot. Slowing down the attack vector by (almost) an order of magnitude is no small feat - i bet you it translates into many orders of magnitude in the difference between the population exhaustion times. Agree, it depends on a multitude of factors, but it could be the difference between a malware that takes the Internet by storm, and a moderate infection that can be contained. > > There are just far too many other ways to identify an ssh host entry/key and > > to then use that the same way the analysed user has. True, but there's no universal cure for anything. You gotta take a first step somewhere. This first step is practically gratis. > Also if the attacker could read the known_hosts file he could also > change the user's environment so it instead of ssh calls a malicious > script/binary which would log user's credentials and only then called > the real ssh binary. Correct, but the hash-armoured known_hosts file has the purpose to stop a potential SSH worm from spreading like wildfire: infect a machine, then in a few seconds infect a dozen more, repeat. It's the same exponential growth mechanism that made so dangerous some Outlook malware that were able to read the address book. The mechanism you describe is entirely different, it's an altogether different attack. -- Florin Andrei http://florin.myip.org/ From jkeating at j2solutions.net Wed May 11 18:56:28 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Wed, 11 May 2005 11:56:28 -0700 Subject: the SSH worm thing In-Reply-To: <1115837128.6718.17.camel@stantz.corp.sgi.com> References: <1115831052.6718.3.camel@stantz.corp.sgi.com> <20050511173400.GA29873@devserv.devel.redhat.com> <1115833666.5882.16.camel@perun.redhat.usu> <1115837128.6718.17.camel@stantz.corp.sgi.com> Message-ID: <1115837788.11675.140.camel@jkeating2.hq.pogolinux.com> On Wed, 2005-05-11 at 11:45 -0700, Florin Andrei wrote: > Correct, but the hash-armoured known_hosts file has the purpose to > stop > a potential SSH worm from spreading like wildfire: infect a machine, > then in a few seconds infect a dozen more, repeat. It's the same > exponential growth mechanism that made so dangerous some Outlook > malware > that were able to read the address book. > The mechanism you describe is entirely different, it's an altogether > different attack. > How would this hash interact with the need to modify known_hosts when systems change and IPs have conflicting mac addresses and such? This is about 5 times a week for me here at work in our lab... -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From florin at andrei.myip.org Wed May 11 19:03:40 2005 From: florin at andrei.myip.org (Florin Andrei) Date: Wed, 11 May 2005 12:03:40 -0700 Subject: the SSH worm thing In-Reply-To: <1115837788.11675.140.camel@jkeating2.hq.pogolinux.com> References: <1115831052.6718.3.camel@stantz.corp.sgi.com> <20050511173400.GA29873@devserv.devel.redhat.com> <1115833666.5882.16.camel@perun.redhat.usu> <1115837128.6718.17.camel@stantz.corp.sgi.com> <1115837788.11675.140.camel@jkeating2.hq.pogolinux.com> Message-ID: <1115838220.6718.21.camel@stantz.corp.sgi.com> On Wed, 2005-05-11 at 11:56 -0700, Jesse Keating wrote: > How would this hash interact with the need to modify known_hosts when > systems change and IPs have conflicting mac addresses and such? This is > about 5 times a week for me here at work in our lab... $ man ssh-keygen [...] -R hostname Removes all keys belonging to hostname from a known_hosts file. This option is useful to delete hashed hosts (see the -H option above). [...] -- Florin Andrei http://florin.myip.org/ From mbneto at gmail.com Wed May 11 19:37:11 2005 From: mbneto at gmail.com (mbneto) Date: Wed, 11 May 2005 15:37:11 -0400 Subject: %postun Error In-Reply-To: <427CD682.5070603@hhs.nl> References: <5cf776b805050707254407d05d@mail.gmail.com> <427CD682.5070603@hhs.nl> Message-ID: <5cf776b8050511123731e9f642@mail.gmail.com> Thanks hans. Unfortunately the new upgrades still complain about post/pre scripts. Is there a better way to solve this ? On 5/7/05, Hans de Goede wrote: > > > mbneto wrote: > > Hi, > > > > The latest updates complain about a postrun. Seems rpm related. > > > > i.e > > ... > > Cleanup : ant ####################### [36/38] > > (ant-1.6.2-3jpp_5fc.i386) scriptlet failed, exit status 255 > > > > It occurs with other packages as well. > > > > Hmm, I had this to when I did a piecemeal upgrade to rawhide, the fix > for me was to install the latest selinux libs (can't remember the proper > package name) or just make sure your system is fully uptodate with yum > -y. Better make sure this is fixex soon, because each post-uninstall > script which fails causes the old version of the package to stay in the > rpmdb. > > After you've got this fixed you can rpm -qa |sort, look for duplictes > and rpm -e --dbonly the dups, this might cause problems if files we're > moved / removed, but not doing --justdb will nuke files from the newer > version. > > Regards, > > Hans > From mattdm at mattdm.org Wed May 11 19:39:55 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Wed, 11 May 2005 15:39:55 -0400 Subject: the SSH worm thing In-Reply-To: <1115833666.5882.16.camel@perun.redhat.usu> References: <1115831052.6718.3.camel@stantz.corp.sgi.com> <20050511173400.GA29873@devserv.devel.redhat.com> <1115833666.5882.16.camel@perun.redhat.usu> Message-ID: <20050511193955.GA20649@jadzia.bu.edu> On Wed, May 11, 2005 at 07:47:45PM +0200, Tomas Mraz wrote: > No, it's currently left as default which means no hashing of known > hosts. Which is nice for tab-completion. -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 84 degrees Fahrenheit. From jbuell at vmware.com Wed May 11 20:08:11 2005 From: jbuell at vmware.com (Jeffrey Buell) Date: Wed, 11 May 2005 13:08:11 -0700 Subject: rpmbuild problem in FC4 release 1290 Message-ID: <58743620D2C0D9439C627C064581E25218D725@PA-ECLUSTER2.vmware.com> I have used rpmbuild several times with recent releases, but have never seen this problem before. I'm using release 1290 (1288 did something similar). The tail end of the output of rpmbuild -bp --target=i686 kernel-2.6.spec is: scripts/kconfig/confdata.c: In function 'conf_get_default_confname': scripts/kconfig/confdata.c:62: warning: pointer targets in passing argument 1 of 'conf_expand_value' differ in signedness scripts/kconfig/confdata.c: In function 'conf_read': scripts/kconfig/confdata.c:88: warning: pointer targets in passing argument 1 of 'conf_expand_value' differ in signedness .config:2477: trying to assign nonexistent symbol NETDUMP CONFIG_X86_REBOOTFIXUPS CONFIG_BLK_DEV_DELKIN CONFIG_BLK_DEV_IT821X CONFIG_SCSI_LPFC CONFIG_USB_SERIAL_AIRPRIME CONFIG_USB_SERIAL_HP4X make[1]: *** [nonint_oldconfig] Error 6 make: *** [nonint_oldconfig] Error 2 error: Bad exit status from /var/tmp/rpm-tmp.22022 (%prep) RPM build errors: Bad exit status from /var/tmp/rpm-tmp.22022 (%prep) The .config file doesn't even have NETDUMP in it. The other parameters listed are all not set. Any known problems with rpmbuild or am I missing something? Jeff From kyrre at solution-forge.net Wed May 11 20:10:55 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Wed, 11 May 2005 22:10:55 +0200 Subject: Java and OOo 2 Message-ID: <1115842255.4197.11.camel@localhost.localdomain> There is still a dependency problem in OpenOffice 2.0, it does require Java to run properly, but does not pull it in as a dependancy. This should be fixed before FC4 is released. https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156419 I am posting here to get some attention to the problem. Kyrre Ness Sj?b?k From wtogami at redhat.com Wed May 11 20:12:51 2005 From: wtogami at redhat.com (Warren Togami) Date: Wed, 11 May 2005 10:12:51 -1000 Subject: yum very slow In-Reply-To: <604aa791050511061263ff1b56@mail.gmail.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> <604aa791050511061263ff1b56@mail.gmail.com> Message-ID: <42826743.8070400@redhat.com> Jeff Spaleta wrote: > On 5/11/05, Richard Hally wrote: > >>Excluding Packages in global exclude list >>Finished >>(wait an hour + ) >>Resolving Dependencies >>--> Populating transaction set with selected packages.. >> >>During the wait, there is continuous disk activity. and no network >>activity. >>This has been going on for a while (it didn't just start today) >>I did a yum clean all to see if it would help but it did not. > > > that's fascinatingly different than what I'm seeing. How much memory > do you have.. and how much memory is the yum process using during that > (wait an hour+) I'm definitely not experiencing this on my rawhide > box. > developmen: ################################################## 3770/3770 Added 135 new packages, deleted 137 old in 8.43 seconds (wait about 2 minutes with 100% CPU and some disk activity) Resolving Dependencies --> Populating transaction set with selected packages. Please wait. ---> Downloading header for kernel-devel to pack into transaction set. Athlon64 3200+ x86_64 1GB RAM Not quite an hour, but that's a long time to get stuck with no visible indication of what's happening. Warren Togami wtogami at redhat.com From vonbrand at inf.utfsm.cl Wed May 11 21:22:27 2005 From: vonbrand at inf.utfsm.cl (Horst von Brand) Date: Wed, 11 May 2005 17:22:27 -0400 Subject: the SSH worm thing In-Reply-To: Your message of "Wed, 11 May 2005 10:04:12 MST." <1115831052.6718.3.camel@stantz.corp.sgi.com> Message-ID: <200505112122.j4BLMRp3014498@laptop11.inf.utfsm.cl> Florin Andrei said: > http://www.schneier.com/blog/archives/2005/05/the_potential_f.html > > I can't test it right now, but i wonder - what's the default setting on > FC4, hash the hosts or not? AFAIK, no OpenSSH has ever used a hash of the host names. Neither has closed SSH, for that matter. And I see this as a very mild problem. Yes, for example I have the same password on a group of machines (small wonder, it's the same account handled via LDAP + NFS), so cracking one gives access to the others. But if they cracked my password here they could just try it on "nearby" machines, with even better results: I haven't connected to all the machines that share my account. Yes, I also do have accounts on remote machines. The accounts are not necesarily called the same as this one, and their passwords are different too. The /real/ risk is having the same account across machines. I'm quite happy with it for my personal use. For managing (some of) the machines themselves I'm not so happy (but they aren't critical, so this is not a huge risk either). -- Dr. Horst H. von Brand User #22616 counter.li.org Departamento de Informatica Fono: +56 32 654431 Universidad Tecnica Federico Santa Maria +56 32 654239 Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513 From skvidal at phy.duke.edu Wed May 11 21:26:41 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Wed, 11 May 2005 17:26:41 -0400 Subject: yum very slow In-Reply-To: <42826743.8070400@redhat.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> <604aa791050511061263ff1b56@mail.gmail.com> <42826743.8070400@redhat.com> Message-ID: <1115846801.23946.0.camel@cutter> > > > > developmen: ################################################## 3770/3770 > Added 135 new packages, deleted 137 old in 8.43 seconds > (wait about 2 minutes with 100% CPU and some disk activity) > Resolving Dependencies > --> Populating transaction set with selected packages. Please wait. > ---> Downloading header for kernel-devel to pack into transaction set. > > Athlon64 3200+ x86_64 1GB RAM > > Not quite an hour, but that's a long time to get stuck with no visible > indication of what's happening. > we've done this before on your system, warren, disable ipv6 support and see if it miraculously speeds up. -sv From wtogami at redhat.com Wed May 11 21:37:58 2005 From: wtogami at redhat.com (Warren Togami) Date: Wed, 11 May 2005 11:37:58 -1000 Subject: yum very slow In-Reply-To: <1115846801.23946.0.camel@cutter> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> <604aa791050511061263ff1b56@mail.gmail.com> <42826743.8070400@redhat.com> <1115846801.23946.0.camel@cutter> Message-ID: <42827B36.6040805@redhat.com> seth vidal wrote: >>developmen: ################################################## 3770/3770 >>Added 135 new packages, deleted 137 old in 8.43 seconds >>(wait about 2 minutes with 100% CPU and some disk activity) >>Resolving Dependencies >>--> Populating transaction set with selected packages. Please wait. >>---> Downloading header for kernel-devel to pack into transaction set. >> >>Athlon64 3200+ x86_64 1GB RAM >> >>Not quite an hour, but that's a long time to get stuck with no visible >>indication of what's happening. >> > > > we've done this before on your system, warren, disable ipv6 support and > see if it miraculously speeds up. > It doesn't, this is a different issue. That point where it stops with 100% CPU usage and disk usage for 2 minutes is apparently the only slow point now, everything else is very fast. Warren Togami wtogami at redhat.com From mharris at www.linux.org.uk Thu May 12 01:55:40 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Wed, 11 May 2005 21:55:40 -0400 Subject: Thunderbird maildir manipulation error In-Reply-To: References: Message-ID: <4282B79C.8060107@www.linux.org.uk> Gianluca Sforna wrote: > Hi, > I am going to submit a bug for thunderbird and, since I have not > installed anything newer than FC3, I would like to ask your help for > testing this in FC4test and/or rawhide. > The bug is related to maildir manipulation (using Dovecot IMAP > server); here are the steps to reproduce: > > 1. create a "test" folder > 2. create a "test2" subfolder inside "test" > 3. try to move test2 to the top level > > With the installed thunderbird-1.0.2-1.3.2 I got this error: > > The current command did not succeed. The mail server responded: Mailbox exists.. > > just to be sure this problem was not related to the server, I tried > with evolution (evolution-2.0.4-4) and it works as expected, that is, > the folder is successfully moved. > I will report this against the FC3 version of thunderbird unless > someone can confirm the same error appears in more recent versions. Aha! I have gotten this bug many times but did not have a reliable 100% way of reproducing it, so never filed a bug report! What happens is you get the error message you've supplied, or you get another one "The mail server responded: Invalid mask". Either error message puts thunderbird into an infinite loop, where it puts the error dialog up, you click OK *or* cancel, and it puts up another dialog with a different message (don't have it handy currently, and I'm not anxious to see it again as it is very difficult to recover from this bug) - then you click ok or cancel again, and the first error comes up again. It makes no difference how many times you click on OK or cancel, thunderbird keeps cycling through these two dialog boxes forever, until you kill the program from the commandline with "killall thunderbird-bin". With the "Invalid mask" error, if I restart thunderbird, it no longer works anymore, and goes right back to the looping dialog boxes. The only way I've been able to fix thunderbird at this point, is to: - killall thunderbird-bin - backup .thunderbird/* - move .thunderbird .thunderbird-old - cd mail ; find . -regex "\.\(imap\|subscriptions\)" -exec rm -rf {} \; - start thunderbird again, and it now acts like you just started it for the first time ever - completely reconfigure all my imap accounts, smtp, etc. - completely reconfigure the UI to work the way I want - completely reconfigure per-folder options for all my folders Then it works normally again, until the next time I move a folder and trigger the bug. The problem is, it is such a hassle to recover from the problem, and very time consuming, that it is hard to make a useful bug report about it, without intentionally causing a lot of mayhem. If you file a bug for this, let me know and I'll add whatever details I think might be useful to it too. For the record - this has happened with every version of thunderbird I've used since 1.0, up to and including the current 1.0.2 from rawhide. It happens with the binary build in rawhide, as well as with thunderbird recompiled for FC1 with a couple of patches disabled (pango mainly). While experiencing this problem, I've used the Windows XP build of thunderbird a few times just so I could access my mail while troubleshooting the problems in Linux that I was experiencing. I've not yet encountered the bug in any of the Windows builds of thunderbird, but it could just be that the problem exists there too and I haven't triggered it yet. I don't use it enough in XP to really know either way, but if it would be helpful in any way to resolving the problem, I could torture myself for a few days to use XP for mail. HTH From mharris at www.linux.org.uk Thu May 12 02:02:35 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Wed, 11 May 2005 22:02:35 -0400 Subject: Thunderbird maildir manipulation error In-Reply-To: <4282B79C.8060107@www.linux.org.uk> References: <4282B79C.8060107@www.linux.org.uk> Message-ID: <4282B93B.2010601@www.linux.org.uk> Mike A. Harris wrote: > Gianluca Sforna wrote: > >> Hi, >> I am going to submit a bug for thunderbird and, since I have not >> installed anything newer than FC3, I would like to ask your help for >> testing this in FC4test and/or rawhide. >> The bug is related to maildir manipulation (using Dovecot IMAP >> server); here are the steps to reproduce: >> >> 1. create a "test" folder 2. create a "test2" subfolder inside "test" >> 3. try to move test2 to the top level >> >> With the installed thunderbird-1.0.2-1.3.2 I got this error: >> >> The current command did not succeed. The mail server responded: >> Mailbox exists.. >> >> just to be sure this problem was not related to the server, I tried >> with evolution (evolution-2.0.4-4) and it works as expected, that is, >> the folder is successfully moved. >> I will report this against the FC3 version of thunderbird unless >> someone can confirm the same error appears in more recent versions. > > > Aha! I have gotten this bug many times but did not have a reliable > 100% way of reproducing it, so never filed a bug report! > > What happens is you get the error message you've supplied, or you > get another one "The mail server responded: Invalid mask". Either > error message puts thunderbird into an infinite loop, where it > puts the error dialog up, you click OK *or* cancel, and it puts up > another dialog with a different message (don't have it handy > currently, and I'm not anxious to see it again as it is very difficult > to recover from this bug) - then you click ok or cancel again, and > the first error comes up again. It makes no difference how many > times you click on OK or cancel, thunderbird keeps cycling through > these two dialog boxes forever, until you kill the program from > the commandline with "killall thunderbird-bin". > > With the "Invalid mask" error, if I restart thunderbird, it no longer > works anymore, and goes right back to the looping dialog boxes. The > only way I've been able to fix thunderbird at this point, is to: Hate to reply to myself, but thought I'd add a couple more data points in case they were useful also. I am using thunderbird exclusively connecting to imap mailboxes when these errors trigger. The imap server is running on the same machine, and is dovecot. I've tried different versions of dovecot and it didn't change anything. Once thunderbird encounters the error, restarting dovecot wont change it. Using different version of dovecot wont change it either. I tried using UW-imap, and that just screwed up my imap mailbox subscriptions in tbird, and left me with a totally unuseable setup. I saw someone mention this could be a dovecot bug, and that is entirely possible, but thunderbird enters into an endless loop of dialog boxes requiring you to kill the binary, so there's definitely a thunderbird bug also, even if it is a mail server bug being the catalyst. HTH From rhally at mindspring.com Thu May 12 02:59:38 2005 From: rhally at mindspring.com (Richard Hally) Date: Wed, 11 May 2005 22:59:38 -0400 Subject: yum very slow In-Reply-To: <604aa791050511061263ff1b56@mail.gmail.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> <604aa791050511061263ff1b56@mail.gmail.com> Message-ID: <4282C69A.4030605@mindspring.com> Jeff Spaleta wrote: >On 5/11/05, Richard Hally wrote: > > >>Excluding Packages in global exclude list >>Finished >>(wait an hour + ) >>Resolving Dependencies >>--> Populating transaction set with selected packages.. >> >>During the wait, there is continuous disk activity. and no network >>activity. >>This has been going on for a while (it didn't just start today) >>I did a yum clean all to see if it would help but it did not. >> >> > >that's fascinatingly different than what I'm seeing. How much memory >do you have.. and how much memory is the yum process using during that >(wait an hour+) I'm definitely not experiencing this on my rawhide >box. > >-jef > > > Below is some of the output with -d 10 option specified. I may have been mistaken when I originally said the wait was while resolving dependencies. The long wait occurs immediately *before* the "Resolving Dependencies" is printed. Once the "Resolving Dependencies" is printed, the header downloads and the rest of the update continue normally. -------------------- Excluding dlm-debuginfo - 1.0-0.pre21.8.i386 Excluding dvgrab-debuginfo - 1.7-3.i386 Finished Excluding Incompatible Archs Finished Reading Local RPMDB Building updates object putting mozilla-dom-inspector in simple update putting procps in simple update putting elfutils in simple update putting glibc in complex update putting mod_auth_kerb in simple update putting sysreport in simple update putting libwnck in simple update putting gtk2-devel in simple update putting php-pear in simple update putting mozilla-nspr in simple update putting system-config-printer-gui in simple update putting php-odbc in simple update putting postgresql-docs in simple update putting statserial in simple update putting php-pgsql in simple update putting system-config-printer in simple update putting postgresql-tcl in simple update putting openssl in complex update putting elfutils-libelf-devel in simple update putting postgresql in simple update putting postgresql-contrib in simple update putting mozilla in simple update putting postgresql-devel in simple update putting vlock in simple update putting gtk2 in simple update putting e2fsprogs in simple update putting bash in simple update putting mozilla-mail in simple update putting ORBit2 in simple update putting fedora-release in simple update putting php-ncurses in simple update putting mozilla-js-debugger in simple update putting lsof in simple update putting php in simple update putting procinfo in simple update putting php-xmlrpc in simple update putting ORBit2-devel in simple update putting postgresql-test in simple update putting mozilla-nspr-devel in simple update putting php-imap in simple update putting kernel in complex update putting php-mysql in simple update putting elinks in simple update putting mozilla-devel in simple update putting postgresql-pl in simple update putting kernel-doc in simple update putting php-mbstring in simple update putting mozilla-nss-devel in simple update putting system-config-bind in simple update putting mozilla-nss in simple update putting php-snmp in simple update putting libwnck-devel in simple update putting e2fsprogs-devel in simple update putting php-gd in simple update putting postgresql-jdbc in simple update putting postgresql-server in simple update putting elfutils-devel in simple update putting mozilla-chat in simple update putting kernel in complex update putting php-devel in simple update putting postgresql-libs in simple update putting php-xml in simple update putting php-ldap in simple update putting elfutils-libelf in simple update putting postgresql-python in simple update processing glibc.i686 processing openssl processing kernel.i686 processing kernel.i686 (waiting here...) (then) Resolving Dependencies (then the rest of the update) while the "wait" is happening, here is what top looks like top - 15:56:04 up 3:23, 3 users, load average: 2.06, 1.97, 1.24 Tasks: 105 total, 2 running, 103 sleeping, 0 stopped, 0 zombie Cpu(s): 2.7% us, 2.7% sy, 0.0% ni, 0.0% id, 94.0% wa, 0.7% hi, 0.0% si Mem: 223316k total, 220440k used, 2876k free, 308k buffers Swap: 458744k total, 860k used, 457884k free, 61788k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 3979 root 18 0 32876 23m 6744 D 2.7 10.6 0:23.48 yum 2915 root 15 0 56632 18m 5448 S 1.3 8.3 2:51.02 X 3154 richard 15 0 38200 13m 8624 S 0.7 6.2 0:04.94 gnome-terminal 4118 root 16 0 2028 988 776 R 0.7 0.4 0:00.17 top 3097 richard 16 0 21820 9428 7332 S 0.3 4.2 0:00.52 wnck-applet 1 root 16 0 1748 560 480 S 0.0 0.3 0:01.85 init 2 root 34 19 0 0 0 S 0.0 0.0 0:00.00 ksoftirqd/0 3 root 10 -5 0 0 0 S 0.0 0.0 0:00.22 events/0 4 root 12 -5 0 0 0 S 0.0 0.0 0:00.02 khelper 5 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 kthread 7 root 20 -5 0 0 0 S 0.0 0.0 0:00.00 kacpid 86 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 kblockd/0 89 root 15 0 0 0 0 S 0.0 0.0 0:00.00 khubd 135 root 15 0 0 0 0 S 0.0 0.0 0:00.00 pdflush 136 root 15 0 0 0 0 S 0.0 0.0 0:00.03 pdflush 138 root 12 -5 0 0 0 S 0.0 0.0 0:00.00 aio/0 137 root 15 0 0 0 0 D 0.0 0.0 0:00.05 kswapd0 From sundaram at redhat.com Thu May 12 06:26:16 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Thu, 12 May 2005 11:56:16 +0530 Subject: I'm experimenting with Kernel Preemption In-Reply-To: <42824EE4.9050308@instant.com.br> References: <42824EE4.9050308@instant.com.br> Message-ID: <4282F708.7020706@redhat.com> Juan Carlos Castro y Castro wrote: > (I posted this in fedora-list and Rahul Sundaram suggested I post it > here instead) > > Not on Fedora's kernel sources, but with 2.6.11ac7. I based my config > on /boot/config-2.6.11-1.14_FC3 and dealt with the additional options > with "make oldconfig". Then I browsed the configuration with "make > menuconfig", just for fun. > > I saw kernel preemption was turned off, so I turned on. Afterwards, I > notice the system is noticeably faster. Bootup is faster. Shutdown is > faster. The Red Hat manu on GNOME pops up WAY faster. OpenOffice.org > loading is faster. I suspect other things are faster too, but I'd have > to time them. > > So my question is: why isn't preemption enabled in the FC3 packaged > kernel? Does it conflict with something I haven't encountered yet? > maybe some esoteric hardware combination? My hardware data is below. > > Another thing: what crucial patch, if any, am I missing by using > 2.6.11ac7 instead of the FC3 packaged kernel? for the record, this has answered by Dave Jones on the fedora users list itself regards Rahul From arjanv at redhat.com Thu May 12 08:04:58 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Thu, 12 May 2005 10:04:58 +0200 Subject: rpmbuild problem in FC4 release 1290 In-Reply-To: <58743620D2C0D9439C627C064581E25218D725@PA-ECLUSTER2.vmware.com> References: <58743620D2C0D9439C627C064581E25218D725@PA-ECLUSTER2.vmware.com> Message-ID: <1115885099.6273.0.camel@laptopd505.fenrus.org> On Wed, 2005-05-11 at 13:08 -0700, Jeffrey Buell wrote: > I have used rpmbuild several times with recent releases, but have never seen > this problem before. I'm using release 1290 (1288 did something similar). > The tail end of the output of > > rpmbuild -bp --target=i686 kernel-2.6.spec which architecture is the host ? -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From aani_avni at yahoo.com Thu May 12 11:09:12 2005 From: aani_avni at yahoo.com (avni thacker) Date: Thu, 12 May 2005 04:09:12 -0700 (PDT) Subject: any help on NISTnet emulators Message-ID: <20050512110912.84492.qmail@web54009.mail.yahoo.com> a small problem have any body of u heard of NISTnet emulator? Who ever had heard pls let me know how to configure it, i m using FC1 and what all changes would be required __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From dhollis at davehollis.com Thu May 12 11:29:55 2005 From: dhollis at davehollis.com (David Hollis) Date: Thu, 12 May 2005 07:29:55 -0400 Subject: rpmbuild problem in FC4 release 1290 In-Reply-To: <58743620D2C0D9439C627C064581E25218D725@PA-ECLUSTER2.vmware.com> References: <58743620D2C0D9439C627C064581E25218D725@PA-ECLUSTER2.vmware.com> Message-ID: <1115897395.18703.3.camel@dhollis-lnx.sunera.com> On Wed, 2005-05-11 at 13:08 -0700, Jeffrey Buell wrote: > I have used rpmbuild several times with recent releases, but have never seen > this problem before. I'm using release 1290 (1288 did something similar). > The tail end of the output of > > rpmbuild -bp --target=i686 kernel-2.6.spec > > is: > > scripts/kconfig/confdata.c: In function 'conf_get_default_confname': > scripts/kconfig/confdata.c:62: warning: pointer targets in passing argument 1 > of 'conf_expand_value' differ in signedness > scripts/kconfig/confdata.c: In function 'conf_read': > scripts/kconfig/confdata.c:88: warning: pointer targets in passing argument 1 > of 'conf_expand_value' differ in signedness > .config:2477: trying to assign nonexistent symbol NETDUMP > CONFIG_X86_REBOOTFIXUPS > CONFIG_BLK_DEV_DELKIN > CONFIG_BLK_DEV_IT821X > CONFIG_SCSI_LPFC > CONFIG_USB_SERIAL_AIRPRIME > CONFIG_USB_SERIAL_HP4X > make[1]: *** [nonint_oldconfig] Error 6 > make: *** [nonint_oldconfig] Error 2 > error: Bad exit status from /var/tmp/rpm-tmp.22022 (%prep) > > RPM build errors: > Bad exit status from /var/tmp/rpm-tmp.22022 (%prep) > > The .config file doesn't even have NETDUMP in it. The other parameters > listed are all not set. Any known problems with rpmbuild or am I missing > something? > > Jeff > > I saw something similar recently as I've rolled my own RPMS based on 1275 & 1287 but adding inotify. I found that just manually adding CONFIG_INOTIFY=y to the kernel-*.config files (the x86 ones anyway) took care of the issue. If I copied one of the configs to .config and ran 'make oldconfig' on it, I did get prompted for quite a few options, such as Processor type and the like. Some of those options were of the select from the list variety instead of Y/m/n style. I'm not sure if nonint_oldconfig can handle those types of options or not. -- David Hollis -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From mfioretti at mclink.it Thu May 12 11:56:52 2005 From: mfioretti at mclink.it (M. Fioretti) Date: Thu, 12 May 2005 13:56:52 +0200 Subject: Converting UNO packages to .rpm format? Message-ID: <20050512115652.GP3286@mclink.it> Greetings, there is a discussion going on at discuss at openoffice.org about how, in general, OO.o add-ons should be packaged for Linux: UNO format, .rpm/deb/ etc... and who should do it. In that context, I asked how to distribute add-ons (macros, templates, clip-art, patches, whatever) in such a way that they can be directly used by others to create native Linux packages in whatever format. Among other comments, I got the following comments: >I think Linux distributors should figure out if there is anything that >provents them from wrapping UNO packages into native packages. I would >imagine that to be rather simple. and: >The UNO package manager is documented in the UNO developers guide > Before going further on this track, I wanted to know if there already is some Fedora developer/packager who has studied this in detail, and has comments on this. Specifically, does anybody already has experience in converting (automatically) those packages to .rpm for Fedora, and integrating them in the whole system? TIA, Marco F. -- Marco Fioretti mfioretti, at the server mclink.it Fedora Core 3 for low memory http://www.rule-project.org/ Real Programmers don't play tennis, or any other sport that requires you to change clothes. Mountain climbing is OK, and real programmers wear their climbing boots to work in case a mountain should suddenly spring up in the middle of the machine room. From buildsys at redhat.com Thu May 12 11:58:15 2005 From: buildsys at redhat.com (Build System) Date: Thu, 12 May 2005 07:58:15 -0400 Subject: rawhide report: 20050512 changes Message-ID: <200505121158.j4CBwF4m010943@porkchop.devel.redhat.com> Updated Packages: booty-0.53-1 ------------ * Wed May 11 2005 Peter Jones 0.53-1 - Make partition installs of grubs first stage behave right. * Wed May 04 2005 Jeremy Katz 0.52-1 - add enablecdboot, enableofboot, and enablenetboot to yaboot.conf - try to detect mac os x and set up dual boot on pmac * Mon May 02 2005 Jeremy Katz - add handling for sx8 device names gaim-1:1.3.0-1.fc4 ------------------ * Tue May 10 2005 Warren Togami 1:1.3.0-1 - 1.3.0 many bug fixes and two security fixes long URL crash fix (#157017) CAN-2005-1261 MSN bad messages crash fix (#157202) CAN-2005-1262 libtiff-3.7.1-6 --------------- * Fri May 06 2005 Matthias Clasen - 3.7.1-6 - Fix a stack overflow mkinitrd-4.2.14-1 ----------------- * Tue May 10 2005 Peter Jones - 4.2.14-1 - Better init argument handling (no uninitialized args) * Fri May 06 2005 Peter Jones - 4.2.13-1 - allow for lvm VGs to be in the /dev/mapper/$VG-$LV format (#154767) nautilus-2.10.0-4 ----------------- * Wed May 11 2005 David Zeuthen 2.10.0-4 - Fix default font for zh_TW (#154185) xscreensaver-1:4.21-4 --------------------- * Wed May 11 2005 Ray Strode 1:4.21-3 - Allow configuration gui to support hacks with absolute paths (bug 157417). From giallu at gmail.com Thu May 12 14:11:49 2005 From: giallu at gmail.com (Gianluca Sforna) Date: Thu, 12 May 2005 16:11:49 +0200 Subject: Thunderbird maildir manipulation error In-Reply-To: <4282B79C.8060107@www.linux.org.uk> References: <4282B79C.8060107@www.linux.org.uk> Message-ID: On 5/12/05, Mike A. Harris wrote: > Aha! I have gotten this bug many times but did not have a reliable > 100% way of reproducing it, so never filed a bug report! > > What happens is you get the error message you've supplied, or you > get another one "The mail server responded: Invalid mask". Either > error message puts thunderbird into an infinite loop, where it > puts the error dialog up, you click OK *or* cancel, and it puts up > another dialog with a different message (don't have it handy > currently, and I'm not anxious to see it again as it is very difficult > to recover from this bug) - then you click ok or cancel again, and > the first error comes up again. It makes no difference how many > times you click on OK or cancel, thunderbird keeps cycling through > these two dialog boxes forever, until you kill the program from > the commandline with "killall thunderbird-bin". Well, this seems much more critical than mine: in my case, after closing the dialog I can work normally with the email and no corruption happens (thanks God...) > If you file a bug for this, let me know and I'll add whatever details > I think might be useful to it too. OK, mine is at: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=157425 I added some additional notes coming from an ethereal capture session, and it really seems a thunderbird problem. Cheers From mbneto at gmail.com Thu May 12 14:25:42 2005 From: mbneto at gmail.com (mbneto) Date: Thu, 12 May 2005 10:25:42 -0400 Subject: yum upgrade errors : %postun Message-ID: <5cf776b8050512072565bfa536@mail.gmail.com> Hi, it has been a while that i can't yum upgrade because of errors. All rpms installed complain about %postrun errors. No clue so far of what should I do to correct it. From skvidal at phy.duke.edu Thu May 12 14:32:24 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Thu, 12 May 2005 10:32:24 -0400 Subject: yum upgrade errors : %postun In-Reply-To: <5cf776b8050512072565bfa536@mail.gmail.com> References: <5cf776b8050512072565bfa536@mail.gmail.com> Message-ID: <1115908344.4604.42.camel@cutter> On Thu, 2005-05-12 at 10:25 -0400, mbneto wrote: > Hi, > > it has been a while that i can't yum upgrade because of errors. > > All rpms installed complain about %postrun errors. > > No clue so far of what should I do to correct it. You need to cite which packages are causing the errors. -sv From mfioretti at mclink.it Thu May 12 15:08:11 2005 From: mfioretti at mclink.it (M. Fioretti) Date: Thu, 12 May 2005 17:08:11 +0200 Subject: Converting UNO packages to .rpm format? In-Reply-To: <28163.192.54.193.28.1115905835.squirrel@rousalka.dyndns.org> References: <20050512115652.GP3286@mclink.it> <28163.192.54.193.28.1115905835.squirrel@rousalka.dyndns.org> Message-ID: <20050512150811.GR3286@mclink.it> On Thu, May 12, 2005 15:50:35 PM +0200, Nicolas Mailhot (nicolas.mailhot at laposte.net) wrote: > > On Jeu 12 mai 2005 13:56, M. Fioretti a ?crit : > > Greetings, > > > > there is a discussion going on at discuss at openoffice.org about how, in > > general, OO.o add-ons should be packaged for Linux: UNO format, > > .rpm/deb/ etc... and who should do it. > > Actually the oo.o developpers asked to move it to > dev at udk.openoffice.org so new messages should go there. I know, but I asked here because I wanted to know if some Fedora developer already has done this. In any case yes, the more Linux packagers participate (and steer in a linux-friendly way, packaging wise) to the dev at udk.openoffice.org discussion, the better. Nicolas or anybody else ho has anything to tell is still welcome to contact me off-list, as I'm writing an article on this issue. Yes, I forgot to say it in the initial message, sorry. Ciao, Marco F. -- Marco Fioretti mfioretti, at the server mclink.it Fedora Core 3 for low memory http://www.rule-project.org/ Real leaders are ordinary people with extraordinary determination From i.pilcher at comcast.net Thu May 12 15:36:33 2005 From: i.pilcher at comcast.net (Ian Pilcher) Date: Thu, 12 May 2005 10:36:33 -0500 Subject: Whither nedit? Message-ID: I just noticed that nedit is no longer present in Fedora Core 4 test 3. Is there a plan to bring it back or move it to Fedora Extras? (I would gladly volunteer to maintain a Fedora Extras package.) -- ======================================================================== Ian Pilcher i.pilcher at comcast.net ======================================================================== From mattdm at mattdm.org Thu May 12 15:55:04 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Thu, 12 May 2005 11:55:04 -0400 Subject: enhance security via private TMP/TMPDIR by default Message-ID: <20050512155504.GA25624@jadzia.bu.edu> Just a thought: /etc/profile.d/tmpdir.sh: # For privacy and security, set temporary directories to ~/tmp on local # filesystems -- but for network filesystems, make up one in /tmp. # The trap statement cleans up the temporary directory if it's empty on exit. if [ -z "$TMP" ]; then if [ $EUID -ne 0 ] && [ -d $HOME/tmp ] && [ $( df --local $HOME | wc -l ) -gt 1 ]; then TMP=$HOME/tmp else TMP=$( mktemp -d /tmp/${HOSTNAME}.tmp.XXXXXXXX ) || exit 1 trap "rmdir --ignore-fail-on-non-empty $TMP" EXIT fi fi if [ -z "$TMPDIR" ]; then TMPDIR=$TMP fi -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 82 degrees Fahrenheit. From ivazquez at ivazquez.net Thu May 12 15:59:28 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Thu, 12 May 2005 11:59:28 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <20050512155504.GA25624@jadzia.bu.edu> References: <20050512155504.GA25624@jadzia.bu.edu> Message-ID: <1115913568.26688.5.camel@ignacio.ignacio.lan> On Thu, 2005-05-12 at 11:55 -0400, Matthew Miller wrote: > Just a thought: > > > /etc/profile.d/tmpdir.sh: > > # For privacy and security, set temporary directories to ~/tmp on local > # filesystems -- but for network filesystems, make up one in /tmp. > # The trap statement cleans up the temporary directory if it's empty on exit. Why not use /tmp/ instead? -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From walters at redhat.com Thu May 12 16:03:28 2005 From: walters at redhat.com (Colin Walters) Date: Thu, 12 May 2005 12:03:28 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <20050512155504.GA25624@jadzia.bu.edu> References: <20050512155504.GA25624@jadzia.bu.edu> Message-ID: <1115913808.3424.5.camel@nexus.verbum.private> On Thu, 2005-05-12 at 11:55 -0400, Matthew Miller wrote: > Just a thought: > > > /etc/profile.d/tmpdir.sh: > > # For privacy and security, set temporary directories to ~/tmp on local There's actually been some work going on on giving each user their own /tmp namespace via the kernel's CLONE_NEWNS capability and a PAM module, AIUI. To the system administrator this could appear as /tmp/. I think the problem is in getting later mounts to actually appear in the cloned namespace. From mattdm at mattdm.org Thu May 12 16:04:28 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Thu, 12 May 2005 12:04:28 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1115913568.26688.5.camel@ignacio.ignacio.lan> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913568.26688.5.camel@ignacio.ignacio.lan> Message-ID: <20050512160428.GA25963@jadzia.bu.edu> On Thu, May 12, 2005 at 11:59:28AM -0400, Ignacio Vazquez-Abrams wrote: > > # For privacy and security, set temporary directories to ~/tmp on local > > # filesystems -- but for network filesystems, make up one in /tmp. > > # The trap statement cleans up the temporary directory if it's empty on > Why not use /tmp/ instead? Because avoiding using predictable filenames is one of the advantages. -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 82 degrees Fahrenheit. From mattdm at mattdm.org Thu May 12 16:05:44 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Thu, 12 May 2005 12:05:44 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1115913808.3424.5.camel@nexus.verbum.private> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> Message-ID: <20050512160544.GB25963@jadzia.bu.edu> On Thu, May 12, 2005 at 12:03:28PM -0400, Colin Walters wrote: > > # For privacy and security, set temporary directories to ~/tmp on local > There's actually been some work going on on giving each user their > own /tmp namespace via the kernel's CLONE_NEWNS capability and a PAM > module, AIUI. To the system administrator this could appear > as /tmp/. I think the problem is in getting later mounts to > actually appear in the cloned namespace. Ooh, that's kinda cool. This is easier, though. :) -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 82 degrees Fahrenheit. From sundaram at redhat.com Thu May 12 16:19:38 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Thu, 12 May 2005 21:49:38 +0530 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1115913808.3424.5.camel@nexus.verbum.private> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> Message-ID: <4283821A.4010008@redhat.com> Colin Walters wrote: >On Thu, 2005-05-12 at 11:55 -0400, Matthew Miller wrote: > > >>Just a thought: >> >> >>/etc/profile.d/tmpdir.sh: >> >># For privacy and security, set temporary directories to ~/tmp on local >> >> > >There's actually been some work going on on giving each user their >own /tmp namespace via the kernel's CLONE_NEWNS capability and a PAM >module, AIUI. To the system administrator this could appear >as /tmp/. I think the problem is in getting later mounts to >actually appear in the cloned namespace. > > > We probably should be putting out such information on a roadmap page for Fedora instead of waiting for related topics to pop out. If we cant get a general roadmap out, we can probably look at short term goals for every Fedora version in a wiki. will this work out? regards Rahul From johnp at redhat.com Thu May 12 16:34:44 2005 From: johnp at redhat.com (John (J5) Palmieri) Date: Thu, 12 May 2005 12:34:44 -0400 Subject: ElementTree vs. lxml (python XML libraries) Message-ID: <1115915684.3351.10.camel@remedyz.boston.redhat.com> This is just an exploratory e-mail as I don't know all the issues involved and don't want to get into a flamewar if someone has strong opinions on either side. I have been looking at what library to use for parsing the XML content in the dbus python bindings. Suggestions were to use lxml (http://codespeak.net/lxml/) however we don't currently ship it. Yum currently uses ElementTree for its parsing. lxml aims to be compatible with the ElementTree API with a few exceptions. On top of that it uses libxml2 as its base library and extends the ElementTree API with things like XPath, Relax NG, XSLT and c14n. While the dbus bindings don't need these features some of the other stuff I want to work on may. Is it a possibility to port yum to lxml? Are we married to ElementTree? Would having both libraries in the distribution be objectionable? Does anyone else have a deeper knowledge of either or both libraries that can shed more light on the issue. -- John (J5) Palmieri Associate Software Engineer Desktop Group Red Hat, Inc. Blog: http://martianrock.com From rhally at mindspring.com Thu May 12 16:07:59 2005 From: rhally at mindspring.com (Richard Hally) Date: Thu, 12 May 2005 12:07:59 -0400 Subject: yum very slow In-Reply-To: <4282C69A.4030605@mindspring.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> <604aa791050511061263ff1b56@mail.gmail.com> <4282C69A.4030605@mindspring.com> Message-ID: <42837F5F.5050501@mindspring.com> sorry for the reply to myself. for the record, Problem solved: see below. Richard Hally wrote: > Jeff Spaleta wrote: > >> On 5/11/05, Richard Hally wrote: >> >> >>> Excluding Packages in global exclude list >>> Finished >>> (wait an hour + ) >>> Resolving Dependencies >>> --> Populating transaction set with selected packages.. >>> >>> During the wait, there is continuous disk activity. and no network >>> activity. >>> This has been going on for a while (it didn't just start today) >>> I did a yum clean all to see if it would help but it did not. >>> >> >> >> that's fascinatingly different than what I'm seeing. How much memory >> do you have.. and how much memory is the yum process using during that >> (wait an hour+) I'm definitely not experiencing this on my rawhide >> box. >> >> -jef >> >> >> > Below is some of the output with -d 10 option specified. I may have > been mistaken when I originally said the wait was while resolving > dependencies. The long wait occurs immediately *before* the "Resolving > Dependencies" is printed. Once the "Resolving Dependencies" is > printed, the header downloads and the rest of the update continue > normally. > after removing primary.xml.gz, primary.xml.gz.sqlite and repomd.xml from /var/cache/yum/development the problem has been solved. It looks like the new primary.xml.gz.sqlite file that was created is substantially small (6.6mb vs 7.2mb) while the other recreated files are exactly the same size. thanks for all the help Richard Hally From remco at rvt.com Thu May 12 17:23:35 2005 From: remco at rvt.com (Remco Treffkorn) Date: Thu, 12 May 2005 10:23:35 -0700 Subject: any help on NISTnet emulators In-Reply-To: <20050512110912.84492.qmail@web54009.mail.yahoo.com> References: <20050512110912.84492.qmail@web54009.mail.yahoo.com> Message-ID: <200505121023.35961.remco@rvt.com> On Thursday 12 May 2005 04:09, avni thacker wrote: > a small problem have any body of u heard of NISTnet emulator? Who ever had > heard pls let me know how to configure it, i m using FC1 and what all > changes would be required > A google on "nistnet" gives you everything you can possibly want. -- Remco Treffkorn (RT445) HAM DC2XT remco at rvt.com (831) 685-1201 From jbuell at vmware.com Thu May 12 17:44:26 2005 From: jbuell at vmware.com (Jeffrey Buell) Date: Thu, 12 May 2005 10:44:26 -0700 Subject: rpmbuild problem in FC4 release 1290 Message-ID: <58743620D2C0D9439C627C064581E25218D72F@PA-ECLUSTER2.vmware.com> > > On Wed, 2005-05-11 at 13:08 -0700, Jeffrey Buell wrote: > > I have used rpmbuild several times with recent releases, > but have never seen > > this problem before. I'm using release 1290 (1288 did > something similar). > > The tail end of the output of > > > > rpmbuild -bp --target=i686 kernel-2.6.spec > > > which architecture is the host ? i686. I just figured out the workaround: removing SOURCES/kernel-2.6.11-i586-smp.config. Apparently I don't need this, but still rpmbuild shouldn't choke on it. Jeff From mricon at gmail.com Thu May 12 17:55:22 2005 From: mricon at gmail.com (Konstantin Ryabitsev) Date: Thu, 12 May 2005 13:55:22 -0400 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: <1115915684.3351.10.camel@remedyz.boston.redhat.com> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> Message-ID: On 5/12/05, John (J5) Palmieri wrote: > Is it a possibility to port yum to lxml? Are we married to ElementTree? > Would having both libraries in the distribution be objectionable? Does > anyone else have a deeper knowledge of either or both libraries that can > shed more light on the issue. Yum is not married to ElementTree, it's married to ${FASTEST_XML_PARSER_FOR_PYTHON}. However, I *really* like the ElementTree API, so if lxml can a) ship with the same API, and b) be faster than ElementTree, I don't see a reason to stay with ElementTree. Regards, -- Konstantin Ryabitsev Zlotniks, INC "? ????? ???? ?? ??? ???? ?????????????." --???? From johnp at redhat.com Thu May 12 18:18:32 2005 From: johnp at redhat.com (John (J5) Palmieri) Date: Thu, 12 May 2005 14:18:32 -0400 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> Message-ID: <1115921912.3351.26.camel@remedyz.boston.redhat.com> On Thu, 2005-05-12 at 13:55 -0400, Konstantin Ryabitsev wrote: > On 5/12/05, John (J5) Palmieri wrote: > > Is it a possibility to port yum to lxml? Are we married to ElementTree? > > Would having both libraries in the distribution be objectionable? Does > > anyone else have a deeper knowledge of either or both libraries that can > > shed more light on the issue. > > Yum is not married to ElementTree, it's married to > ${FASTEST_XML_PARSER_FOR_PYTHON}. However, I *really* like the > ElementTree API, so if lxml can a) ship with the same API, and b) be > faster than ElementTree, I don't see a reason to stay with > ElementTree. >From what I heard from people I respect in the python community lxml is a lot faster (by virtue of using libxml2 as its backend) but I have yet to see any real data on the issue. Luckily the API is so similar it will just require a couple of tweaks like changing the name of the imported package to test. I'll investigate further. -- John (J5) Palmieri Associate Software Engineer Desktop Group Red Hat, Inc. Blog: http://martianrock.com From skvidal at phy.duke.edu Thu May 12 18:24:21 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Thu, 12 May 2005 14:24:21 -0400 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: <1115921912.3351.26.camel@remedyz.boston.redhat.com> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> <1115921912.3351.26.camel@remedyz.boston.redhat.com> Message-ID: <1115922261.13475.16.camel@cutter> > >From what I heard from people I respect in the python community lxml is > a lot faster (by virtue of using libxml2 as its backend) but I have yet > to see any real data on the issue. Luckily the API is so similar it > will just require a couple of tweaks like changing the name of the > imported package to test. I'll investigate further. > Well cElementTree, which is really what yum is using, has tested out faster for the yum use case than libxml2. Like about 3 times faster. -sv From skvidal at phy.duke.edu Thu May 12 18:30:10 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Thu, 12 May 2005 14:30:10 -0400 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: <1115915684.3351.10.camel@remedyz.boston.redhat.com> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> Message-ID: <1115922610.13475.20.camel@cutter> > I have been looking at what library to use for parsing the XML content > in the dbus python bindings. Suggestions were to use lxml > (http://codespeak.net/lxml/) however we don't currently ship it. Yum > currently uses ElementTree for its parsing. lxml aims to be compatible > with the ElementTree API with a few exceptions. On top of that it uses > libxml2 as its base library and extends the ElementTree API with things > like XPath, Relax NG, XSLT and c14n. While the dbus bindings don't need > these features some of the other stuff I want to work on may. Oh - one thing tangentially related - you've seen the simple dbus event notifier that nasrat wrote for doing this, right? https://www.redhat.com/archives/fedora-config-list/2005-March/msg00008.html instead of parsing it on its own it lets yum do the work for it and send the events that way. -sv From johnp at redhat.com Thu May 12 19:11:09 2005 From: johnp at redhat.com (John (J5) Palmieri) Date: Thu, 12 May 2005 15:11:09 -0400 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: <1115922610.13475.20.camel@cutter> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> <1115922610.13475.20.camel@cutter> Message-ID: <1115925069.3351.31.camel@remedyz.boston.redhat.com> On Thu, 2005-05-12 at 14:30 -0400, seth vidal wrote: > > I have been looking at what library to use for parsing the XML content > > in the dbus python bindings. Suggestions were to use lxml > > (http://codespeak.net/lxml/) however we don't currently ship it. Yum > > currently uses ElementTree for its parsing. lxml aims to be compatible > > with the ElementTree API with a few exceptions. On top of that it uses > > libxml2 as its base library and extends the ElementTree API with things > > like XPath, Relax NG, XSLT and c14n. While the dbus bindings don't need > > these features some of the other stuff I want to work on may. > > Oh - one thing tangentially related - you've seen the simple dbus event > notifier that nasrat wrote for doing this, right? > > https://www.redhat.com/archives/fedora-config-list/2005-March/msg00008.html > > instead of parsing it on its own it lets yum do the work for it and send > the events that way. > Looks cool. I'm actually trying to parse introspect data not yum data it is just yum uses an xml library and it would be nice if both dbus and yum used the same one. Hmm, I should send a patch to that to Paul to keep up to date with the new bindings. Looking further into it it seems that yum uses some lowlevel stuff (iterparse) that is not emulated by lxml. -- John (J5) Palmieri Associate Software Engineer Desktop Group Red Hat, Inc. Blog: http://martianrock.com From skvidal at phy.duke.edu Thu May 12 19:13:18 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Thu, 12 May 2005 15:13:18 -0400 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: <1115925069.3351.31.camel@remedyz.boston.redhat.com> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> <1115922610.13475.20.camel@cutter> <1115925069.3351.31.camel@remedyz.boston.redhat.com> Message-ID: <1115925198.13475.32.camel@cutter> On Thu, 2005-05-12 at 15:11 -0400, John (J5) Palmieri wrote: > On Thu, 2005-05-12 at 14:30 -0400, seth vidal wrote: > > > I have been looking at what library to use for parsing the XML content > > > in the dbus python bindings. Suggestions were to use lxml > > > (http://codespeak.net/lxml/) however we don't currently ship it. Yum > > > currently uses ElementTree for its parsing. lxml aims to be compatible > > > with the ElementTree API with a few exceptions. On top of that it uses > > > libxml2 as its base library and extends the ElementTree API with things > > > like XPath, Relax NG, XSLT and c14n. While the dbus bindings don't need > > > these features some of the other stuff I want to work on may. > > > > Oh - one thing tangentially related - you've seen the simple dbus event > > notifier that nasrat wrote for doing this, right? > > > > https://www.redhat.com/archives/fedora-config-list/2005-March/msg00008.html > > > > instead of parsing it on its own it lets yum do the work for it and send > > the events that way. > > > > Looks cool. I'm actually trying to parse introspect data not yum data > it is just yum uses an xml library and it would be nice if both dbus and > yum used the same one. Hmm, I should send a patch to that to Paul to > keep up to date with the new bindings. > > Looking further into it it seems that yum uses some lowlevel stuff > (iterparse) that is not emulated by lxml. > if you don't use iterparse your memory size EXPLODES. -sv From jbuell at vmware.com Thu May 12 19:44:39 2005 From: jbuell at vmware.com (Jeffrey Buell) Date: Thu, 12 May 2005 12:44:39 -0700 Subject: FC4 release 1290 build fails with ARCH=xen Message-ID: <58743620D2C0D9439C627C064581E25218D730@PA-ECLUSTER2.vmware.com> The regular FC4 release 1290 build is fine on my i686 box, but building with ARCH=xen (with the xen0 config file included with the source) fails. The file asm/synch_bitops.h does not exist but is needed by asm-xen/evtchn.h. Similarly, asm/hypercall.h does not exist but is needed by asm-xen/hypervisor.h. Header files with these names are in asm-xen/asm-i386 but using these leads to compile errors involving the structs evtchn_op_t and physdev_op_t. Is this a known problem and are there any workarounds? The only change I made was to remove SOURCES/kernel-2.6.11-i586-smp.config in order to get rpmbuild to work. Jeff From jspaleta at gmail.com Thu May 12 19:50:07 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Thu, 12 May 2005 15:50:07 -0400 Subject: yum very slow In-Reply-To: <42837F5F.5050501@mindspring.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> <604aa791050511061263ff1b56@mail.gmail.com> <4282C69A.4030605@mindspring.com> <42837F5F.5050501@mindspring.com> Message-ID: <604aa79105051212502040019f@mail.gmail.com> On 5/12/05, Richard Hally wrote: > after removing primary.xml.gz, primary.xml.gz.sqlite and repomd.xml from > /var/cache/yum/development the problem has been solved. It looks like > the new primary.xml.gz.sqlite file that was created is substantially > small (6.6mb vs 7.2mb) while the other recreated files are exactly the > same size. Ah, man .... you didn't keep the old version of that primary.xml.gz.sqlite file did you? Now that you have narrowed down the problem.. it would have been interesting to poke at that file to see what was going wrong. -jef From davej at redhat.com Thu May 12 19:52:03 2005 From: davej at redhat.com (Dave Jones) Date: Thu, 12 May 2005 15:52:03 -0400 Subject: FC4 release 1290 build fails with ARCH=xen In-Reply-To: <58743620D2C0D9439C627C064581E25218D730@PA-ECLUSTER2.vmware.com> References: <58743620D2C0D9439C627C064581E25218D730@PA-ECLUSTER2.vmware.com> Message-ID: <20050512195202.GA17005@redhat.com> On Thu, May 12, 2005 at 12:44:39PM -0700, Jeffrey Buell wrote: > The regular FC4 release 1290 build is fine on my i686 box, but building with > ARCH=xen (with the xen0 config file included with the source) fails. The > file asm/synch_bitops.h does not exist but is needed by asm-xen/evtchn.h. > Similarly, asm/hypercall.h does not exist but is needed by > asm-xen/hypervisor.h. Header files with these names are in asm-xen/asm-i386 > but using these leads to compile errors involving the structs evtchn_op_t and > physdev_op_t. Is this a known problem and are there any workarounds? The > only change I made was to remove SOURCES/kernel-2.6.11-i586-smp.config in > order to get rpmbuild to work. Yes its known (Read the changelog in the spec). It's being worked on. Dave From veillard at redhat.com Thu May 12 20:01:07 2005 From: veillard at redhat.com (Daniel Veillard) Date: Thu, 12 May 2005 16:01:07 -0400 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: <1115915684.3351.10.camel@remedyz.boston.redhat.com> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> Message-ID: <20050512200106.GS14726@redhat.com> On Thu, May 12, 2005 at 12:34:44PM -0400, John (J5) Palmieri wrote: > This is just an exploratory e-mail as I don't know all the issues > involved and don't want to get into a flamewar if someone has strong > opinions on either side. > > I have been looking at what library to use for parsing the XML content > in the dbus python bindings. Suggestions were to use lxml why not libxml2 python bindings directly ? At least if you have a problem I know the full stack there. Daniel -- Daniel Veillard | Red Hat Desktop team http://redhat.com/ veillard at redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/ From rhally at mindspring.com Thu May 12 20:08:19 2005 From: rhally at mindspring.com (Richard Hally) Date: Thu, 12 May 2005 16:08:19 -0400 Subject: yum very slow In-Reply-To: <604aa79105051212502040019f@mail.gmail.com> References: <4281A8F6.1070503@mindspring.com> <1115795716.10234.130.camel@cutter> <4281C547.5090506@mindspring.com> <604aa791050511061263ff1b56@mail.gmail.com> <4282C69A.4030605@mindspring.com> <42837F5F.5050501@mindspring.com> <604aa79105051212502040019f@mail.gmail.com> Message-ID: <4283B7B3.3080206@mindspring.com> Jeff Spaleta wrote: >On 5/12/05, Richard Hally wrote: > > >>after removing primary.xml.gz, primary.xml.gz.sqlite and repomd.xml from >>/var/cache/yum/development the problem has been solved. It looks like >>the new primary.xml.gz.sqlite file that was created is substantially >>small (6.6mb vs 7.2mb) while the other recreated files are exactly the >>same size. >> >> > >Ah, man .... you didn't keep the old version of that >primary.xml.gz.sqlite file did you? >Now that you have narrowed down the problem.. it would have been >interesting to poke at that file to see what was going wrong. > >-jef > > > yup, I kept it = 8-) it is 7218176 bytes. Richard From nicolas.mailhot at laposte.net Thu May 12 13:50:35 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Thu, 12 May 2005 15:50:35 +0200 (CEST) Subject: Converting UNO packages to .rpm format? In-Reply-To: <20050512115652.GP3286@mclink.it> References: <20050512115652.GP3286@mclink.it> Message-ID: <28163.192.54.193.28.1115905835.squirrel@rousalka.dyndns.org> On Jeu 12 mai 2005 13:56, M. Fioretti a ?crit : > Greetings, > > there is a discussion going on at discuss at openoffice.org about how, in > general, OO.o add-ons should be packaged for Linux: UNO format, > .rpm/deb/ etc... and who should do it. Actually the oo.o developpers asked to move it to dev at udk.openoffice.org so new messages should go there. Regards, -- Nicolas Mailhot From shahms at shahms.com Thu May 12 20:59:04 2005 From: shahms at shahms.com (Shahms King) Date: Thu, 12 May 2005 13:59:04 -0700 Subject: ElementTree vs. lxml (python XML libraries) In-Reply-To: <20050512200106.GS14726@redhat.com> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> <20050512200106.GS14726@redhat.com> Message-ID: <4283C398.4000802@shahms.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Daniel Veillard wrote: | On Thu, May 12, 2005 at 12:34:44PM -0400, John (J5) Palmieri wrote: | |>This is just an exploratory e-mail as I don't know all the issues |>involved and don't want to get into a flamewar if someone has strong |>opinions on either side. |> |>I have been looking at what library to use for parsing the XML content |>in the dbus python bindings. Suggestions were to use lxml | | | why not libxml2 python bindings directly ? At least if you have a | problem I know the full stack there. | | Daniel | This is just a summary from the last time this dicussion came up, some of the details may be incorrect. I happen to like libxml2 and would try to fix these problems "If I Only Had The Time". 1. The bindings are very un-pythonic. The biggest wart off the top of ~ my head is that you have to manually free the document. The lxml ~ wrapper is an attempt to fix this problem. 2. "They are slow". As much as libxml2 is usually faster than expat, ~ but something about how the python bindings deal with strings makes ~ libxml2 slower in this case. There may have been a third, but those two are the biggest problems that I remember. As for lxml not implementing the "iterparse" interface, I'd need to look into it more, but I'm pretty sure lxml could use libxml's XmlTextReader interface to implement this... - -- Shahms E. King Multnomah ESD Public Key: http://shahms.mesd.k12.or.us/~sking/shahms.asc Fingerprint: 1612 054B CE92 8770 F1EA AB1B FEAB 3636 45B2 D75B -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFCg8OY/qs2NkWy11sRArYQAKCTxp1tu7pF4BJm5HX8VImXwX1X2ACglyX8 ChyrWxymMEGmo32oH1ElHV0= =hbGR -----END PGP SIGNATURE----- From jbuell at vmware.com Thu May 12 20:57:57 2005 From: jbuell at vmware.com (Jeffrey Buell) Date: Thu, 12 May 2005 13:57:57 -0700 Subject: FC4 release 1290 build fails with ARCH=xen Message-ID: <58743620D2C0D9439C627C064581E25218D732@PA-ECLUSTER2.vmware.com> > On Thu, May 12, 2005 at 12:44:39PM -0700, Jeffrey Buell wrote: > > The regular FC4 release 1290 build is fine on my i686 box, > but building with > > ARCH=xen (with the xen0 config file included with the > source) fails. The > > file asm/synch_bitops.h does not exist but is needed by > asm-xen/evtchn.h. > > Similarly, asm/hypercall.h does not exist but is needed by > > asm-xen/hypervisor.h. Header files with these names are > in asm-xen/asm-i386 > > but using these leads to compile errors involving the > structs evtchn_op_t and > > physdev_op_t. Is this a known problem and are there any > workarounds? The > > only change I made was to remove > SOURCES/kernel-2.6.11-i586-smp.config in > > order to get rpmbuild to work. > > Yes its known (Read the changelog in the spec). > It's being worked on. > > Dave > My bad. Thanks for the pointer, Dave. From johnp at redhat.com Thu May 12 21:20:05 2005 From: johnp at redhat.com (John (J5) Palmieri) Date: Thu, 12 May 2005 17:20:05 -0400 Subject: Resolution for now (Re: ElementTree vs. lxml (python XML libraries)) In-Reply-To: <1115915684.3351.10.camel@remedyz.boston.redhat.com> References: <1115915684.3351.10.camel@remedyz.boston.redhat.com> Message-ID: <1115932805.3351.50.camel@remedyz.boston.redhat.com> After talking to a bunch of people and reading the responses here is my take on things: * I have made the decision for right now to go with the libxml2 bindings or perhaps put the parsing in the C layer and use libxml2 directly. The reason for this is simple - most all distros have libxml2 and in the end dbus just needs a fast parser that everyone uses. * The original thought was to make lxml the default if it was ready and a drop in replacement for yum (there really should be one defacto way to parse XML documents in Python or at least one clear winner). This however is not true yet though it may be in the future so I am going to go ahead and package lxml for extras during the FC5 cycle and start playing with it, offering help where I can. * I have other projects I want to work on in my free time that can use the enhanced functionality and nice Pythonic API that lxml gives. * If at one point we can swap out ElementTree in yum and there is no loss in performance we have the option to do so. Thanks for everyones input. On Thu, 2005-05-12 at 12:34 -0400, John (J5) Palmieri wrote: > This is just an exploratory e-mail as I don't know all the issues > involved and don't want to get into a flamewar if someone has strong > opinions on either side. > > I have been looking at what library to use for parsing the XML content > in the dbus python bindings. Suggestions were to use lxml > (http://codespeak.net/lxml/) however we don't currently ship it. Yum > currently uses ElementTree for its parsing. lxml aims to be compatible > with the ElementTree API with a few exceptions. On top of that it uses > libxml2 as its base library and extends the ElementTree API with things > like XPath, Relax NG, XSLT and c14n. While the dbus bindings don't need > these features some of the other stuff I want to work on may. > > Is it a possibility to port yum to lxml? Are we married to ElementTree? > Would having both libraries in the distribution be objectionable? Does > anyone else have a deeper knowledge of either or both libraries that can > shed more light on the issue. > > -- > John (J5) Palmieri > Associate Software Engineer > Desktop Group > Red Hat, Inc. > Blog: http://martianrock.com > -- John (J5) Palmieri Associate Software Engineer Desktop Group Red Hat, Inc. Blog: http://martianrock.com From moe at blagblagblag.org Fri May 13 02:17:33 2005 From: moe at blagblagblag.org (jeff) Date: Thu, 12 May 2005 20:17:33 -0600 Subject: 3w-9xxx module version in FC4 In-Reply-To: <1114708461.6151.171.camel@jkeating2.hq.pogolinux.com> References: <1114635647.6151.116.camel@jkeating2.hq.pogolinux.com> <1114694983.3265.11.camel@dhollis-lnx.sunera.com> <42711410.8020207@blagblagblag.org> <1114708461.6151.171.camel@jkeating2.hq.pogolinux.com> Message-ID: <42840E3D.8030509@blagblagblag.org> Jesse Keating wrote: > On Thu, 2005-04-28 at 10:49 -0600, jeff wrote: > >>Do either of these versions address the "Extremely high iowait with >>3Ware array and moderate disk activity" bug? It's been open for just >>over a year & has 259 comments. >> >>https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=121434 >> >>I can confirm this bug with v2.26.02.001, fwiw. > > > Most likely not as much as you'd like. That unfortunately is an > artifact of 3ware's design. We never use these systems for high usage > scenarios like a database server or sometimes even a home directory > server. Nearline backup and slow storage is what we consider them > useful for. The '9.2' release does have significant performance > increases (firmware and driver need to be updated) but I don't think > they'll solve the IO Wait state any time soon. I can't benchmark it since it's a live system, but the following helped with the iowait problem under FC2: echo 512 > /sys/block/sda/queue/nr_requests It was suggested on LKML that nr_requests be double queue_depth. I get much more even performance now. It no longer bogs down & goes out to lunch when I throw a DVD ISO at it, for example. I also use: blockdev --setra 16384 /dev/sda -Jeff From rodd at clarkson.id.au Fri May 13 03:09:17 2005 From: rodd at clarkson.id.au (Rodd Clarkson) Date: Fri, 13 May 2005 13:09:17 +1000 Subject: Whither nedit? In-Reply-To: References: Message-ID: <1115953757.4724.7.camel@jellyfish.redfishdemo.com> Ian, > I just noticed that nedit is no longer present in Fedora Core 4 test 3. > Is there a plan to bring it back or move it to Fedora Extras? (I would > gladly volunteer to maintain a Fedora Extras package.) If you read the RELEASE_NOTES_en you'll notice that nedit is listed in Packages that have be moved out of core. A quick look at ftp://download.fedora.redhat.com/pub/fedora/linux/extras/development/i386 shows that nedit isn't in extras (yet) so I'm sure you would be most welcome to offer to maintain it in extras. You'll need to sponser the package, but I'm not sure on the mechanics of how this works. Hopefully someone will point you in the right direction. R. -- "It's a fine line between denial and faith. It's much better on my side" From nphilipp at redhat.com Fri May 13 10:20:21 2005 From: nphilipp at redhat.com (Nils Philippsen) Date: Fri, 13 May 2005 12:20:21 +0200 Subject: rawhide report: 20050510 changes In-Reply-To: <1115739006.30474.94.camel@localhost.localdomain> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> <1115739006.30474.94.camel@localhost.localdomain> Message-ID: <1115979622.3448.10.camel@gibraltar.stuttgart.redhat.com> On Tue, 2005-05-10 at 11:30 -0400, Owen Taylor wrote: > On Tue, 2005-05-10 at 08:26 -0400, Jakub Jelinek wrote: > > > a) on x86-64 this is unnecessary, all x86-64's have -msse2 by default > > b) -msse2 implies -msse > > c) if you use -msse2 in CFLAGS for all files, you can't run the latest > > GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. > > -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, > > and GIMP should make sure that no routine from those sources will be > > ever called on pre-SSE2 chips > > This sounds like a big pain ... shouldn't there be a way to say > "use sse2/sse only for builtins" ? > > Without that, there is no way to use builtins in single functions; you > need to have one file for sse2, one for sse, one for mmx, and do the > detection of the current processor somewhere else entirely. I concur, with gcc <= 4.0.0-2 it was possible to use MMX/SSE calls in inline assembly without having to use -mmmx/-msse/-msse2. In order to build the gimp so that it uses MMX/SSE where available, I had to resort to serious autofoo munging because it doesn't allow for setting per-object compiler flags. Nils -- Nils Philippsen / Red Hat / nphilipp at redhat.com "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- B. Franklin, 1759 PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011 From buildsys at redhat.com Fri May 13 11:56:40 2005 From: buildsys at redhat.com (Build System) Date: Fri, 13 May 2005 07:56:40 -0400 Subject: rawhide report: 20050513 changes Message-ID: <200505131156.j4DBueR6026468@porkchop.devel.redhat.com> Updated Packages: dump-0.4b40-2 ------------- * Wed May 11 2005 Jindrich Novy 0.4b40-2 - Don't strip binaries to get valid debuginfo eclipse-1:3.1.0_fc-0.M6.16 -------------------------- * Mon May 09 2005 Ben Konrath 3.1.0_fc-0.M6.16 - Add Requires junit >= 3.8.1-3jpp_4fc to JDT. - Add -g to gcj calls. * Thu May 05 2005 Andrew Overholt 3.1.0_fc-0.M6.15 - Rebuild with new gjdoc (rh#152049). * Wed May 04 2005 Ben Konrath - Re-enable jdt.ui/jdt.jar.so and require gcj 4.0.0-2 (rh#151296). eclipse-cdt-1:3.0.0_fc-0.M6.6 ----------------------------- * Wed May 11 2005 Ben Konrath 3.0.0_fc-0.M6.6 - Temporarily disable org.eclipse.cdt.managedbuilder.core_3.0.0/libmngbuildcore.jar.so. eject-2.0.13-15 --------------- * Thu May 12 2005 Than Ngo 2.0.13-15 - add better translation for zh_TW #157519, thanks to Wei-Lun Chao gimp-2:2.2.7-2 -------------- * Wed May 11 2005 Nils Philippsen - use -mmmx/sse/sse2/... only for the relevant source files so that extended instruction sets only get used on suitable CPUs (#157409) * Mon May 09 2005 Nils Philippsen - version 2.2.7, fixes bug in SSE2 assembly for Lighten Only layer mode (#145771) and various other bugs - on x86 and x86_64, use -msse and -msse2 to accomodate newer compilers * Wed Apr 27 2005 Jeremy Katz - 2:2.2.6-2 - silence %post gnome-panel-2.10.1-10 --------------------- * Wed May 11 2005 Mark McLoughlin 2.10.1-10 - Fix "dialogs pop up under panel dialogs" issue (bug #156425) hal-0.5.2-1 ----------- * Thu May 12 2005 David Zeuthen 0.5.2-1 - Update to upstream release 0.5.2 * Wed Apr 27 2005 David Zeuthen 0.5.1-1 - Update to upstream release 0.5.1 * Tue Apr 19 2005 Florian La Roche - exclude usb reqs for mainframe (#154616) hotplug-3:2004_09_23-6 ---------------------- * Tue May 10 2005 Bill Nottingham 3:2004_09_23-6 - fix mistriggering on HWADDRs of sit0, etc (#153669, #157252) htmlview-3.0.0-11 ----------------- * Wed May 11 2005 Warren Togami - 3.0.0-11 - also ignore -remote (#154033) initscripts-8.11-1 ------------------ * Tue May 10 2005 Bill Nottingham 8.11-1 - fix mis-bringup of interfaces due to accidentally matched HWADDR (a.k.a. ONBOOT=no not working) (#153669, #157252) - support automatic relabeling later if rebooted w/o SELinux () - rc.sysinit: fix fixfiles invocation (#157182) - btmp should be 0600 (#156900) - translation updates: fr, bg, ru, mk, pa, es kernel-2.6.11-1.1303_FC4 ------------------------ * Wed May 11 2005 Dave Jones - Add Ingo's patch to detect soft lockups. - Thread exits siliently via __RESTORE_ALL exception for iret. (#154369) * Wed May 11 2005 David Woodhouse - Import post-rc4 audit fixes from git, including ppc syscall auditing * Wed May 11 2005 Dave Jones - Revert NMI watchdog changes. nc-1.78-2 --------- * Wed May 11 2005 David Woodhouse 1.78-2 - Don't ignore POLLHUP and go into an endless loop (#156835) * Mon Apr 11 2005 Radek Vokal 1.78-1 - update from CVS, using glib functions * Thu Mar 31 2005 Radek Vokal 1.77-1 - switching to new OpenBSD version of netcat pciutils-2.1.99.test8-9 ----------------------- * Tue May 10 2005 Bill Nottingham - 2.1.99.test8-9 - fix debuginfo generation perl-PDL-2.4.1-11 ----------------- * Wed May 11 2005 Jose Pedro Oliveira - 2.4.1-11 - Add missing perl(PDL::Graphics::TriD*) provides. (#156482) - Explicitly filter perl(Tk). (#156482) redhat-artwork-0.122-9 ---------------------- * Tue May 10 2005 John (J5) Palmieri 0.122-9 - Add new stock volume icons squid-7:2.5.STABLE9-6 --------------------- * Wed May 11 2005 Jay Fenlason 7:2.5.STABLE9-6 - More upstream patches, including a fix for bz#157456 CAN-2005-1519 DNS lookups unreliable on untrusted networks * Tue Apr 26 2005 Jay Fenlason 7:2.5.STABLE9-5 - more upstream patches, including a fix for CVE-1999-0710 cachemgr malicious use system-switch-mail-0.5.25-4 --------------------------- * Wed May 11 2005 Than Ngo 0.5.25-4 - fix location for menu item #157173 usermode-1.80-1 --------------- * Wed May 11 2005 Jindrich Novy 1.80-1 - fix "Unknown error" when password is mistyped in userpasswd (#135500) - add icons to windows for usermode-gtk applications (#155867) - add missing checks for some PAM error codes - fix ungettextized error message - update translations From dragoran at feuerpokemon.de Fri May 13 14:15:42 2005 From: dragoran at feuerpokemon.de (dragoran) Date: Fri, 13 May 2005 16:15:42 +0200 Subject: rawhide report: 20050510 changes In-Reply-To: <1115979622.3448.10.camel@gibraltar.stuttgart.redhat.com> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> <1115739006.30474.94.camel@localhost.localdomain> <1115979622.3448.10.camel@gibraltar.stuttgart.redhat.com> Message-ID: <4284B68E.2070102@feuerpokemon.de> Nils Philippsen wrote: >On Tue, 2005-05-10 at 11:30 -0400, Owen Taylor wrote: > > >>On Tue, 2005-05-10 at 08:26 -0400, Jakub Jelinek wrote: >> >> >> >>>a) on x86-64 this is unnecessary, all x86-64's have -msse2 by default >>>b) -msse2 implies -msse >>>c) if you use -msse2 in CFLAGS for all files, you can't run the latest >>> GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. >>> -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, >>> and GIMP should make sure that no routine from those sources will be >>> ever called on pre-SSE2 chips >>> >>> >>This sounds like a big pain ... shouldn't there be a way to say >>"use sse2/sse only for builtins" ? >> >>Without that, there is no way to use builtins in single functions; you >>need to have one file for sse2, one for sse, one for mmx, and do the >>detection of the current processor somewhere else entirely. >> >> > >I concur, with gcc <= 4.0.0-2 it was possible to use MMX/SSE calls in >inline assembly without having to use -mmmx/-msse/-msse2. In order to >build the gimp so that it uses MMX/SSE where available, I had to resort >to serious autofoo munging because it doesn't allow for setting >per-object compiler flags. > >Nils > > why aren't sse/mmx/sse2 enabled by default on x86_64 ? all those chips supports it. From d.jacobfeuerborn at conversis.de Fri May 13 20:56:34 2005 From: d.jacobfeuerborn at conversis.de (Dennis Jacobfeuerborn) Date: Fri, 13 May 2005 22:56:34 +0200 Subject: gnome-panel eats a lot of memory Message-ID: <42851482.6040108@conversis.de> I noticed that gnome-panel eats 25% (=256mb) of my memory which is quite a lot for a simple panel. Does anybody else see this too? I installed FC4Test2 and then upgraded to Rawhide. Regards, Dennis From mbneto at gmail.com Sat May 14 11:48:43 2005 From: mbneto at gmail.com (mbneto) Date: Sat, 14 May 2005 07:48:43 -0400 Subject: yum upgrade errors : %postun In-Reply-To: <1115908344.4604.42.camel@cutter> References: <5cf776b8050512072565bfa536@mail.gmail.com> <1115908344.4604.42.camel@cutter> Message-ID: <5cf776b805051404484608c8a2@mail.gmail.com> All of them... For instance. # yum -y upgrade rpm Setting up Upgrade Process Setting up repositories development 100% |=========================| 1.1 kB 00:00 base 100% |=========================| 1.1 kB 00:00 Reading repository metadata in from local files Resolving Dependencies --> Populating transaction set with selected packages. Please wait. ---> Package rpm.i386 0:4.4.1-18.1 set to be updated --> Running transaction check Dependencies Resolved ============================================================================= Package Arch Version Repository Size ============================================================================= Updating: rpm i386 4.4.1-18.1 development 580 k Transaction Summary ============================================================================= Install 0 Package(s) Update 1 Package(s) Remove 0 Package(s) Total download size: 580 k Downloading Packages: (1/1): rpm-4.4.1-18.1.i38 100% |=========================| 580 kB 00:00 Running Transaction Test warning: rpm-4.4.1-18.1: Header V3 DSA signature: NOKEY, key ID 30c9ecf8 Finished Transaction Test Transaction Test Succeeded Running Transaction error: %pre(rpm-4.4.1-18.1.i386) scriptlet failed, exit status 255 error: install: %pre scriptlet failed (2), skipping rpm-4.4.1-18.1 Updated: rpm.i386 0:4.4.1-18.1 Complete! # rpm -qi rpm Name : rpm Relocations: (not relocatable) Version : 4.4.1 Vendor: Red Hat, Inc. Release : 17 Build Date: Tue 03 May 2005 04:09:54 PM AMT Install Date: Wed 04 May 2005 09:54:31 AM AMT Build Host: decompose.build.redhat.com Group : System Environment/Base Source RPM: rpm-4.4.1-17.src.rpm Size : 1499457 License: GPL Signature : DSA/SHA1, Wed 04 May 2005 01:36:59 AM AMT, Key ID da84cbd430c9ecf8 Packager : Red Hat, Inc. Summary : The RPM package management system. Description : The RPM Package Manager (RPM) is a powerful command line driven package management system capable of installing, uninstalling, verifying, querying, and updating software packages. Each software package consists of an archive of files along with information about the package like its version, a description, etc. - mb On 5/12/05, seth vidal wrote: > On Thu, 2005-05-12 at 10:25 -0400, mbneto wrote: > > Hi, > > > > it has been a while that i can't yum upgrade because of errors. > > > > All rpms installed complain about %postrun errors. > > > > No clue so far of what should I do to correct it. > > You need to cite which packages are causing the errors. > > -sv > > From buildsys at redhat.com Sat May 14 11:58:08 2005 From: buildsys at redhat.com (Build System) Date: Sat, 14 May 2005 07:58:08 -0400 Subject: rawhide report: 20050514 changes Message-ID: <200505141158.j4EBw8VP003562@porkchop.devel.redhat.com> Updated Packages: crypto-utils-2.2-5 ------------------ * Fri May 13 2005 Joe Orton 2.2-5 - genkey(1): fix paths to use /etc/pki devhelp-0.9.3-7 --------------- * Fri May 13 2005 Christopher Aillon 0.9.3-7 - Depend on mozilla 1.7.8 epiphany-1.6.1-3 ---------------- * Fri May 13 2005 Christopher Aillon - 1.6.1-3 - Depend on mozilla 1.7.8 firefox-0:1.0.4-2 ----------------- * Wed May 11 2005 Christopher Aillon 0:1.0.4-2 - Update to 1.0.4 gcc-4.0.0-5 ----------- * Thu May 12 2005 Jakub Jelinek 4.0.0-5 - update from CVS - PRs bootstrap/21403, c++/19203, c++/20723, c++/21352, c/21160, c/21342, c/21502, fortran/19478, fortran/21260, java/19285, java/20309, libffi/21285, libfortran/18958, libfortran/19155, libfortran/20788, libfortran/21471, libstdc++/18604, libstdc++/21238, middle-end/21085, other/21052, target/16925, target/19051, target/21325, target/21397, target/21412, target/21416, tree-optimization/21380 - use INTERNAL_SYSCALL* macros in libgcc_post_upgrade.c, so that the statically linked binary doesn't need TLS (#155701) - implement java.awt.Frame.getBufferStrategy (Thomas Fitzsimmons, #153266) - put C++ const POD arrays with size determined from initializer into .rodata section (PR c++/21454) - allow limited recursive hash tab use (#157308) - fix C++ "storage class specified for" error reporting (PR c++/21495) - fix crossjumping (#157243, PR middle-end/21492) gimp-2:2.2.7-4 -------------- * Fri May 13 2005 Nils Philippsen - fix inline asm of MMX/SSE optimizations instead of using -mmmx and the like * Fri May 13 2005 Nils Philippsen - fix cpuinstructionset patch so that it actually uses CPU-specific optimizations gnome-applets-1:2.10.1-7 ------------------------ * Fri May 13 2005 Ray Strode 1:2.10.1-7 - Don't disable battstat on some platforms (bug 157683). kernel-2.6.11-1.1305_FC4 ------------------------ * Thu May 12 2005 David Woodhouse - Enable CONFIG_ISA on ppc32 to make the RS/6000 user happy. - Update audit patches mdadm-1.9.0-4.fc4 ----------------- mozilla-37:1.7.8-2 ------------------ * Fri May 13 2005 Christopher Aillon 37:1.7.8-2 - Update to 1.7.8 perl-XML-Encoding-1.01-27 ------------------------- * Sat Apr 30 2005 Jose Pedro Oliveira - 1.01-27 - Bring up to date with current Fedora.Extras perl spec template. (#156510) sox-12.17.7-3 ------------- * Thu May 12 2005 Thomas Woerner 12.17.7-3 - fixed bad link for man/man1/rec.1.gz (#154089) - using /usr/include instead of kernel-devel includes thunderbird-0:1.0.2-6 --------------------- * Fri May 13 2005 Christopher Aillon 1.0.2-6 - Change the Exec line in the desktop file to `thunderbird` * Fri May 13 2005 Christopher Aillon 1.0.2-5 - Update pango patche, MOZ_DISABLE_PANGO now works as advertised. * Mon May 09 2005 Christopher Aillon 1.0.2-4 - Add temporary workaround to not create files in the user's $HOME (#149664) xorg-x11-6.8.2-31 ----------------- * Fri May 13 2005 Mike A. Harris 6.8.2-31 - Added xorg-x11-6.8.2-ati-ragexl-ia64-avoidcpiofix.patch to workaround issue on ia64 with CPIO disabled in ati Mach64 driver (#155609,155610). For future reference, this is also included in RHEL4_U1 build 6.8.2-1.EL.13.5. From jspaleta at gmail.com Sat May 14 13:40:35 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Sat, 14 May 2005 09:40:35 -0400 Subject: yum upgrade errors : %postun In-Reply-To: <5cf776b8050512072565bfa536@mail.gmail.com> References: <5cf776b8050512072565bfa536@mail.gmail.com> Message-ID: <604aa79105051406403c5dd0ad@mail.gmail.com> On 5/12/05, mbneto wrote: > Hi, > > it has been a while that i can't yum upgrade because of errors. > > All rpms installed complain about %postrun errors. Have you filled up a system partition perhaps? well.. first thing you can do is attempt to recreate the problem using rpm command directly instead of yum. You'll find the rawhide updates that you downloaded, cached in /var/cache/yum/development/packages If the errors still occure when using rpm directly... then you can look at the scripts in the package for example: rpm -q --scripts -p /pathto/rpm-4.4.1-18.1.i386.rpm Since your error is a "pre" error and not a "post" error for the rpm package you should concentrate on the preinstall scriptlet for rpm-4.4.1-18.1.i386.rpm, and attempt to find which command in the preinstall scriptlet is failing. The scriptlet is just a series of shell commands using /bin/sh as the shell. You can go through the scriptlet one shell command at a time and find which command is failing. That will most likely point you to a problem with your system. For completeness here is the part of the rm -q --scripts -p /pathto/rpm-4.4.1-18.1.i386.rpm that you are interested in: preinstall scriptlet (using /bin/sh): <-this is the section header, shell commands start below if [ -f /var/lib/rpm/packages.rpm ]; then echo " You have (unsupported) /var/lib/rpm/packages.rpm db1 format installed package headers Please install rpm-4.0.4 first, and do rpm --rebuilddb to convert your database from db1 to db3 format. " exit 1 fi /usr/sbin/groupadd -g 37 rpm > /dev/null 2>&1 /usr/sbin/useradd -r -d /var/lib/rpm -u 37 -g 37 rpm -s /sbin/nologin > /dev/null 2>&1 exit 0 -jef From denis at poolshark.org Sat May 14 16:48:41 2005 From: denis at poolshark.org (Denis Leroy) Date: Sat, 14 May 2005 09:48:41 -0700 Subject: FC4t3 not booting in VMWare Message-ID: <42862BE9.50805@poolshark.org> Hi, I'm running FC4t3 as guest OS in a VMWare virtual machine. Kernel 2.6.11-1.1287_FC4 is the last kernel that was booting correctly for me, the subsequent kernel updates (290 and 303) fail right away with this error : PCI: Cannot allocate resource region 4 of device 0000:00:07.1 audit(1116064003.996:0): initialized Red Hat nash version 4.2.12 starting mknod: failed to create /dev/console: 17 mknod: failed to create /dev/null: 17 mknod: failed to create /dev/zero: 17 mkrootdev: label / not found mount: error 2 mounting ext3 ERROR opening /dev/console!!!!: 2 error dup2'ing fd of 0 to 0 error dup2'ing fd of 0 to 1 error dup2'ing fd of 0 to 2 switchroot: mount failed: 22 Kernel panic - not syncing: Attempted to kill init! Note the PCI warning occurs even when the kernel boots correctly. Any ideas ? I'll try to poke one of my contacts at VMWare... -denis From pmatilai at welho.com Sun May 15 07:18:16 2005 From: pmatilai at welho.com (Panu Matilainen) Date: Sun, 15 May 2005 10:18:16 +0300 (EEST) Subject: yum upgrade errors : %postun In-Reply-To: <5cf776b805051404484608c8a2@mail.gmail.com> References: <5cf776b8050512072565bfa536@mail.gmail.com> <1115908344.4604.42.camel@cutter> <5cf776b805051404484608c8a2@mail.gmail.com> Message-ID: On Sat, 14 May 2005, mbneto wrote: > All of them... > > For instance. > > # yum -y upgrade rpm ... > Running Transaction Test > warning: rpm-4.4.1-18.1: Header V3 DSA signature: NOKEY, key ID 30c9ecf8 > Finished Transaction Test > Transaction Test Succeeded > Running Transaction > error: %pre(rpm-4.4.1-18.1.i386) scriptlet failed, exit status 255 > error: install: %pre scriptlet failed (2), skipping rpm-4.4.1-18.1 When *everything* fails like this it's quite possibly a SELinux problem, similar things happen if you try to use mach with selinux enabled. Try this: # touch /.autorelabel # reboot ..and see if things start working again. However the fact that all those scriptlets have failed the system might be in somewhat messed up state... - Panu - From buildsys at redhat.com Sun May 15 12:10:55 2005 From: buildsys at redhat.com (Build System) Date: Sun, 15 May 2005 08:10:55 -0400 Subject: rawhide report: 20050515 changes Message-ID: <200505151210.j4FCAtbc015389@porkchop.devel.redhat.com> Updated Packages: From russell at coker.com.au Sun May 15 15:06:27 2005 From: russell at coker.com.au (Russell Coker) Date: Mon, 16 May 2005 01:06:27 +1000 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505140246.36294.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> Message-ID: <200505160106.32177.russell@coker.com.au> On Saturday 14 May 2005 02:46, Russell Coker wrote: > I am seeing /etc/ld.so.cache getting type etc_t for an initial install of > FC4T3. Is anyone else seeing this? > > At this stage I'm not sure whether I messed up my install process or > whether it's a more general thing. I've found the problem. The domain anaconda_t seems to be unused (we should probably just delete anaconda.te). The installation process runs all initial programs from an initrd (gzip compressed cpio file). cpio has no support for SE Linux labels so no domain transitions occur and everything runs in kernel_t. Everything that's not in an initrd is in a cramfs file system (which also has no support for SE Linux labelling). This means that created files get the type of the directory - etc_t in the case of /etc/ld.so.cache. One possible method of dealing with this would be the following: domain_auto_trans(kernel_t, ldconfig_exec_t, ldconfig_t) The other option is to run restorecon at the end of the install. Both options are ugly hacks. Given that we aren't doing anything with SE Linux at the install the best option is probably to create a policy that defines all file types with a single domain that can have read/write access to them, this will save space in the stage2 files and also precious RAM (currently installing to a machine with 64M of RAM is almost impossible, and 4176K of that problem is the SE Linux policy). I've attached a little Perl script that will munge a targeted policy. It replaces most type and domain definitions with typealias rules and reduces the policy binary size from 4176K to 60K. That saves 4116K of kernel memory and almost 700K on the cramfs. The saving of 4M of kernel memory will make a huge difference to the install on small machines. Currently it's almost impossible to install a FC4 test version on a machine with 64M of RAM, this change will give the same result as adding another 4M of RAM to machines for the installer (particularly important for machines that run out of RAM before completing the partitioning process). -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page -------------- next part -------------- A non-text attachment was scrubbed... Name: tiny.pl Type: application/x-perl Size: 1170 bytes Desc: not available URL: From nphilipp at redhat.com Sun May 15 15:47:31 2005 From: nphilipp at redhat.com (Nils Philippsen) Date: Sun, 15 May 2005 17:47:31 +0200 Subject: rawhide report: 20050510 changes In-Reply-To: <4284B68E.2070102@feuerpokemon.de> References: <200505101212.j4ACCZQL028902@porkchop.devel.redhat.com> <20050510121932.GA20078@ryoko.camperquake.de> <20050510122653.GC17420@devserv.devel.redhat.com> <1115739006.30474.94.camel@localhost.localdomain> <1115979622.3448.10.camel@gibraltar.stuttgart.redhat.com> <4284B68E.2070102@feuerpokemon.de> Message-ID: <1116172051.3451.11.camel@gibraltar.stuttgart.redhat.com> On Fri, 2005-05-13 at 16:15 +0200, dragoran wrote: > Nils Philippsen wrote: > > >On Tue, 2005-05-10 at 11:30 -0400, Owen Taylor wrote: > > > > > >>On Tue, 2005-05-10 at 08:26 -0400, Jakub Jelinek wrote: > >> > >> > >> > >>>a) on x86-64 this is unnecessary, all x86-64's have -msse2 by default > >>>b) -msse2 implies -msse > >>>c) if you use -msse2 in CFLAGS for all files, you can't run the latest > >>> GIMP on e.g. Pentium2, or pre-x86_64 AMD chips. > >>> -msse2 should be ONLY used on sources that have SSE/SSE2 stuff in it, > >>> and GIMP should make sure that no routine from those sources will be > >>> ever called on pre-SSE2 chips > >>> > >>> > >>This sounds like a big pain ... shouldn't there be a way to say > >>"use sse2/sse only for builtins" ? > >> > >>Without that, there is no way to use builtins in single functions; you > >>need to have one file for sse2, one for sse, one for mmx, and do the > >>detection of the current processor somewhere else entirely. > >> > >> > > > >I concur, with gcc <= 4.0.0-2 it was possible to use MMX/SSE calls in > >inline assembly without having to use -mmmx/-msse/-msse2. In order to > >build the gimp so that it uses MMX/SSE where available, I had to resort > >to serious autofoo munging because it doesn't allow for setting > >per-object compiler flags. > > > >Nils > > > > > why aren't sse/mmx/sse2 enabled by default on x86_64 ? all those chips > supports it. I got some things mixed up in my original assessment of the problem: - I thought that gcc only allowed MMX/SSE/SSE2-specific inline assembler when using -mmmx/-msse/-msse2. Actually the real problem was that gcc only allows MMX/SSE registers in clobber lists when using -mmmx/-msse/-msse2 and complains about them being present if these options aren't used. Masking them out with #ifdef __MMX__/__SSE__ ... #endif is the real solution. - I thought the problem would show on x86_64 as well while it doesn't due to the aforementioned reasons. gimp-2.2.7-4 contains the real fix. Nils -- Nils Philippsen / Red Hat / nphilipp at redhat.com "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- B. Franklin, 1759 PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011 From russell at coker.com.au Sun May 15 17:09:06 2005 From: russell at coker.com.au (Russell Coker) Date: Mon, 16 May 2005 03:09:06 +1000 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505160106.32177.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> Message-ID: <200505160309.09114.russell@coker.com.au> On Monday 16 May 2005 01:06, Russell Coker wrote: > I've attached a little Perl script that will munge a targeted policy. It > replaces most type and domain definitions with typealias rules and reduces > the policy binary size from 4176K to 60K. That saves 4116K of kernel > memory and almost 700K on the cramfs. The saving of 4M of kernel memory > will make a huge difference to the install on small machines. Currently > it's almost impossible to install a FC4 test version on a machine with 64M > of RAM, this change will give the same result as adding another 4M of RAM > to machines for the installer (particularly important for machines that run > out of RAM before completing the partitioning process). I've attached a new version, my first version had a bug that caused files created in the post install scripts of packages and the post install for kickstart get the wrong type. For reference, if the type on a directory is an alias it seems that new objects created under the directory get the base type in the security.selinux xattr not the alias name. Anyway with this change the result is correct (verified by running setfiles -v on a fresh install - I found evidence of other bugs but no bugs caused by my code). The policy.19 file will now be 444K in size, this saves 3732K of kernel memory which is still worth doing. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page -------------- next part -------------- A non-text attachment was scrubbed... Name: tiny.pl Type: application/x-perl Size: 1177 bytes Desc: not available URL: From buildsys at redhat.com Mon May 16 12:09:23 2005 From: buildsys at redhat.com (Build System) Date: Mon, 16 May 2005 08:09:23 -0400 Subject: rawhide report: 20050516 changes Message-ID: <200505161209.j4GC9NMH019930@porkchop.devel.redhat.com> Updated Packages: From russell at coker.com.au Mon May 16 12:16:14 2005 From: russell at coker.com.au (Russell Coker) Date: Mon, 16 May 2005 22:16:14 +1000 Subject: .pyo files and cups Message-ID: <200505162216.17416.russell@coker.com.au> After doing a default install of FC4T3 I relabeled the file system. Ideally such an operation should be a no-op as the install should have used the same labels that the relabel operation does. Below is part of the output of setfiles when performing the relabel. It seems that the pyo files were created by a post-install script in the rpm package or by a system boot script. Could someone please advise on what programs are likely to need the access to create such files? setfiles: relabeling /usr/share/printconf/util/cups_import.pyo from system_u:object_r:usr_t to system_u:object_r:printconf_t setfiles: relabeling /usr/share/printconf/util/backend.pyo from system_u:object_r:usr_t to system_u:object_r:printconf_t setfiles: relabeling /usr/share/printconf/util/scan_usb_devices.pyo from system_u:object_r:usr_t to system_u:object_r:printconf_t setfiles: relabeling /usr/share/printconf/util/printconf_conf.pyo from system_u:object_r:usr_t to system_u:object_r:printconf_t -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From sds at tycho.nsa.gov Mon May 16 12:11:27 2005 From: sds at tycho.nsa.gov (Stephen Smalley) Date: Mon, 16 May 2005 08:11:27 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505160106.32177.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> Message-ID: <1116245487.28782.42.camel@moss-spartans.epoch.ncsc.mil> On Mon, 2005-05-16 at 01:06 +1000, Russell Coker wrote: > I've found the problem. > > The domain anaconda_t seems to be unused (we should probably just delete > anaconda.te). The installation process runs all initial programs from an > initrd (gzip compressed cpio file). cpio has no support for SE Linux labels > so no domain transitions occur and everything runs in kernel_t. Everything > that's not in an initrd is in a cramfs file system (which also has no support > for SE Linux labelling). This means that created files get the type of the > directory - etc_t in the case of /etc/ld.so.cache. initrd or initramfs? Sounds like the latter from your description. An initrd should be able to support a labeled filesystem like ext2, unlike initramfs. By created files, you just mean runtime-generated files, right? Any files set down by rpm should be labeled explicitly and correctly by it. > One possible method of dealing with this would be the following: > domain_auto_trans(kernel_t, ldconfig_exec_t, ldconfig_t) Is ldconfig labeled correctly at this point? Is this an ldconfig that is installed by rpm on the disk (vs. one from the initramfs)? > The other option is to run restorecon at the end of the install. Both options > are ugly hacks. Applying restorecon selectively to runtime-generated files wouldn't be too bad. > Given that we aren't doing anything with SE Linux at the install the best > option is probably to create a policy that defines all file types with a > single domain that can have read/write access to them, this will save space > in the stage2 files and also precious RAM (currently installing to a machine > with 64M of RAM is almost impossible, and 4176K of that problem is the SE > Linux policy). In that case, do you need to enable SELinux in the kernel at all for installation? As long as the kernel provides the xattr API and the filesystem supports the security xattrs, rpm should be able to label all files it installs even with a SELinux-disabled kernel (although you might have to remove an is_selinux_enabled() test from rpm to make it do that). With SELinux disabled, you don't need to load a policy at all to set the file contexts. But how does simplifying the install policy (or disabling SELinux altogether during install) address the problem of runtime-created files like /etc/ld.so.cache? > I've attached a little Perl script that will munge a targeted policy. It > replaces most type and domain definitions with typealias rules and reduces > the policy binary size from 4176K to 60K. That saves 4116K of kernel memory > and almost 700K on the cramfs. The saving of 4M of kernel memory will make a > huge difference to the install on small machines. Currently it's almost > impossible to install a FC4 test version on a machine with 64M of RAM, this > change will give the same result as adding another 4M of RAM to machines for > the installer (particularly important for machines that run out of RAM before > completing the partitioning process). As you discovered, SELinux only uses type aliases to allow applications and users to specify alternative names; internally, it always canonicalizes to the primary type name. It is true that you can presently set the context to a type alias on disk (but only because the xattr API goes straight to the filesystem code and doesn't give SELinux an opportunity to canonicalize the attribute value, unlike the old SELinux API), but newly created files will always get the primary name. Aside from that, I'm not adverse to compressing the policy (although I would think that collapsing all domains and reducing to a minimal set of allow rules would achieve that) for installation, although it would be even simpler if you could just install with SELinux disabled in the kernel and thus avoid requiring loading a policy at all. -- Stephen Smalley National Security Agency From russell at coker.com.au Mon May 16 12:45:08 2005 From: russell at coker.com.au (Russell Coker) Date: Mon, 16 May 2005 22:45:08 +1000 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116245487.28782.42.camel@moss-spartans.epoch.ncsc.mil> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> <1116245487.28782.42.camel@moss-spartans.epoch.ncsc.mil> Message-ID: <200505162245.13035.russell@coker.com.au> On Monday 16 May 2005 22:11, Stephen Smalley wrote: > On Mon, 2005-05-16 at 01:06 +1000, Russell Coker wrote: > > I've found the problem. > > > > The domain anaconda_t seems to be unused (we should probably just delete > > anaconda.te). The installation process runs all initial programs from an > > initrd (gzip compressed cpio file). cpio has no support for SE Linux > > labels so no domain transitions occur and everything runs in kernel_t. > > Everything that's not in an initrd is in a cramfs file system (which also > > has no support for SE Linux labelling). This means that created files > > get the type of the directory - etc_t in the case of /etc/ld.so.cache. > > initrd or initramfs? Sounds like the latter from your description. An > initrd should be able to support a labeled filesystem like ext2, unlike > initramfs. initrd. Sure an initrd can support ext2 with labels, but that's not being done at the moment and such a significant change is unlikely to be made to the installer in a hurry. > By created files, you just mean runtime-generated files, right? Any > files set down by rpm should be labeled explicitly and correctly by it. Yes. > > One possible method of dealing with this would be the following: > > domain_auto_trans(kernel_t, ldconfig_exec_t, ldconfig_t) > > Is ldconfig labeled correctly at this point? Is this an ldconfig that > is installed by rpm on the disk (vs. one from the initramfs)? Yes, it SHOULD be labeled correctly. I haven't verified it (verifying exactly what happens at each stage of the install is painful), but it's supposed to be labeled and I have no reason to expect it not to be. > > The other option is to run restorecon at the end of the install. Both > > options are ugly hacks. > > Applying restorecon selectively to runtime-generated files wouldn't be > too bad. True. It seems likely that we will have to do that regardless (see my next message). > > Given that we aren't doing anything with SE Linux at the install the best > > option is probably to create a policy that defines all file types with a > > single domain that can have read/write access to them, this will save > > space in the stage2 files and also precious RAM (currently installing to > > a machine with 64M of RAM is almost impossible, and 4176K of that problem > > is the SE Linux policy). > > In that case, do you need to enable SELinux in the kernel at all for > installation? As long as the kernel provides the xattr API and the > filesystem supports the security xattrs, rpm should be able to label all > files it installs even with a SELinux-disabled kernel (although you > might have to remove an is_selinux_enabled() test from rpm to make it do > that). With SELinux disabled, you don't need to load a policy at all to > set the file contexts. True. But with SE Linux disabled I believe that files which are created by programs that are not SE Linux aware (IE the %post scripts run from RPM packages) will not have any labels. Thus we would create another problem of how to get those files labeled. Currently we have about a dozen files that have issues and an "everything" install might turn up another few dozen. If we have SE Linux disabled then we can expect hundreds or maybe thousands. It will be significantly more painful. We can manage a list of a dozen problem files that occasionally changes. We can't manage a list of a few thousand files that change more frequently in proportion to their number. > But how does simplifying the install policy (or disabling SELinux > altogether during install) address the problem of runtime-created files > like /etc/ld.so.cache? It doesn't. I investigated the issue and went off on a tangent. But these things are linked and need to be solved at the same time. > Aside from that, I'm not adverse to compressing the policy (although I > would think that collapsing all domains and reducing to a minimal set of > allow rules would achieve that) for installation, although it would be > even simpler if you could just install with SELinux disabled in the > kernel and thus avoid requiring loading a policy at all. My script does collapse all domains. As for a minimal set of allow rules, I guess if we turn off auditing of AVC messages we could just remove all allow and dontaudit rules. It's a hack, but that's the way the install process goes. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From sds at tycho.nsa.gov Mon May 16 12:41:15 2005 From: sds at tycho.nsa.gov (Stephen Smalley) Date: Mon, 16 May 2005 08:41:15 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505162245.13035.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> <1116245487.28782.42.camel@moss-spartans.epoch.ncsc.mil> <200505162245.13035.russell@coker.com.au> Message-ID: <1116247275.28782.64.camel@moss-spartans.epoch.ncsc.mil> On Mon, 2005-05-16 at 22:45 +1000, Russell Coker wrote: > My script does collapse all domains. As for a minimal set of allow rules, I > guess if we turn off auditing of AVC messages we could just remove all allow > and dontaudit rules. It's a hack, but that's the way the install process > goes. I don't think that there is a way to turn off auditing of AVC messages globally (you need dontaudit rules, and they have to specify the relevant types, classes, and permissions). -- Stephen Smalley National Security Agency From twaugh at redhat.com Mon May 16 13:04:59 2005 From: twaugh at redhat.com (Tim Waugh) Date: Mon, 16 May 2005 14:04:59 +0100 Subject: .pyo files and cups In-Reply-To: <200505162216.17416.russell@coker.com.au> References: <200505162216.17416.russell@coker.com.au> Message-ID: <20050516130459.GF8706@redhat.com> On Mon, May 16, 2005 at 10:16:14PM +1000, Russell Coker wrote: > After doing a default install of FC4T3 I relabeled the file system. Ideally > such an operation should be a no-op as the install should have used the same > labels that the relabel operation does. Below is part of the output of > setfiles when performing the relabel. It seems that the pyo files were > created by a post-install script in the rpm package or by a system boot > script. No, they are created when Python executes the corresponding .py files. There is a bug in bugzilla about this. What *should* happen is that, at RPM build time, the pyo/pyc files get generated and packaged in the RPM. https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=129025 This is an FC4 target bug. It was filed back in August last year(!). Not only that, but apparently the fix has been sitting in CVS since last month. Tim. */ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From russell at coker.com.au Mon May 16 14:11:39 2005 From: russell at coker.com.au (Russell Coker) Date: Tue, 17 May 2005 00:11:39 +1000 Subject: files without SE Linux labels on a default install - no Anaconda labeling Message-ID: <200505170011.41762.russell@coker.com.au> On an SE Linux system barring file system corruption and quota issues every file on a regular file system (Ext3 etc) should have a SE Linux label. As a test I did a default install of FC4T3 in the "Personal Workstation" configuration and checked this. Below is the relevant output from setfiles -v when relabelling the root file system. It seems to me that /usr/share/apps/ksplash, /usr/share/apps/ksplash/Themes, /usr/share/anaconda, /usr/share/anaconda/pixmaps, /usr/lib/anaconda-runtime, /usr/lib/anaconda-runtime/boot, and the install logs are created by Anaconda which doesn't apply SE Linux labels. Would it be possible to get Anaconda changed to apply labels to files and directories that it creates? I have no idea why the Portuguese Brazilian language file didn't get a label when all the other language files did. I have attached a list of all the files which aren't correctly labeled after a default targeted install which I haven't dealt with in other messages. NB this includes /etc/shadow... setfiles: relabeling /usr/share/apps/ksplash from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/share/apps/ksplash/Themes from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/share/anaconda from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/share/anaconda/pixmaps from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/lib/anaconda-runtime from system_u:object_r:file_t to system_u:object_r:lib_t setfiles: relabeling /usr/lib/anaconda-runtime/boot from system_u:object_r:file_t to system_u:object_r:lib_t setfiles: relabeling /usr/X11R6/lib/X11/locale/pt_BR.UTF-8 from system_u:object_r:file_t to system_u:object_r:lib_t setfiles: relabeling /root/install.log from system_u:object_r:file_t to root:object_r:user_home_t setfiles: relabeling /root/install.log.syslog from system_u:object_r:file_t to root:object_r:user_home_t -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page -------------- next part -------------- setfiles: relabeling /usr/share/apps/ksplash from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/share/apps/ksplash/Themes from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/share/anaconda from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/share/anaconda/pixmaps from system_u:object_r:file_t to system_u:object_r:usr_t setfiles: relabeling /usr/lib/anaconda-runtime from system_u:object_r:file_t to system_u:object_r:lib_t setfiles: relabeling /usr/lib/anaconda-runtime/boot from system_u:object_r:file_t to system_u:object_r:lib_t setfiles: relabeling /usr/X11R6/lib/X11/locale/pt_BR.UTF-8 from system_u:object_r:file_t to system_u:object_r:lib_t setfiles: relabeling /root/install.log from system_u:object_r:file_t to root:object_r:user_home_t setfiles: relabeling /root/install.log.syslog from system_u:object_r:file_t to root:object_r:user_home_t setfiles: relabeling /etc/ssh/ssh_host_key from system_u:object_r:etc_runtime_t to system_u:object_r:sshd_key_t setfiles: relabeling /etc/ssh/ssh_host_rsa_key from system_u:object_r:etc_runtime_t to system_u:object_r:sshd_key_t setfiles: relabeling /etc/ssh/ssh_host_dsa_key from system_u:object_r:etc_runtime_t to system_u:object_r:sshd_key_t setfiles: relabeling /etc/asound.conf from system_u:object_r:etc_runtime_t to system_u:object_r:etc_t setfiles: relabeling /etc/shadow from system_u:object_r:etc_t to system_u:object_r:shadow_t setfiles: relabeling /etc/gshadow- from system_u:object_r:etc_t to system_u:object_r:shadow_t setfiles: relabeling /etc/cups/cupsd.conf from system_u:object_r:cupsd_etc_t to system_u:object_r:cupsd_rw_etc_t setfiles: relabeling /etc/cups/printers.conf from system_u:object_r:cupsd_etc_t to system_u:object_r:cupsd_rw_etc_t setfiles: relabeling /etc/cups/cupsd.conf.save from system_u:object_r:cupsd_etc_t to system_u:object_r:cupsd_rw_etc_t setfiles: relabeling /etc/aliases.db from system_u:object_r:etc_t to system_u:object_r:etc_aliases_t setfiles: relabeling /etc/shadow- from system_u:object_r:etc_t to system_u:object_r:shadow_t setfiles: relabeling /etc/gshadow from system_u:object_r:etc_t to system_u:object_r:shadow_t setfiles: relabeling /etc/.pwd.lock from system_u:object_r:etc_t to system_u:object_r:shadow_t setfiles: relabeling /etc/dhclient-eth0.conf from system_u:object_r:etc_runtime_t to system_u:object_r:dhcp_etc_t setfiles: relabeling /etc/sysconfig/mouse from system_u:object_r:etc_runtime_t to system_u:object_r:etc_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.dep from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.ieee1394map from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.usbmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.inputmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.isapnpmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.symbols from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.ccwmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.alias from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.pcimap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t setfiles: relabeling /home/rjc from system_u:object_r:home_root_t to user_u:object_r:user_home_dir_t setfiles: relabeling /var/run/sm-client.pid from system_u:object_r:initrc_var_run_t to system_u:object_r:sendmail_var_run_t setfiles: relabeling /var/log/lastlog from system_u:object_r:var_log_t to system_u:object_r:lastlog_t setfiles: relabeling /var/log/btmp from system_u:object_r:var_log_t to system_u:object_r:faillog_t setfiles: relabeling /var/log/mail from system_u:object_r:var_log_t to system_u:object_r:sendmail_log_t From toshio at tiki-lounge.com Mon May 16 14:30:46 2005 From: toshio at tiki-lounge.com (Toshio) Date: Mon, 16 May 2005 10:30:46 -0400 Subject: workaround gcc-4 assembly constraint bug Message-ID: <1116253847.23731.19.camel@Madison.badger.com> I've been working on getting the ocaml extras package compiled on devel: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=156238 There appears to be a bug in gcc4's inline assembly constraints: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21291 that's keeping the compiler from selecting memory when we run out of registers. Here's the affected ocaml code from: ocaml-3.08.3/otherlibs/num/bng_ia32.c:bngdigit bng_ia32_mult_sub_digit: asm("1: \n\t" "movl (%1), %%eax \n\t" "movl (%0), %4 \n\t" "mull %5\n\t" /* edx:eax = d * next digit of b */ "subl %%eax, %4 \n\t" /* subtract eax from next digit of a */ "adcl $0, %%edx \n\t" /* accumulate carry in edx */ "subl %3, %4 \n\t" /* subtract out */ "adcl $0, %%edx \n\t" /* accumulate carry in edx */ "movl %4, (%0) \n\t" /* store next digit of result */ "movl %%edx, %3 \n\t" /* edx is next out */ "leal 4(%0), %0 \n\t" "leal 4(%1), %1 \n\t" "decl %2 \n\t" "jnz 1b" : "+&r" (a), "+&r" (b), "+&rm" (blen), "+&rm" (out), "=&r" (tmp) : "rm" (d) : "eax", "edx"); I think that the solution should be to change the constraints to specify memory only rather than a register with memory fallback: : "+&r" (a), "+&r" (b), "+m" (blen), "+m" (out), "=&r" (tmp) : "rm" (d) : "eax", "edx"); I'm not too familiar with gcc's constraints, though. Would anyone with a bit more experience be willing to tell me if that looks right or not? Thanks, Toshio -- toshio \ 25th March, 1999: Cold rain in Georgia. Blustery blowing wind. @tiki \ Freezing fingers and tired body. Hiking on because the shelters -lounge \ are filled with other hikers. .com \__________Life is miserable -- Life is grand!_______ GA->ME 1999 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From pjones at redhat.com Mon May 16 15:13:04 2005 From: pjones at redhat.com (Peter Jones) Date: Mon, 16 May 2005 11:13:04 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505162245.13035.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> <1116245487.28782.42.camel@moss-spartans.epoch.ncsc.mil> <200505162245.13035.russell@coker.com.au> Message-ID: <1116256384.9888.12.camel@localhost.localdomain> On Mon, 2005-05-16 at 22:45 +1000, Russell Coker wrote: > On Monday 16 May 2005 22:11, Stephen Smalley wrote: > > On Mon, 2005-05-16 at 01:06 +1000, Russell Coker wrote: > > > I've found the problem. > > > > > > The domain anaconda_t seems to be unused (we should probably just delete > > > anaconda.te). The installation process runs all initial programs from an > > > initrd (gzip compressed cpio file). cpio has no support for SE Linux > > > labels so no domain transitions occur and everything runs in kernel_t. > > > Everything that's not in an initrd is in a cramfs file system (which also > > > has no support for SE Linux labelling). This means that created files > > > get the type of the directory - etc_t in the case of /etc/ld.so.cache. > > > > initrd or initramfs? Sounds like the latter from your description. An > > initrd should be able to support a labeled filesystem like ext2, unlike > > initramfs. > > initrd. Sure an initrd can support ext2 with labels, but that's not being > done at the moment and such a significant change is unlikely to be made to > the installer in a hurry. Anaconda has been using initramfs for boot media since November. Are you sure you mean initrd? Regardless of that, why isn't ld.so.cache's context getting set correctly from the data in the glibc package? -- Peter From sds at tycho.nsa.gov Mon May 16 15:27:49 2005 From: sds at tycho.nsa.gov (Stephen Smalley) Date: Mon, 16 May 2005 11:27:49 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116256384.9888.12.camel@localhost.localdomain> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> <1116245487.28782.42.camel@moss-spartans.epoch.ncsc.mil> <200505162245.13035.russell@coker.com.au> <1116256384.9888.12.camel@localhost.localdomain> Message-ID: <1116257269.28782.79.camel@moss-spartans.epoch.ncsc.mil> On Mon, 2005-05-16 at 11:13 -0400, Peter Jones wrote: > Anaconda has been using initramfs for boot media since November. Are > you sure you mean initrd? > > Regardless of that, why isn't ld.so.cache's context getting set > correctly from the data in the glibc package? It is a runtime-created file, and ldconfig is not specifically modified to set the security context on it, so it just follows the default behavior, i.e. if there is a file type transition rule for the creating domain and the parent directory type, then apply the resulting type (which is what normally happens when ldconfig is run in the ldconfig_t domain); otherwise, inherit the type from the parent directory. In this case, it seems that ldconfig is not running in its domain because the caller isn't in the expected domain because the calling sequence never transitioned out of kernel_t due to the lack of labeling on the initramfs. At least that is what I gleaned from Russell's posting. -- Stephen Smalley National Security Agency From russell at coker.com.au Mon May 16 15:44:41 2005 From: russell at coker.com.au (Russell Coker) Date: Tue, 17 May 2005 01:44:41 +1000 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116256384.9888.12.camel@localhost.localdomain> References: <200505140246.36294.russell@coker.com.au> <200505162245.13035.russell@coker.com.au> <1116256384.9888.12.camel@localhost.localdomain> Message-ID: <200505170144.46594.russell@coker.com.au> On Tuesday 17 May 2005 01:13, Peter Jones wrote: > > initrd. Sure an initrd can support ext2 with labels, but that's not > > being done at the moment and such a significant change is unlikely to be > > made to the installer in a hurry. > > Anaconda has been using initramfs for boot media since November. Are > you sure you mean initrd? That was my understanding of it, I thought that initrd=whatever for the boot loaded made it use initrd. Could you please give me a URL for the correct information. > Regardless of that, why isn't ld.so.cache's context getting set > correctly from the data in the glibc package? The cache file is created by ldconfig. So it's not an issue of the glibc package or RPM. We could patch ldconfig to specifically request the context we desire (using the same mechanism that rpm uses to determine the correct file type), but that seems like a waste as such code would only be needed for the install. file_type_auto_trans(ldconfig_t, etc_t, ld_so_cache_t, file) In normal operation the ldconfig program runs in domain ldconfig_t. The above SE Linux policy specifies that when domain ldconfig_t creates a file in a directory of type etc_t the file type should be ld_so_cache_t. Currently during the install everything runs in kernel_t (including ldconfig) so the policy in question does not apply. The options to solve this are to hack the policy or to run restorecon at the end of the install. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From russell at coker.com.au Mon May 16 15:46:37 2005 From: russell at coker.com.au (Russell Coker) Date: Tue, 17 May 2005 01:46:37 +1000 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116257269.28782.79.camel@moss-spartans.epoch.ncsc.mil> References: <200505140246.36294.russell@coker.com.au> <1116256384.9888.12.camel@localhost.localdomain> <1116257269.28782.79.camel@moss-spartans.epoch.ncsc.mil> Message-ID: <200505170146.42815.russell@coker.com.au> On Tuesday 17 May 2005 01:27, Stephen Smalley wrote: > It is a runtime-created file, and ldconfig is not specifically modified > to set the security context on it, so it just follows the default > behavior, i.e. if there is a file type transition rule for the creating > domain and the parent directory type, then apply the resulting type > (which is what normally happens when ldconfig is run in the ldconfig_t > domain); otherwise, inherit the type from the parent directory. In this > case, it seems that ldconfig is not running in its domain because the > caller isn't in the expected domain because the calling sequence never > transitioned out of kernel_t due to the lack of labeling on the > initramfs. At least that is what I gleaned from Russell's posting. Yes. However although the kernel_t domain is used for everything the programs being run will all be from the chroot environment and thus have the correct types. Therefore ldconfig_exec_t will be used for the ldconfig program and we can do a domain transition on it. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From ronny-vlug at vlugnet.org Mon May 16 16:49:49 2005 From: ronny-vlug at vlugnet.org (Ronny Buchmann) Date: Mon, 16 May 2005 18:49:49 +0200 Subject: files without SE Linux labels on a default install - no Anaconda labeling In-Reply-To: <200505170011.41762.russell@coker.com.au> References: <200505170011.41762.russell@coker.com.au> Message-ID: <200505161849.49334.ronny-vlug@vlugnet.org> On Monday 16 May 2005 16:11, Russell Coker wrote: > On an SE Linux system barring file system corruption and quota issues every > file on a regular file system (Ext3 etc) should have a SE Linux label. > > As a test I did a default install of FC4T3 in the "Personal Workstation" > configuration and checked this. Below is the relevant output from setfiles > -v when relabelling the root file system. This was after the first start I assume (= at the second reboot)? > setfiles: relabeling /usr/share/apps/ksplash from system_u:object_r:file_t > to system_u:object_r:usr_t > setfiles: relabeling /usr/share/apps/ksplash/Themes from > system_u:object_r:file_t to system_u:object_r:usr_t owned by kdebase > setfiles: relabeling /usr/share/anaconda from system_u:object_r:file_t to > system_u:object_r:usr_t > setfiles: relabeling /usr/share/anaconda/pixmaps from > system_u:object_r:file_t to system_u:object_r:usr_t > setfiles: relabeling /usr/lib/anaconda-runtime from > system_u:object_r:file_t to system_u:object_r:lib_t > setfiles: relabeling /usr/lib/anaconda-runtime/boot from > system_u:object_r:file_t to system_u:object_r:lib_t owned by anaconda, anaconda-runtime rpm issue? > setfiles: relabeling /usr/X11R6/lib/X11/locale/pt_BR.UTF-8 from > system_u:object_r:file_t to system_u:object_r:lib_t should be owned by xorg-x11-libs > setfiles: relabeling /root/install.log from system_u:object_r:file_t to > root:object_r:user_home_t > setfiles: relabeling /root/install.log.syslog from > system_u:object_r:file_t to root:object_r:user_home_t this is from anaconda > setfiles: relabeling /etc/ssh/ssh_host_key from > system_u:object_r:etc_runtime_t to system_u:object_r:sshd_key_t > setfiles: relabeling /etc/ssh/ssh_host_rsa_key from > system_u:object_r:etc_runtime_t to system_u:object_r:sshd_key_t > setfiles: relabeling /etc/ssh/ssh_host_dsa_key from > system_u:object_r:etc_runtime_t to system_u:object_r:sshd_key_t these are generated on first start of sshd > setfiles: relabeling /etc/asound.conf from system_u:object_r:etc_runtime_t to system_u:object_r:etc_t kudzu or firstboot? > setfiles: relabeling /etc/shadow from system_u:object_r:etc_t to system_u:object_r:shadow_t > setfiles: relabeling /etc/gshadow- from system_u:object_r:etc_t to system_u:object_r:shadow_t anaconda > setfiles: relabeling /etc/cups/cupsd.conf from system_u:object_r:cupsd_etc_t to system_u:object_r:cupsd_rw_etc_t > setfiles: relabeling /etc/cups/printers.conf from system_u:object_r:cupsd_etc_t to system_u:object_r:cupsd_rw_etc_t > setfiles: relabeling /etc/cups/cupsd.conf.save from system_u:object_r:cupsd_etc_t to system_u:object_r:cupsd_rw_etc_t ? > setfiles: relabeling /etc/aliases.db from system_u:object_r:etc_t to system_u:object_r:etc_aliases_t sendmail init script? > setfiles: relabeling /etc/shadow- from system_u:object_r:etc_t to system_u:object_r:shadow_t > setfiles: relabeling /etc/gshadow from system_u:object_r:etc_t to system_u:object_r:shadow_t anaconda > setfiles: relabeling /etc/.pwd.lock from system_u:object_r:etc_t to system_u:object_r:shadow_t ? > setfiles: relabeling /etc/dhclient-eth0.conf from system_u:object_r:etc_runtime_t to system_u:object_r:dhcp_etc_t anaconda? > setfiles: relabeling /etc/sysconfig/mouse from system_u:object_r:etc_runtime_t to system_u:object_r:etc_t anaconda? > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.dep from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.ieee1394map from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.usbmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.inputmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.isapnpmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.symbols from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.ccwmap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.alias from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /lib/modules/2.6.11-1.1286_FC4/modules.pcimap from system_u:object_r:modules_object_t to system_u:object_r:modules_dep_t > setfiles: relabeling /home/rjc from system_u:object_r:home_root_t to user_u:object_r:user_home_dir_t firstboot? > setfiles: relabeling /var/run/sm-client.pid from system_u:object_r:initrc_var_run_t to system_u:object_r:sendmail_var_run_t sendmail init script > setfiles: relabeling /var/log/lastlog from system_u:object_r:var_log_t to system_u:object_r:lastlog_t > setfiles: relabeling /var/log/btmp from system_u:object_r:var_log_t to system_u:object_r:faillog_t initscripts? > setfiles: relabeling /var/log/mail from system_u:object_r:var_log_t to system_u:object_r:sendmail_log_t should be owned by some package (i.e. sendmail) -- http://LinuxWiki.org/RonnyBuchmann From mattdm at mattdm.org Mon May 16 17:43:59 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Mon, 16 May 2005 13:43:59 -0400 Subject: tmpwatch and removing directories.... Message-ID: <20050516174359.GA4622@jadzia.bu.edu> tmpwatch never seems to remove directories unless I give it the -c flag, which isn't always what I want to do. (In fact, usually it's not.) I think the problem is still the one described by Aleksey Nogin in this bug from July, 2000: This problem existed for as long as tmpwatch was in RedHat distribution. The problem is that by default tmpwatch uses atime for both files and directories. But because tmpwatch is ran daily, the atime for directories is always <= 1day, so it never removes them... I believe that by default, tmpwatch should use atime for files and mtime for empty directories... Yet, that's marked as fixed in rawhide in August, 2000. Am I missing something? Thanks.... -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 72 degrees Fahrenheit. From katzj at redhat.com Mon May 16 19:35:26 2005 From: katzj at redhat.com (Jeremy Katz) Date: Mon, 16 May 2005 15:35:26 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505160106.32177.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> Message-ID: <1116272126.3360.18.camel@bree.local.net> On Mon, 2005-05-16 at 01:06 +1000, Russell Coker wrote: > The domain anaconda_t seems to be unused (we should probably just delete > anaconda.te). The installation process runs all initial programs from an > initrd (gzip compressed cpio file). cpio has no support for SE Linux labels > so no domain transitions occur and everything runs in kernel_t. Everything > that's not in an initrd is in a cramfs file system (which also has no support > for SE Linux labelling). This means that created files get the type of the > directory - etc_t in the case of /etc/ld.so.cache. We never used label'ing of things in the initrd when it was an ext2 image. The loader explicitly sets the exec context before running anaconda to be system_u:object_r:anaconda_t if policy doesn't fail to load. Look in /tmp/anaconda.log (or tty3) for errors about loading the policy or setting the context. Jeremy From katzj at redhat.com Mon May 16 19:36:29 2005 From: katzj at redhat.com (Jeremy Katz) Date: Mon, 16 May 2005 15:36:29 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505170144.46594.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505162245.13035.russell@coker.com.au> <1116256384.9888.12.camel@localhost.localdomain> <200505170144.46594.russell@coker.com.au> Message-ID: <1116272189.3360.20.camel@bree.local.net> On Tue, 2005-05-17 at 01:44 +1000, Russell Coker wrote: > On Tuesday 17 May 2005 01:13, Peter Jones wrote: > > > initrd. Sure an initrd can support ext2 with labels, but that's not > > > being done at the moment and such a significant change is unlikely to be > > > made to the installer in a hurry. > > > > Anaconda has been using initramfs for boot media since November. Are > > you sure you mean initrd? > > That was my understanding of it, I thought that initrd=whatever for the boot > loaded made it use initrd. Could you please give me a URL for the correct > information. The initramfs is loaded using the protocol for loading initrds (ie, placing a blob at a certain location in memory by the bootloader instead of munging the kernel binary to have the initrd stuck at the end of it). Jeremy From sopwith at redhat.com Mon May 16 20:21:14 2005 From: sopwith at redhat.com (Elliot Lee) Date: Mon, 16 May 2005 16:21:14 -0400 (EDT) Subject: Raising the bar for FC4 Message-ID: Hi all, Fedora Core 4 public release is currently targeted for Monday, June 6. To hit that, we need some time in advance doing QA and final polishing. For test releases that is usually about a week, but this is the final FC4 release so we need a bit of extra time to do the final steps. This means that Monday, May 23 is the final final freeze for FC4. After that, the bar for accepting changes into FC4 is even higher. Fixes will need to be for showstopper bugs (data corruption, crashing programs, and other things that impact a large percentage of users in a major way). I've sent out reminder e-mails to people who currently own FC4Target and FC4Blocker bugs in bugzilla, so you should know if you have specific bugs to address. If you don't have any bugs to address, you can still help by doing test installs of rawhide. I'll also try to get a few intermediate test trees out for people to install. If the plan needs clarification, please let me know! -- Elliot From mitr at volny.cz Mon May 16 20:57:57 2005 From: mitr at volny.cz (Miloslav Trmac) Date: Mon, 16 May 2005 22:57:57 +0200 Subject: tmpwatch and removing directories.... In-Reply-To: <20050516174359.GA4622@jadzia.bu.edu> References: <20050516174359.GA4622@jadzia.bu.edu> Message-ID: <20050516205752.GB4281@chrys.ms.mff.cuni.cz> On Mon, May 16, 2005 at 01:43:59PM -0400, Matthew Miller wrote: > tmpwatch never seems to remove directories unless I give it the -c flag, > which isn't always what I want to do. (In fact, usually it's not.) I think > the problem is still the one described by Aleksey Nogin in this bug from > July, 2000: ("This" bug is #14930.) > This problem existed for as long as tmpwatch was in RedHat distribution. > The problem is that by default tmpwatch uses atime for both files and > directories. But because tmpwatch is ran daily, the atime for > directories is always <= 1day, so it never removes them... I believe > that by default, tmpwatch should use atime for files and mtime for empty > directories... > > Yet, that's marked as fixed in rawhide in August, 2000. Am I missing > something? Thanks.... #91096, I think. (Using mtime doesn't look safe enough; e.g. a daemon could create a "queue" directory and poll it for files added by other processes; that can lead to an often used directory with old mtime.) Mirek From mattdm at mattdm.org Mon May 16 21:12:28 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Mon, 16 May 2005 17:12:28 -0400 Subject: tmpwatch and removing directories.... In-Reply-To: <20050516205752.GB4281@chrys.ms.mff.cuni.cz> References: <20050516174359.GA4622@jadzia.bu.edu> <20050516205752.GB4281@chrys.ms.mff.cuni.cz> Message-ID: <20050516211228.GA14403@jadzia.bu.edu> On Mon, May 16, 2005 at 10:57:57PM +0200, Miloslav Trmac wrote: > > which isn't always what I want to do. (In fact, usually it's not.) I think > > the problem is still the one described by Aleksey Nogin in this bug from > > July, 2000: > ("This" bug is #14930.) Oh, sorry -- I meant to include that. > > Yet, that's marked as fixed in rawhide in August, 2000. Am I missing > > something? Thanks.... > #91096, I think. Yep, that's it. Thanks! > (Using mtime doesn't look safe enough; e.g. a daemon could create a > "queue" directory and poll it for files added by other processes; that > can lead to an often used directory with old mtime.) > Mirek Or the race condition within a program itself which you mention in the bug. Although an awfully slow race -- arguably, a daemon shouldn't *expect* that a directory it created in /tmp yesterday will still be there today. But yeah, I understand the need for safety here. Unfortunately, it means that tmpwatch never removes directories. What about adding an *option* to make tmpwatch do this? I run tmpwatch on my own ~/tmp, and I can be pretty sure any such race is not going to happen. Right now, there's no way to say "use atime for regular files, and mtime for directories", and I think that's a very common desirable case. It might also be worth mentioning what's going on in the man pages. -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 74 degrees Fahrenheit. From mitr at volny.cz Mon May 16 21:17:56 2005 From: mitr at volny.cz (Miloslav Trmac) Date: Mon, 16 May 2005 23:17:56 +0200 Subject: tmpwatch and removing directories.... In-Reply-To: <20050516211228.GA14403@jadzia.bu.edu> References: <20050516174359.GA4622@jadzia.bu.edu> <20050516205752.GB4281@chrys.ms.mff.cuni.cz> <20050516211228.GA14403@jadzia.bu.edu> Message-ID: <20050516211756.GC4281@chrys.ms.mff.cuni.cz> On Mon, May 16, 2005 at 05:12:28PM -0400, Matthew Miller wrote: > What about adding an *option* to make tmpwatch do this? Sounds reasonable, please file it in bugzilla. Mirek From mattdm at mattdm.org Mon May 16 21:26:10 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Mon, 16 May 2005 17:26:10 -0400 Subject: tmpwatch and removing directories.... In-Reply-To: <20050516211756.GC4281@chrys.ms.mff.cuni.cz> References: <20050516174359.GA4622@jadzia.bu.edu> <20050516205752.GB4281@chrys.ms.mff.cuni.cz> <20050516211228.GA14403@jadzia.bu.edu> <20050516211756.GC4281@chrys.ms.mff.cuni.cz> Message-ID: <20050516212610.GA15473@jadzia.bu.edu> On Mon, May 16, 2005 at 11:17:56PM +0200, Miloslav Trmac wrote: > On Mon, May 16, 2005 at 05:12:28PM -0400, Matthew Miller wrote: > > What about adding an *option* to make tmpwatch do this? > Sounds reasonable, please file it in bugzilla. Got a preference for new bug or reopenning #91096? Thanks! -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 74 degrees Fahrenheit. From mike.honeyfield at gmail.com Mon May 16 22:29:46 2005 From: mike.honeyfield at gmail.com (Mike Honeyfield) Date: Tue, 17 May 2005 10:29:46 +1200 Subject: "use register arguments" option enabled. Message-ID: I am just wondering the reasoning behind having the experimental option "Use register arguments" on in the kernel config. It has caused some issues for me, have since fixed it, but was wondering why an experimental feature that could/can break the ABI for 3rd party binary only modules would be enabled? I have noted this on FC2/3, not yet looked at FC4test and also RHEL4. Thanks in advance. Mike From ivazquez at ivazquez.net Mon May 16 22:42:53 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Mon, 16 May 2005 18:42:53 -0400 Subject: "use register arguments" option enabled. In-Reply-To: References: Message-ID: <1116283373.7161.3.camel@ignacio.ignacio.lan> On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: > I am just wondering the reasoning behind having the experimental > option "Use register arguments" on in the kernel config. > > It has caused some issues for me, have since fixed it, but was > wondering why an experimental feature that could/can break the ABI for > 3rd party binary only modules would be enabled? Speed. Pushing data onto the stack takes longer than just shoving it into a register. -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From mike.honeyfield at gmail.com Mon May 16 22:45:57 2005 From: mike.honeyfield at gmail.com (Mike Honeyfield) Date: Tue, 17 May 2005 10:45:57 +1200 Subject: "use register arguments" option enabled. In-Reply-To: <1116283373.7161.3.camel@ignacio.ignacio.lan> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> Message-ID: On 5/17/05, Ignacio Vazquez-Abrams wrote: > On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: > > I am just wondering the reasoning behind having the experimental > > option "Use register arguments" on in the kernel config. > > > > It has caused some issues for me, have since fixed it, but was > > wondering why an experimental feature that could/can break the ABI for > > 3rd party binary only modules would be enabled? > > Speed. Pushing data onto the stack takes longer than just shoving it > into a register. > So the speed improvement is enough to justify this? I could understand Fedora Core's kernels have this feature, but was a bit suprised to see the same issue in RHEL. Cheers Mike From ivazquez at ivazquez.net Mon May 16 23:01:59 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Mon, 16 May 2005 19:01:59 -0400 Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> Message-ID: <1116284519.7161.7.camel@ignacio.ignacio.lan> On Tue, 2005-05-17 at 10:45 +1200, Mike Honeyfield wrote: > On 5/17/05, Ignacio Vazquez-Abrams wrote: > > On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: > > > I am just wondering the reasoning behind having the experimental > > > option "Use register arguments" on in the kernel config. > > > > > > It has caused some issues for me, have since fixed it, but was > > > wondering why an experimental feature that could/can break the ABI for > > > 3rd party binary only modules would be enabled? > > > > Speed. Pushing data onto the stack takes longer than just shoving it > > into a register. > > > > So the speed improvement is enough to justify this? Even a 2 clock cycle delay happening hundreds of thousands or even millions of times a second builds up. > I could understand Fedora Core's kernels have this feature, but was a > bit suprised to see the same issue in RHEL. Agreed. But then again it really shouldn't affect a large number of modules. -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From wtogami at redhat.com Tue May 17 00:09:24 2005 From: wtogami at redhat.com (Warren Togami) Date: Mon, 16 May 2005 14:09:24 -1000 Subject: "use register arguments" option enabled. In-Reply-To: <1116284519.7161.7.camel@ignacio.ignacio.lan> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116284519.7161.7.camel@ignacio.ignacio.lan> Message-ID: <42893634.7040501@redhat.com> Ignacio Vazquez-Abrams wrote: > On Tue, 2005-05-17 at 10:45 +1200, Mike Honeyfield wrote: > >>On 5/17/05, Ignacio Vazquez-Abrams wrote: >> >>>On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: >>> >>>>I am just wondering the reasoning behind having the experimental >>>>option "Use register arguments" on in the kernel config. >>>> >>>>It has caused some issues for me, have since fixed it, but was >>>>wondering why an experimental feature that could/can break the ABI for >>>>3rd party binary only modules would be enabled? >>> >>>Speed. Pushing data onto the stack takes longer than just shoving it >>>into a register. >>> >> >>So the speed improvement is enough to justify this? > > > Even a 2 clock cycle delay happening hundreds of thousands or even > millions of times a second builds up. Isn't using system memory much longer than 2 clock cycles? Warren From ivazquez at ivazquez.net Tue May 17 00:14:58 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Mon, 16 May 2005 20:14:58 -0400 Subject: "use register arguments" option enabled. In-Reply-To: <42893634.7040501@redhat.com> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116284519.7161.7.camel@ignacio.ignacio.lan> <42893634.7040501@redhat.com> Message-ID: <1116288898.7161.10.camel@ignacio.ignacio.lan> On Mon, 2005-05-16 at 14:09 -1000, Warren Togami wrote: > Ignacio Vazquez-Abrams wrote: > > Even a 2 clock cycle delay happening hundreds of thousands or even > > millions of times a second builds up. > > Isn't using system memory much longer than 2 clock cycles? Absolutely. But I figure that if the concept can be grasped with as low a number as 2 then there's no need to worry about the actual number (which would in fact be somewhat higher) ;) -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From pjones at redhat.com Mon May 16 17:36:55 2005 From: pjones at redhat.com (Peter Jones) Date: Mon, 16 May 2005 13:36:55 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505170144.46594.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505162245.13035.russell@coker.com.au> <1116256384.9888.12.camel@localhost.localdomain> <200505170144.46594.russell@coker.com.au> Message-ID: <1116265016.15526.3.camel@localhost.localdomain> On Tue, 2005-05-17 at 01:44 +1000, Russell Coker wrote: > On Tuesday 17 May 2005 01:13, Peter Jones wrote: > > > initrd. Sure an initrd can support ext2 with labels, but that's not > > > being done at the moment and such a significant change is unlikely to be > > > made to the installer in a hurry. > > > > Anaconda has been using initramfs for boot media since November. Are > > you sure you mean initrd? > > That was my understanding of it, I thought that initrd=whatever for the boot > loaded made it use initrd. Could you please give me a URL for the correct > information. I don't have a url, but if you look at linux/init/initramfs.c, it appears to just check the magic bytes for the filesystem in whatever you pass it; if it's a cramfs image, it uses initramfs. > > Regardless of that, why isn't ld.so.cache's context getting set > > correctly from the data in the glibc package? > > The cache file is created by ldconfig. So it's not an issue of the glibc > package or RPM. We could patch ldconfig to specifically request the context > we desire (using the same mechanism that rpm uses to determine the correct > file type), but that seems like a waste as such code would only be needed for > the install. Why can't the glibc package provide a ld.so.cache which simply indicates no libraries? This seems more correct to me, especially as it claims ownership of the file. -- Peter From katzj at redhat.com Tue May 17 00:43:31 2005 From: katzj at redhat.com (Jeremy Katz) Date: Mon, 16 May 2005 20:43:31 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116265016.15526.3.camel@localhost.localdomain> References: <200505140246.36294.russell@coker.com.au> <200505162245.13035.russell@coker.com.au> <1116256384.9888.12.camel@localhost.localdomain> <200505170144.46594.russell@coker.com.au> <1116265016.15526.3.camel@localhost.localdomain> Message-ID: <1116290611.3360.33.camel@bree.local.net> On Mon, 2005-05-16 at 13:36 -0400, Peter Jones wrote: > On Tue, 2005-05-17 at 01:44 +1000, Russell Coker wrote: > > On Tuesday 17 May 2005 01:13, Peter Jones wrote: > > > > initrd. Sure an initrd can support ext2 with labels, but that's not > > > > being done at the moment and such a significant change is unlikely to be > > > > made to the installer in a hurry. > > > > > > Anaconda has been using initramfs for boot media since November. Are > > > you sure you mean initrd? > > > > That was my understanding of it, I thought that initrd=whatever for the boot > > loaded made it use initrd. Could you please give me a URL for the correct > > information. > > I don't have a url, but if you look at linux/init/initramfs.c, it > appears to just check the magic bytes for the filesystem in whatever you > pass it; if it's a cramfs image, it uses initramfs. Correction: cramfs has nothing to do with an initramfs. An initramfs is just a gzipped cpio archive that gets exploded out onto the default kernel rootfs. Jeremy From russell at coker.com.au Tue May 17 04:05:29 2005 From: russell at coker.com.au (Russell Coker) Date: Tue, 17 May 2005 14:05:29 +1000 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116272126.3360.18.camel@bree.local.net> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> <1116272126.3360.18.camel@bree.local.net> Message-ID: <200505171405.34672.russell@coker.com.au> On Tuesday 17 May 2005 05:35, Jeremy Katz wrote: > On Mon, 2005-05-16 at 01:06 +1000, Russell Coker wrote: > > The domain anaconda_t seems to be unused (we should probably just delete > > anaconda.te). The installation process runs all initial programs from an > > initrd (gzip compressed cpio file). cpio has no support for SE Linux > > labels so no domain transitions occur and everything runs in kernel_t. > > Everything that's not in an initrd is in a cramfs file system (which also > > has no support for SE Linux labelling). This means that created files > > get the type of the directory - etc_t in the case of /etc/ld.so.cache. > > We never used label'ing of things in the initrd when it was an ext2 > image. The loader explicitly sets the exec context before running > anaconda to be system_u:object_r:anaconda_t if policy doesn't fail to > load. Look in /tmp/anaconda.log (or tty3) for errors about loading the > policy or setting the context. That's an invalid context. The correct value should be system_u:system_r:anaconda_t. The role object_r is only suitable for files on disk. The kernel does allow setting it though. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From arjanv at redhat.com Tue May 17 06:59:34 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Tue, 17 May 2005 08:59:34 +0200 Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> Message-ID: <1116313174.6442.18.camel@laptopd505.fenrus.org> On Tue, 2005-05-17 at 10:45 +1200, Mike Honeyfield wrote: > On 5/17/05, Ignacio Vazquez-Abrams wrote: > > On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: > > > I am just wondering the reasoning behind having the experimental > > > option "Use register arguments" on in the kernel config. > > > > > > It has caused some issues for me, have since fixed it, but was > > > wondering why an experimental feature that could/can break the ABI for > > > 3rd party binary only modules would be enabled? > > > > Speed. Pushing data onto the stack takes longer than just shoving it > > into a register. > > > > So the speed improvement is enough to justify this? > > I could understand Fedora Core's kernels have this feature, but was a > bit suprised to see the same issue in RHEL. why? This only impacts binary modules, and in the process of porting those to 2.6 the vendor needs to "fix" this up just as well. That is the same for FC and RHEL. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From mike.honeyfield at gmail.com Tue May 17 07:29:55 2005 From: mike.honeyfield at gmail.com (Mike Honeyfield) Date: Tue, 17 May 2005 19:29:55 +1200 Subject: "use register arguments" option enabled. In-Reply-To: <1116313174.6442.18.camel@laptopd505.fenrus.org> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> Message-ID: On 5/17/05, Arjan van de Ven wrote: > On Tue, 2005-05-17 at 10:45 +1200, Mike Honeyfield wrote: > > On 5/17/05, Ignacio Vazquez-Abrams wrote: > > > On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: > > > > I am just wondering the reasoning behind having the experimental > > > > option "Use register arguments" on in the kernel config. > > > > > > > > It has caused some issues for me, have since fixed it, but was > > > > wondering why an experimental feature that could/can break the ABI for > > > > 3rd party binary only modules would be enabled? > > > > > > Speed. Pushing data onto the stack takes longer than just shoving it > > > into a register. > > > > > > > So the speed improvement is enough to justify this? > > > > I could understand Fedora Core's kernels have this feature, but was a > > bit suprised to see the same issue in RHEL. > > why? > > This only impacts binary modules, and in the process of porting those to > 2.6 the vendor needs to "fix" this up just as well. That is the same for > FC and RHEL. > So you disagree that this is actually "experimental" as per the kernel config? Doesn't the term experimental indirectly imply possible stability issues? Not reasuring IMHO. Yes, RH's choice in kernel configure option is theirs and yes, it is true, binary only vendor can "fix" (not sure waht the quotes imply), we have "fixed" our drivers. Worth noting, our driver was written for 2.6 and in RHEL it was broken, but not other platforms we support, like debian and suse. However, I was curious because it seemed ironic to possibly break ABI for binary only vendors in the enterprise product. This might be a bold assumption to make, but from the market I have seen where RHEL fits in, people are using closed sourced apps and drivers because that apps they need run on RHEL or are offically supported on RHEL. So it seems odd that one of the reasons that someone would use RHEL could also be the reason they cant run that app/driver. Hence my comment, I understood FC, but not RHEL. Cheers Mike From arjanv at redhat.com Tue May 17 08:07:40 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Tue, 17 May 2005 10:07:40 +0200 Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> Message-ID: <1116317261.6442.28.camel@laptopd505.fenrus.org> On Tue, 2005-05-17 at 19:29 +1200, Mike Honeyfield wrote: > On 5/17/05, Arjan van de Ven wrote: > > This only impacts binary modules, and in the process of porting those to > > 2.6 the vendor needs to "fix" this up just as well. That is the same for > > FC and RHEL. > > > > So you disagree that this is actually "experimental" as per the kernel config? yes I do. EXPERIMENTAL in the kernel config in practice is meaningless; it hardly ever gets removed even after years of high production use of stuff. I wonder if the EXPERIMENTAL got removed from EXT3 yet .. ;) > Yes, RH's choice in kernel configure option is theirs and yes, it is > true, binary only vendor can "fix" (not sure waht the quotes imply), > we have "fixed" our drivers. Worth noting, our driver was written for > 2.6 and in RHEL it was broken, but not other platforms we support, > like debian and suse. I assume that since your driver was written with linux in mind, that it is gpl, right? So then you wouldn't have this problem.... (if it's not, make sure your lawyers have read the COPYING.modules file from the kernel rpms) > However, I was curious because it seemed ironic to possibly break ABI > for binary only vendors in the enterprise product. The linux kernel has no ABI. So it's also not possible to break it. SLES uses this option too afaik, and it has been there for a long time in 2.6, any driver not capable of running with it is of course very buggy or at least not adjusted to 2.6 properly. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From mike.honeyfield at gmail.com Tue May 17 08:23:07 2005 From: mike.honeyfield at gmail.com (Mike Honeyfield) Date: Tue, 17 May 2005 20:23:07 +1200 Subject: "use register arguments" option enabled. In-Reply-To: <1116315398.6442.23.camel@laptopd505.fenrus.org> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> <1116315398.6442.23.camel@laptopd505.fenrus.org> Message-ID: On 5/17/05, Arjan van de Ven wrote: > > So you disagree that this is actually "experimental" as per the kernel config? > > yes I do. > EXPERIMENTAL in the kernel config in practice is meaningless; it hardly > ever gets removed even after years of high production use of stuff. I > wonder if the EXPERIMENTAL got removed from EXT3 yet .. ;) I see. > > Yes, RH's choice in kernel configure option is theirs and yes, it is > > true, binary only vendor can "fix" (not sure waht the quotes imply), > > we have "fixed" our drivers. Worth noting, our driver was written for > > 2.6 and in RHEL it was broken, but not other platforms we support, > > like debian and suse. > > I assume that since your driver was written with linux in mind, that it > is gpl, right? So then you wouldn't have this problem.... > (if it's not, make sure your lawyers have read the COPYING.modules file > from the kernel rpms) Yes, we are fine there. > > However, I was curious because it seemed ironic to possibly break ABI > > for binary only vendors in the enterprise product. > > The linux kernel has no ABI. So it's also not possible to break it. SLES > uses this option too afaik, and it has been there for a long time in > 2.6, any driver not capable of running with it is of course very buggy > or at least not adjusted to 2.6 properly. Sorry, I confused myself, re: ABI. The kernel config states its uses a different ABI when this option is set, not break. IIRC, SLES 9 ran fine. Debian has been the platform of choice by the developers, largely, but customers also use RHEL and SLES. As previously stated, the issue was corrected. The question was raised to find out why. As I understand it, it offers some performance improvements, not sure how much. Thanks for the clarification all. Mike From byte at aeon.com.my Tue May 17 08:50:57 2005 From: byte at aeon.com.my (Colin Charles) Date: Tue, 17 May 2005 18:50:57 +1000 Subject: Whither nedit? In-Reply-To: References: Message-ID: <1116319857.4564.237.camel@arena.soho.bytebot.net> On Thu, 2005-05-12 at 10:36 -0500, Ian Pilcher wrote: > I just noticed that nedit is no longer present in Fedora Core 4 test > 3. > Is there a plan to bring it back or move it to Fedora Extras? (I > would > gladly volunteer to maintain a Fedora Extras package.) Get a sponsor (by posting to fedora-extras-list at redhat.com), and you can import the package from Core CVS directly into Extras CVS without too much issue as the package maintainer -- Colin Charles, http://www.bytebot.net/ From aani_avni at yahoo.com Tue May 17 09:08:44 2005 From: aani_avni at yahoo.com (avni thacker) Date: Tue, 17 May 2005 02:08:44 -0700 (PDT) Subject: mounting ntfs on fedora Message-ID: <20050517090845.42215.qmail@web54001.mail.yahoo.com> i m using FC1 can any body knows why ntfs partion cannot be mounted on linux or why does linux does not recognize ntfs partion. i hv grub loader with windows 2000 professional installed. --------------------------------- Yahoo! Mail Mobile Take Yahoo! Mail with you! Check email on your mobile phone. -------------- next part -------------- An HTML attachment was scrubbed... URL: From paul at city-fan.org Tue May 17 09:14:09 2005 From: paul at city-fan.org (Paul Howarth) Date: Tue, 17 May 2005 10:14:09 +0100 Subject: mounting ntfs on fedora In-Reply-To: <20050517090845.42215.qmail@web54001.mail.yahoo.com> References: <20050517090845.42215.qmail@web54001.mail.yahoo.com> Message-ID: <4289B5E1.2070207@city-fan.org> avni thacker wrote: > i m using FC1 can any body knows why ntfs partion cannot be mounted on > linux or why does linux does not recognize ntfs partion. i hv grub > loader with windows 2000 professional installed. Solution: http://www.fedorafaq.org/fc1/#NTFS Please note that this mailing list is for development discussions only, and as such your question was posted to an inappropriate list. The first step in resolving this sort of issue should be to search (e.g. using google) for solutions, and only if you still cannot find a solution, post to a user-support forum such as fedoraforum.org or the fedora-list mailing list. Regards, Paul. From nicolas.mailhot at laposte.net Tue May 17 09:44:19 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Tue, 17 May 2005 11:44:19 +0200 (CEST) Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> Message-ID: <22564.192.54.193.28.1116323059.squirrel@rousalka.dyndns.org> On Mar 17 mai 2005 9:29, Mike Honeyfield a ?crit : > On 5/17/05, Arjan van de Ven wrote: >> On Tue, 2005-05-17 at 10:45 +1200, Mike Honeyfield wrote: >> > On 5/17/05, Ignacio Vazquez-Abrams wrote: >> > > On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: >> > > > I am just wondering the reasoning behind having the experimental >> > > > option "Use register arguments" on in the kernel config. >> > > > >> > > > It has caused some issues for me, have since fixed it, but was >> > > > wondering why an experimental feature that could/can break the ABI >> for >> > > > 3rd party binary only modules would be enabled? >> > > >> > > Speed. Pushing data onto the stack takes longer than just shoving it >> > > into a register. >> > > >> > >> > So the speed improvement is enough to justify this? >> > >> > I could understand Fedora Core's kernels have this feature, but was a >> > bit suprised to see the same issue in RHEL. >> >> why? >> >> This only impacts binary modules, and in the process of porting those to >> 2.6 the vendor needs to "fix" this up just as well. That is the same for >> FC and RHEL. >> > > So you disagree that this is actually "experimental" as per the kernel > config? > > Doesn't the term experimental indirectly imply possible stability > issues? Not reasuring IMHO. Large parts of the Linux kernel are marked "Experimental". It really does not mean much - people are so conservative removing those warnings they are kept till the code is old and crufty. They're about as informative as all the commercial EULAs that say no one will be held responsible if $BIG_VENDOR_APP eats your data. > Yes, RH's choice in kernel configure option is theirs and yes, it is > true, binary only vendor can "fix" (not sure waht the quotes imply), > we have "fixed" our drivers. Worth noting, our driver was written for > 2.6 and in RHEL it was broken, but not other platforms we support, > like debian and suse. Different linux vendors make changes at a different pace. I'm sure you can find "experimental" options turned on is Suse that are still disabled in RHEL (at one point this was the case for ALSA for example) There is little point complaining about it - something that appears in RHEL will hit Suse and Debian after a few months too (similarly something appearing in FC will hit RHEL later) If you want seamless Linux support you have to make sure your drivers work on the Distribution development versions (Rawhide->FC->RHEL, Mdk cooker, etc) not wait till they are deployed on RHEL and big iron. The whole development process is public - you have no excuse not being ready by RHEL time. RHEL is about stability, not stagnation. Stability is achieved by giving loads of forwarning about what will end up in RHEL so everyone can prepare for it not by freezing RHEL for a decade. Proprietary vendors that start looking at the RHEL featureset months after it has been released and blame RH for they poor support make me mad. (we have costly support, and it will tell you we're not ready - right) RHEL release is not the start of the adaptation race. It's the end of it. -- Nicolas Mailhot From mike.honeyfield at gmail.com Tue May 17 09:50:10 2005 From: mike.honeyfield at gmail.com (Mike Honeyfield) Date: Tue, 17 May 2005 21:50:10 +1200 Subject: "use register arguments" option enabled. In-Reply-To: <22564.192.54.193.28.1116323059.squirrel@rousalka.dyndns.org> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> <22564.192.54.193.28.1116323059.squirrel@rousalka.dyndns.org> Message-ID: On 5/17/05, Nicolas Mailhot wrote: > RHEL is about stability, not stagnation. Stability is achieved by giving > loads of forwarning about what will end up in RHEL so everyone can prepare > for it not by freezing RHEL for a decade. Proprietary vendors that start > looking at the RHEL featureset months after it has been released and blame > RH for they poor support make me mad. (we have costly support, and it will > tell you we're not ready - right) Wasn't blaming RH for anything, was curious about the whys... not pointing fingers ;) and the whys have been said. Mike From nicolas.mailhot at laposte.net Tue May 17 09:58:50 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Tue, 17 May 2005 11:58:50 +0200 (CEST) Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> <22564.192.54.193.28.1116323059.squirrel@rousalka.dyndns.org> Message-ID: <57245.192.54.193.28.1116323930.squirrel@rousalka.dyndns.org> On Mar 17 mai 2005 11:50, Mike Honeyfield a ?crit : > On 5/17/05, Nicolas Mailhot wrote: >> RHEL is about stability, not stagnation. Stability is achieved by giving >> loads of forwarning about what will end up in RHEL so everyone can >> prepare >> for it not by freezing RHEL for a decade. Proprietary vendors that start >> looking at the RHEL featureset months after it has been released and >> blame >> RH for they poor support make me mad. (we have costly support, and it >> will >> tell you we're not ready - right) > > Wasn't blaming RH for anything, was curious about the whys... not > pointing fingers ;) Oh I'm not writing this especially about you, I've had bad experiences in the past that's all :(. Userspace linux commercial support is even worse - some companies still spec new versions on the RH 7.2 compatibility libraries. -- Nicolas Mailhot From mike at navi.cx Tue May 17 11:32:03 2005 From: mike at navi.cx (Mike Hearn) Date: Tue, 17 May 2005 12:32:03 +0100 Subject: "use register arguments" option enabled. References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116284519.7161.7.camel@ignacio.ignacio.lan> Message-ID: On Mon, 16 May 2005 19:01:59 -0400, Ignacio Vazquez-Abrams wrote: > Even a 2 clock cycle delay happening hundreds of thousands or even > millions of times a second builds up. I'd really like to see hard statistics about this. If I recall correctly, Ulrich Drepper did some experiments with using -mregparm for internal calls inside glibc and found it made no difference, or in some cases made things worse. There are enough "intuitively" correct optimisations that slow things down out there already ... thanks -mike From arjanv at redhat.com Tue May 17 11:44:32 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Tue, 17 May 2005 13:44:32 +0200 Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116284519.7161.7.camel@ignacio.ignacio.lan> Message-ID: <1116330272.6442.48.camel@laptopd505.fenrus.org> On Tue, 2005-05-17 at 12:32 +0100, Mike Hearn wrote: > On Mon, 16 May 2005 19:01:59 -0400, Ignacio Vazquez-Abrams wrote: > > Even a 2 clock cycle delay happening hundreds of thousands or even > > millions of times a second builds up. > > I'd really like to see hard statistics about this. If I recall correctly, > Ulrich Drepper did some experiments with using -mregparm for internal > calls inside glibc and found it made no difference, or in some cases made > things worse. There are enough "intuitively" correct optimisations that > slow things down out there already .. actually glibc mostly is lots of function wrappers, where it indeed helps little. For the kernel it actually shrunk code some, and at the time I looked over the assembly generated and it looked a lot better too. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From buildsys at redhat.com Tue May 17 12:00:53 2005 From: buildsys at redhat.com (Build System) Date: Tue, 17 May 2005 08:00:53 -0400 Subject: rawhide report: 20050517 changes Message-ID: <200505171200.j4HC0r6m024570@porkchop.devel.redhat.com> Updated Packages: NetworkManager-0.4-15.cvs20050404 --------------------------------- * Mon May 16 2005 Dan Williams - 0.4-15.cvs30050404 - Fix dispatcher and applet CFLAGS so they gets compiled with FORTIFY_SOURCE * Mon May 16 2005 Dan Williams - 0.4-14.cvs30050404 - Fix segfault in NetworkManagerDispatcher, add an initscript for it * Mon May 16 2005 Dan Williams - 0.4-13.cvs30050404 - Fix condition that may have resulted in DHCP client returning success when it really timed out alsa-utils-1.0.9rc2-2 --------------------- * Mon May 16 2005 Bill Nottingham 1.0.9rc2-2 - make sure 'Wave' playback channel isn't muted (#157850) hwdata-0.158-1 -------------- * Mon May 16 2005 Bill Nottingham - 0.158-1 - add a orinoco card (#157482) kernel-2.6.11-1.1312_FC4 ------------------------ * Sat May 14 2005 Dave Jones - Update E1000 driver from netdev-2.6 tree. * Fri May 13 2005 Dave Jones - Bump maximum supported CPUs on x86-64 to 32. - Tickle the NMI watchdog when we're doing serial writes. - SCSI CAM geometry fix. - Slab debug single-bit error improvement. libsepol-1.5.8-1 ---------------- * Mon May 16 2005 Dan Walsh 1.5.8-1 - Upgrade to latest from NSA * Added sepol_ prefix to Flask types to avoid namespace collision with libselinux. * Fri May 13 2005 Dan Walsh 1.5.7-1 - Upgrade to latest from NSA * Added sepol_compute_av_reason() for audit2why. mdadm-1.11.0-4.fc4 ------------------ * Mon May 16 2005 Doug Ledford 1.11.0-4.fc4 - Make the mdmonitor init script use the pid-file option, major cleanup of the script now possible (#134459) * Mon May 16 2005 Doug Ledford 1.11.0-3.fc4 - Put back the obsoletes: raidtools that was present in 1.11.0-1.fc4 * Mon May 16 2005 Doug Ledford 1.11.0-2.fc4 - Change the default auto= mode so it need not be on the command line to work with udev, however it is still supported on the command line (#132706) - Add a man page (from Luca Berra) for mdassemble openssh-4.0p1-3 --------------- * Mon May 16 2005 Tomas Mraz 4.0p1-3 - link libselinux only to sshd (#157678) python-2.4.1-2 -------------- * Wed Apr 20 2005 Mihai Ibanescu 2.4.1-2 - Fixed bug #143667 (python should own /usr/lib/python* on 64-bit systems, for noarch packages) - Fixed bug #143419 (BuildRequires db4 is not versioned) squid-7:2.5.STABLE9-7 --------------------- * Mon May 16 2005 Jay Fenlason 7:2.5.STABLE9-7 - Upgrade the upstream -dns_query patch from -4 to -5 From bgerst at didntduck.org Tue May 17 12:19:07 2005 From: bgerst at didntduck.org (Brian Gerst) Date: Tue, 17 May 2005 08:19:07 -0400 Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> Message-ID: <4289E13B.1030407@didntduck.org> Mike Honeyfield wrote: > On 5/17/05, Arjan van de Ven wrote: > >>On Tue, 2005-05-17 at 10:45 +1200, Mike Honeyfield wrote: >> >>>On 5/17/05, Ignacio Vazquez-Abrams wrote: >>> >>>>On Tue, 2005-05-17 at 10:29 +1200, Mike Honeyfield wrote: >>>> >>>>>I am just wondering the reasoning behind having the experimental >>>>>option "Use register arguments" on in the kernel config. >>>>> >>>>>It has caused some issues for me, have since fixed it, but was >>>>>wondering why an experimental feature that could/can break the ABI for >>>>>3rd party binary only modules would be enabled? >>>> >>>>Speed. Pushing data onto the stack takes longer than just shoving it >>>>into a register. >>>> >>> >>>So the speed improvement is enough to justify this? >>> >>>I could understand Fedora Core's kernels have this feature, but was a >>>bit suprised to see the same issue in RHEL. >> >>why? >> >>This only impacts binary modules, and in the process of porting those to >>2.6 the vendor needs to "fix" this up just as well. That is the same for >>FC and RHEL. >> > > > So you disagree that this is actually "experimental" as per the kernel config? > > Doesn't the term experimental indirectly imply possible stability > issues? Not reasuring IMHO. It is no longer experimental, even though the tag wasn't removed. It was only marked experimental because old versions of GCC had bugs with respect to -mregparm. -- Brian Gerst From sds at tycho.nsa.gov Tue May 17 12:17:39 2005 From: sds at tycho.nsa.gov (Stephen Smalley) Date: Tue, 17 May 2005 08:17:39 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116265016.15526.3.camel@localhost.localdomain> References: <200505140246.36294.russell@coker.com.au> <200505162245.13035.russell@coker.com.au> <1116256384.9888.12.camel@localhost.localdomain> <200505170144.46594.russell@coker.com.au> <1116265016.15526.3.camel@localhost.localdomain> Message-ID: <1116332259.917.33.camel@moss-spartans.epoch.ncsc.mil> On Mon, 2005-05-16 at 13:36 -0400, Peter Jones wrote: > Why can't the glibc package provide a ld.so.cache which simply indicates > no libraries? This seems more correct to me, especially as it claims > ownership of the file. That won't help if ldconfig re-creates the file each time (vs. just rewriting the existing file in place), as the new file would still be labeled in accordance with the default behavior unless ldconfig were modified. But in any event, it sounds like we need to determine why ldconfig isn't running the proper domain, as that would suffice to ensuring that ld.so.cache is labeled properly. -- Stephen Smalley National Security Agency From spam at tachegroup.com Tue May 17 13:40:44 2005 From: spam at tachegroup.com (TGS) Date: Tue, 17 May 2005 09:40:44 -0400 Subject: Rebuild installation CDROMs Message-ID: Hello, How do I get information for rebuilding the CDROMs used for installation? I need to rebuild the kernel used during boot, and install into the isolinux directory. I have found a bit of information on the web, and am mainly looking to see how to build the initrd.img located in the isolinux directory. It seems quite a bit bigger than the initrd that I build from source. From spam at tachegroup.com Tue May 17 13:44:28 2005 From: spam at tachegroup.com (TGS) Date: Tue, 17 May 2005 09:44:28 -0400 Subject: How to rebuild the kernel src rpm. Message-ID: How do I rebuild the kernel src rpm, after I have rebuilt a new version from the src? I have created a patch to the kernel, tweaked the kernel-2.6.spec, and built the binary rpms for the kernel just fine. Problem is that I cannot see how I build the src version of the rpm with my patch included. From sundaram at redhat.com Tue May 17 13:50:58 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Tue, 17 May 2005 19:20:58 +0530 Subject: How to rebuild the kernel src rpm. In-Reply-To: References: Message-ID: <4289F6C2.2070509@redhat.com> TGS wrote: > How do I rebuild the kernel src rpm, after I have rebuilt a new > version from the src? > > I have created a patch to the kernel, tweaked the kernel-2.6.spec, and > built the binary rpms for the kernel just fine. Problem is that I > cannot see how I build the src version of the rpm with my patch included. > use rpmbuild -ba --target=// kernel.spec instead of rpmbuild -bp as specified in the release notes. Note that this is off topic here. http://fedoraproject.org/wiki/PostIsOffTopic regards Rahul From sundaram at redhat.com Tue May 17 13:51:44 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Tue, 17 May 2005 19:21:44 +0530 Subject: Rebuild installation CDROMs In-Reply-To: References: Message-ID: <4289F6F0.7030901@redhat.com> TGS wrote: > Hello, > > How do I get information for rebuilding the CDROMs used for > installation? I need to rebuild the kernel used during boot, and > install into the isolinux directory. I have found a bit of information > on the web, and am mainly looking to see how to build the initrd.img > located in the isolinux directory. It seems quite a bit bigger than > the initrd that I build from source. > Off topic here. Use the Fedora users list instead http://fedoraproject.org/wiki/PostIsOffTopic regards Rahul From paul at city-fan.org Tue May 17 13:51:42 2005 From: paul at city-fan.org (Paul Howarth) Date: Tue, 17 May 2005 14:51:42 +0100 Subject: How to rebuild the kernel src rpm. In-Reply-To: References: Message-ID: <4289F6EE.5090003@city-fan.org> TGS wrote: > How do I rebuild the kernel src rpm, after I have rebuilt a new version > from the src? > > I have created a patch to the kernel, tweaked the kernel-2.6.spec, and > built the binary rpms for the kernel just fine. Problem is that I cannot > see how I build the src version of the rpm with my patch included. Use "rpmbuild -ba" instead of "rpmbuild -bb"? Paul. From spam at tachegroup.com Tue May 17 14:01:01 2005 From: spam at tachegroup.com (TGS) Date: Tue, 17 May 2005 10:01:01 -0400 Subject: How to rebuild the kernel src rpm. In-Reply-To: <4289F6EE.5090003@city-fan.org> References: <4289F6EE.5090003@city-fan.org> Message-ID: Thanks, I thought I was doing that, so I will double check - sorry for the postings to this group instead of the user group. On May 17, 2005, at 9:51 AM, Paul Howarth wrote: > TGS wrote: >> How do I rebuild the kernel src rpm, after I have rebuilt a new >> version from the src? >> I have created a patch to the kernel, tweaked the kernel-2.6.spec, >> and built the binary rpms for the kernel just fine. Problem is that I >> cannot see how I build the src version of the rpm with my patch >> included. > > Use "rpmbuild -ba" instead of "rpmbuild -bb"? > > Paul. > > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list > From dwmw2 at infradead.org Tue May 17 14:47:29 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 17 May 2005 15:47:29 +0100 Subject: "use register arguments" option enabled. In-Reply-To: <42893634.7040501@redhat.com> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116284519.7161.7.camel@ignacio.ignacio.lan> <42893634.7040501@redhat.com> Message-ID: <1116341249.23972.91.camel@hades.cambridge.redhat.com> On Mon, 2005-05-16 at 14:09 -1000, Warren Togami wrote: > > Even a 2 clock cycle delay happening hundreds of thousands or even > > millions of times a second builds up. > > Isn't using system memory much longer than 2 clock cycles? If accessing the top of your stack is "using system memory" you're screwed already. It ought to be in L1 cache. -- dwmw2 From alan at redhat.com Tue May 17 14:56:41 2005 From: alan at redhat.com (Alan Cox) Date: Tue, 17 May 2005 10:56:41 -0400 Subject: "use register arguments" option enabled. In-Reply-To: <1116341249.23972.91.camel@hades.cambridge.redhat.com> References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116284519.7161.7.camel@ignacio.ignacio.lan> <42893634.7040501@redhat.com> <1116341249.23972.91.camel@hades.cambridge.redhat.com> Message-ID: <20050517145641.GA12925@devserv.devel.redhat.com> On Tue, May 17, 2005 at 03:47:29PM +0100, David Woodhouse wrote: > If accessing the top of your stack is "using system memory" you're > screwed already. It ought to be in L1 cache. It will normally be in L1, but you've still got penalties compared to register access because of things like %esp interlocks as well as being longer and using more cache From mike at flyn.org Tue May 17 15:03:30 2005 From: mike at flyn.org (W. Michael Petullo) Date: Tue, 17 May 2005 10:03:30 -0500 (CDT) Subject: pam_ccreds and Fedora Message-ID: <4894.66.151.13.104.1116342210.squirrel@zero.voxel.net> I have been using Fedora Core's pam_ccreds package to allow my laptop to authenticate users even when it is disconnected from my network's LDAP server[1]. Recently, logging in to my computer when disconnected began to fail. It seems that I was incorrectly relying on nscd to cache information for long periods of time. Bug 150748 fixed nscd, but made it difficult to abuse it in the way I require. After doing some research, I found nss_updatedb, a utility that maintains a local cache of network directory user and group information. However, nss_updatedb is not included in Fedora Core. What is the preferred way to use pam_ccreds on Fedora? Is anyone else using this PAM module? Is nss_updatedb a prerequisite and, if so, will it be packaged for Fedora? I think disconnected authentication is an important feature for Fedora and would like to help work on it. [1] See the thread that starts at: http://www.redhat.com/archives/fedora-devel-list/2004-May/msg00368.html -- Mike From Fedora at TQMcube.com Tue May 17 15:10:02 2005 From: Fedora at TQMcube.com (David Cary Hart) Date: Tue, 17 May 2005 11:10:02 -0400 Subject: Evolution 2.3 Message-ID: <1116342602.15557.12.camel@dch.tqmcube.com> Evolution 2.3.1 was released on April 26 and is not yet in rawhide. Are the odd numbered releases considered test versions? -- Multi-RBL Check: http://www.TQMcube.com/rblcheck.htm Kill Spam at the Source: http://www.TQMcube.com/spam_trap.htm Today's Spam Trap Adds: http://www.TQMcube.com/BlockedToday RBLDNSD HowTo: http://www.TQMcube.com/rbldnsd.htm From sundaram at redhat.com Tue May 17 15:18:51 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Tue, 17 May 2005 20:48:51 +0530 Subject: Evolution 2.3 In-Reply-To: <1116342602.15557.12.camel@dch.tqmcube.com> References: <1116342602.15557.12.camel@dch.tqmcube.com> Message-ID: <428A0B5B.6010306@redhat.com> David Cary Hart wrote: >Evolution 2.3.1 was released on April 26 and is not yet in rawhide. Are >the odd numbered releases considered test versions? > > Yes. Evolution is part of GNOME now and follows the classical Linux kernel naming scheme regards Rahul From pjones at redhat.com Tue May 17 17:45:40 2005 From: pjones at redhat.com (Peter Jones) Date: Tue, 17 May 2005 13:45:40 -0400 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <200505171405.34672.russell@coker.com.au> References: <200505140246.36294.russell@coker.com.au> <200505160106.32177.russell@coker.com.au> <1116272126.3360.18.camel@bree.local.net> <200505171405.34672.russell@coker.com.au> Message-ID: <1116351940.8907.6.camel@localhost.localdomain> On Tue, 2005-05-17 at 14:05 +1000, Russell Coker wrote: > On Tuesday 17 May 2005 05:35, Jeremy Katz wrote: > > We never used label'ing of things in the initrd when it was an ext2 > > image. The loader explicitly sets the exec context before running > > anaconda to be system_u:object_r:anaconda_t if policy doesn't fail to > > load. Look in /tmp/anaconda.log (or tty3) for errors about loading the > > policy or setting the context. > > That's an invalid context. The correct value should be > system_u:system_r:anaconda_t. The role object_r is only suitable for files > on disk. The kernel does allow setting it though. Fixed in cvs. -- Peter From pmatilai at welho.com Tue May 17 18:53:44 2005 From: pmatilai at welho.com (Panu Matilainen) Date: Tue, 17 May 2005 21:53:44 +0300 Subject: pam_ccreds and Fedora In-Reply-To: <4894.66.151.13.104.1116342210.squirrel@zero.voxel.net> References: <4894.66.151.13.104.1116342210.squirrel@zero.voxel.net> Message-ID: <1116356024.7953.9.camel@weasel.laiskiainen.org> On Tue, 2005-05-17 at 10:03 -0500, W. Michael Petullo wrote: > I have been using Fedora Core's pam_ccreds package to allow my laptop to > authenticate users even when it is disconnected from my network's LDAP > server[1]. Recently, logging in to my computer when disconnected began to > fail. > > It seems that I was incorrectly relying on nscd to cache information for > long periods of time. Bug 150748 fixed nscd, but made it difficult to > abuse it in the way I require. > > After doing some research, I found nss_updatedb, a utility that maintains > a local cache of network directory user and group information. However, > nss_updatedb is not included in Fedora Core. > > What is the preferred way to use pam_ccreds on Fedora? Is anyone else > using this PAM module? Is nss_updatedb a prerequisite and, if so, will it > be packaged for Fedora? > > I think disconnected authentication is an important feature for Fedora and > would like to help work on it. You don't really need nss_updatedb, in fact nss_updatedb is totally unusable in *big* environments), nscd does all the necessary caching as of FC3 and beyond. What IS missing is integration of pam_ccreds into authconfig. There's a bug about it somewhere in RH bugzilla and apparently there's been (an RH internal) patch to authconfig floating around to add the support for configuring pam_ccreds, too bad it hasn't made the broad daylights so far despite me asking on a few occasions :-/ - Panu - From jmorris at redhat.com Tue May 17 19:01:59 2005 From: jmorris at redhat.com (James Morris) Date: Tue, 17 May 2005 15:01:59 -0400 (EDT) Subject: Kernel panic during boot with latest rawhide kernel Message-ID: I'm getting this with kernel-smp-2.6.11-1.1312_FC4, system fully updated to rawhide: Linux version 2.6.11-1.1312_FC4smp (bhcompile at decompose.build.redhat.com) (gcc version 4.0.0 20050512 (Red Hat 4.0.0-5)) #1 SMP Sat May 14 16:33:46 EDT 2005 BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 00000000000a0000 (usable) BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved) BIOS-e820: 0000000000100000 - 000000003ff75000 (usable) BIOS-e820: 000000003ff75000 - 000000003ff77000 (ACPI NVS) BIOS-e820: 000000003ff77000 - 000000003ff98000 (ACPI data) BIOS-e820: 000000003ff98000 - 0000000040000000 (reserved) BIOS-e820: 00000000fec00000 - 00000000fec90000 (reserved) BIOS-e820: 00000000fee00000 - 00000000fee10000 (reserved) BIOS-e820: 00000000ffb00000 - 0000000100000000 (reserved) 127MB HIGHMEM available. 896MB LOWMEM available. found SMP MP-table at 000fe710 Using x86 segment limits to approximate NX protection DMI 2.3 present. Using APIC driver default ACPI: PM-Timer IO Port: 0x808 ACPI: LAPIC (acpi_id[0x01] lapic_id[0x00] enabled) Processor #0 15:2 APIC version 20 ACPI: LAPIC (acpi_id[0x02] lapic_id[0x02] enabled) Processor #2 15:2 APIC version 20 ACPI: LAPIC (acpi_id[0x03] lapic_id[0x01] enabled) Processor #1 15:2 APIC version 20 ACPI: LAPIC (acpi_id[0x04] lapic_id[0x03] enabled) Processor #3 15:2 APIC version 20 ACPI: IOAPIC (id[0x04] address[0xfec00000] gsi_base[0]) IOAPIC[0]: apic_id 4, version 32, address 0xfec00000, GSI 0-23 ACPI: IOAPIC (id[0x05] address[0xfec80000] gsi_base[24]) IOAPIC[1]: apic_id 5, version 32, address 0xfec80000, GSI 24-47 ACPI: IOAPIC (id[0x06] address[0xfec80800] gsi_base[48]) IOAPIC[2]: apic_id 6, version 32, address 0xfec80800, GSI 48-71 ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl) ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level) Enabling APIC mode: Flat. Using 3 I/O APICs Using ACPI (MADT) for SMP configuration information Allocating PCI resources starting at 40000000 (gap: 40000000:bec00000) Built 1 zonelists Kernel command line: ro root=LABEL=/1 console=tty1 console=ttyS0,115200 selinux=1 Initializing CPU#0 CPU 0 irqstacks, hard=c042c000 soft=c040c000 PID hash table entries: 4096 (order: 12, 65536 bytes) Detected 1994.223 MHz processor. Using pmtmr for high-res timesource Console: colour VGA+ 80x25 Dentry cache hash table entries: 131072 (order: 7, 524288 bytes) Inode-cache hash table entries: 65536 (order: 6, 262144 bytes) Memory: 1032276k/1048020k available (2086k kernel code, 14900k reserved, 776k data, 232k init, 130516k highmem) Checking if this processor honours the WP bit even in supervisor mode... Ok. Security Framework v1.0.0 initialized SELinux: Initializing. SELinux: Starting in permissive mode selinux_register_security: Registering secondary module capability Capability LSM initialized as secondary Mount-cache hash table entries: 512 CPU: Trace cache: 12K uops, L1 D cache: 8K CPU: L2 cache: 512K CPU: Physical Processor ID: 0 Intel machine check architecture supported. Intel machine check reporting enabled on CPU#0. CPU0: Intel P4/Xeon Extended MCE MSRs (12) available CPU0: Thermal monitoring enabled Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Checking 'hlt' instruction... OK. CPU0: Intel(R) Xeon(TM) CPU 2.00GHz stepping 07 Booting processor 1/1 eip 3000 CPU 1 irqstacks, hard=c042d000 soft=c040d000 Initializing CPU#1 CPU: Trace cache: 12K uops, L1 D cache: 8K CPU: L2 cache: 512K CPU: Physical Processor ID: 0 Intel machine check architecture supported. Intel machine check reporting enabled on CPU#1. CPU1: Intel P4/Xeon Extended MCE MSRs (12) available CPU1: Thermal monitoring enabled CPU1: Intel(R) Xeon(TM) CPU 2.00GHz stepping 07 Booting processor 2/2 eip 3000 CPU 2 irqstacks, hard=c042e000 soft=c040e000 Initializing CPU#2 CPU: Trace cache: 12K uops, L1 D cache: 8K CPU: L2 cache: 512K CPU: Physical Processor ID: 3 Intel machine check architecture supported. Intel machine check reporting enabled on CPU#2. CPU2: Intel P4/Xeon Extended MCE MSRs (12) available CPU2: Thermal monitoring enabled CPU2: Intel(R) Xeon(TM) CPU 2.00GHz stepping 07 Booting processor 3/3 eip 3000 CPU 3 irqstacks, hard=c042f000 soft=c040f000 Initializing CPU#3 CPU: Trace cache: 12K uops, L1 D cache: 8K CPU: L2 cache: 512K CPU: Physical Processor ID: 3 Intel machine check architecture supported. Intel machine check reporting enabled on CPU#3. CPU3: Intel P4/Xeon Extended MCE MSRs (12) available CPU3: Thermal monitoring enabled CPU3: Intel(R) Xeon(TM) CPU 2.00GHz stepping 07 Total of 4 processors activated (15884.28 BogoMIPS). ENABLING IO-APIC IRQs ..TIMER: vector=0x31 pin1=2 pin2=-1 checking TSC synchronization across 4 CPUs: passed. softlockup thread 0 started up. softlockup thread 1 started up. softlockup thread 2 started up. Brought up 4 CPUs softlockup thread 3 started up. checking if image is initramfs... it is Freeing initrd memory: 1068k freed NET: Registered protocol family 16 PCI: PCI BIOS revision 2.10 entry at 0xfbdf9, last bus=5 PCI: Using configuration type 1 mtrr: v2.0 (20020519) ACPI: Subsystem revision 20050309 ACPI: Interpreter enabled ACPI: Using IOAPIC for interrupt routing ACPI: PCI Root Bridge [PCI0] (0000:00) PCI: Probing PCI hardware (bus 00) PCI: Ignoring BAR0-3 of IDE controller 0000:00:1f.1 PCI: Transparent bridge - 0000:00:1e.0 ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 5 6 7 9 10 *11 12 15) ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 6 7 9 10 *11 12 15) ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 5 6 7 *9 10 11 12 15) ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 6 7 9 *10 11 12 15) ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 5 6 7 9 10 11 12 15) *0, disabled. ACPI: PCI Interrupt Link [LNKF] (IRQs 3 4 5 6 7 9 10 *11 12 15) ACPI: PCI Interrupt Link [LNKG] (IRQs 3 4 5 6 7 9 10 11 12 15) *0, disabled. ACPI: PCI Interrupt Link [LNKH] (IRQs 3 4 5 6 7 9 10 *11 12 15) Linux Plug and Play Support v0.97 (c) Adam Belay pnp: PnP ACPI init pnp: PnP ACPI: found 11 devices usbcore: registered new driver usbfs usbcore: registered new driver hub PCI: Using ACPI for IRQ routing PCI: If a device doesn't work, try "pci=routeirq". If it helps, post a report pnp: 00:0a: ioport range 0x800-0x85f could not be reserved pnp: 00:0a: ioport range 0xc00-0xc7f has been reserved pnp: 00:0a: ioport range 0x860-0x8ff could not be reserved Simple Boot Flag value 0x87 read from CMOS RAM was invalid Simple Boot Flag at 0x7a set to 0x1 apm: BIOS version 1.2 Flags 0x03 (Driver version 1.16ac) apm: disabled - APM is not SMP safe. audit: initializing netlink socket (disabled) audit(1116354771.207:0): initialized highmem bounce pool size: 64 pages Total HugeTLB memory allocated, 0 VFS: Disk quotas dquot_6.5.1 Dquot-cache hash table entries: 1024 (order 0, 4096 bytes) SELinux: Registering netfilter hooks Initializing Cryptographic API ksign: Installing public key data Loading keyring - Added public key 84EB29B9E4457CD - User ID: Red Hat, Inc. (Kernel Module GPG key) pci_hotplug: PCI Hot Plug PCI Core version: 0.5 isapnp: Scanning for PnP cards... isapnp: No Plug & Play device found Real Time Clock Driver v1.12 Linux agpgart interface v0.101 (c) Dave Jones agpgart: Detected an Intel E7505 Chipset. agpgart: AGP aperture is 128M @ 0xe8000000 PNP: PS/2 controller doesn't have KBD irq; using default 0x1 PNP: PS/2 Controller [PNP0f13:MOU] at 0x60,0x64 irq 1,12 serio: i8042 AUX port at 0x60,0x64 irq 12 serio: i8042 KBD port at 0x60,0x64 irq 1 Serial: 8250/16550 driver $Revision: 1.90 $ 76 ports, IRQ sharing enabled ?ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A io scheduler noop registered io scheduler anticipatory registered io scheduler deadline registered io scheduler cfq registered RAMDISK driver initialized: 16 RAM disks of 16384K size 1024 blocksize Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx ICH4: IDE controller at PCI slot 0000:00:1f.1 PCI: Enabling device 0000:00:1f.1 (0005 -> 0007) ACPI: PCI Interrupt 0000:00:1f.1[A] -> GSI 18 (level, low) -> IRQ 169 ICH4: chipset revision 1 ICH4: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xffa0-0xffa7, BIOS settings: hda:DMA, hdb:pio ide1: BM-DMA at 0xffa8-0xffaf, BIOS settings: hdc:DMA, hdd:pio hda: Maxtor 6E040L0, ATA DISK drive ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 hdc: HL-DT-ST GCE-8481B, ATAPI CD/DVD-ROM drive ide1 at 0x170-0x177,0x376 on irq 15 hda: max request size: 128KiB hda: 80293248 sectors (41110 MB) w/2048KiB Cache, CHS=65535/16/63, UDMA(100) hda: cache flushes supported hda: hda1 hda2 hda3 hda4 < hda5 hda6 > hdc: ATAPI 40X CD-ROM CD-R/RW drive, 2048kB Cache, UDMA(33) Uniform CD-ROM driver Revision: 3.20 ide-floppy driver 0.99.newide usbcore: registered new driver hiddev usbcore: registered new driver usbhid drivers/usb/input/hid-core.c: v2.01:USB HID core driver mice: PS/2 mouse device common for all mice md: md driver 0.90.1 MAX_MD_DEVS=256, MD_SB_DISKS=27 NET: Registered protocol family 2 IP: routing cache hash table of 4096 buckets, 64Kbytes TCP established hash table entries: 262144 (order: 10, 4194304 bytes) TCP bind hash table entries: 65536 (order: 7, 786432 bytes) TCP: Hash tables configured (established 262144 bind 65536) Initializing IPsec netlink socket NET: Registered protocol family 1 NET: Registered protocol family 17 ACPI wakeup devices: VBTN PCI0 USB0 USB1 USB2 PCI1 PCI2 PCI3 PCI4 MOU ACPI: (supports S0 S1 S3 S4 S5) Freeing unused kernel memory: 232k freed Red Hat nash version 4.2.14 starting Mounted /proc filesystem Mounting sysfs Creating /dev Starting udev Loading jbd.ko module Loading ext3.ko module Creating root device input: ImPS/2 Generic Wheel Mouse on isa0060/serio1 Mounting root filesystem kjournald starting. Commit interval 5 seconds EXT3-fs: mounted filesystem with ordered data mode. Kernel panic - not syncing: Attempted to kill init! [] panic+0x45/0x1e8 [] profile_task_exit+0x31/0x45 [] do_exit+0x25a/0x369 [] do_munmap+0xca/0xf7 [] do_group_exit+0x29/0x90 [] syscall_call+0x7/0xb From arjan at infradead.org Tue May 17 07:36:37 2005 From: arjan at infradead.org (Arjan van de Ven) Date: Tue, 17 May 2005 09:36:37 +0200 Subject: "use register arguments" option enabled. In-Reply-To: References: <1116283373.7161.3.camel@ignacio.ignacio.lan> <1116313174.6442.18.camel@laptopd505.fenrus.org> Message-ID: <1116315398.6442.23.camel@laptopd505.fenrus.org> On Tue, 2005-05-17 at 19:29 +1200, Mike Honeyfield wrote: > On 5/17/05, Arjan van de Ven wrote: > > This only impacts binary modules, and in the process of porting those to > > 2.6 the vendor needs to "fix" this up just as well. That is the same for > > FC and RHEL. > > > > So you disagree that this is actually "experimental" as per the kernel config? yes I do. EXPERIMENTAL in the kernel config in practice is meaningless; it hardly ever gets removed even after years of high production use of stuff. I wonder if the EXPERIMENTAL got removed from EXT3 yet .. ;) > Yes, RH's choice in kernel configure option is theirs and yes, it is > true, binary only vendor can "fix" (not sure waht the quotes imply), > we have "fixed" our drivers. Worth noting, our driver was written for > 2.6 and in RHEL it was broken, but not other platforms we support, > like debian and suse. I assume that since your driver was written with linux in mind, that it is gpl, right? So then you wouldn't have this problem.... (if it's not, make sure your lawyers have read the COPYING.modules file from the kernel rpms) > However, I was curious because it seemed ironic to possibly break ABI > for binary only vendors in the enterprise product. The linux kernel has no ABI. So it's also not possible to break it. SLES uses this option too afaik, and it has been there for a long time in 2.6, any driver not capable of running with it is of course very buggy or at least not adjusted to 2.6 properly. From russell at coker.com.au Wed May 18 06:32:02 2005 From: russell at coker.com.au (Russell Coker) Date: Wed, 18 May 2005 16:32:02 +1000 Subject: SE Linux installer changes needed - was Re: /etc/ld.so.cache and FC4T3 In-Reply-To: <1116351940.8907.6.camel@localhost.localdomain> References: <200505140246.36294.russell@coker.com.au> <200505171405.34672.russell@coker.com.au> <1116351940.8907.6.camel@localhost.localdomain> Message-ID: <200505181632.07374.russell@coker.com.au> On Wednesday 18 May 2005 03:45, Peter Jones wrote: > On Tue, 2005-05-17 at 14:05 +1000, Russell Coker wrote: > > On Tuesday 17 May 2005 05:35, Jeremy Katz wrote: > > > We never used label'ing of things in the initrd when it was an ext2 > > > image. The loader explicitly sets the exec context before running > > > anaconda to be system_u:object_r:anaconda_t if policy doesn't fail to > > > load. Look in /tmp/anaconda.log (or tty3) for errors about loading the > > > policy or setting the context. > > > > That's an invalid context. The correct value should be > > system_u:system_r:anaconda_t. The role object_r is only suitable for > > files on disk. The kernel does allow setting it though. > > Fixed in cvs. I've discovered the root cause of the problem. anaconda.te seems to have been removed from the targeted policy so there is no anaconda_t domain in the policy used for installation. Ideally we want anaconda.te included in the policy for installation but excluded from the policy used for running the system. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From NOS at Utel.no Wed May 18 07:45:08 2005 From: NOS at Utel.no (=?ISO-8859-1?Q?=22Nils_O=2E_Sel=E5sdal=22?=) Date: Wed, 18 May 2005 09:45:08 +0200 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <000001c5570b$b3f1c430$14aaa8c0@utelsystems.local> References: <000001c5570b$b3f1c430$14aaa8c0@utelsystems.local> Message-ID: <428AF284.5070104@Utel.no> Matthew Miller wrote: > Just a thought: > > > /etc/profile.d/tmpdir.sh: Why not try to make use of bind mounts ? mount --bind /home/user/tmp /tmp From russell at coker.com.au Wed May 18 08:48:21 2005 From: russell at coker.com.au (Russell Coker) Date: Wed, 18 May 2005 18:48:21 +1000 Subject: "use register arguments" option enabled. In-Reply-To: <1116341249.23972.91.camel@hades.cambridge.redhat.com> References: <42893634.7040501@redhat.com> <1116341249.23972.91.camel@hades.cambridge.redhat.com> Message-ID: <200505181848.25171.russell@coker.com.au> On Wednesday 18 May 2005 00:47, David Woodhouse wrote: > On Mon, 2005-05-16 at 14:09 -1000, Warren Togami wrote: > > > Even a 2 clock cycle delay happening hundreds of thousands or even > > > millions of times a second builds up. > > > > Isn't using system memory much longer than 2 clock cycles? > > If accessing the top of your stack is "using system memory" you're > screwed already. It ought to be in L1 cache. I presume that the top of the stack will eventually be written back to main memory. Aren't memory writes more of a performance hit than reads, especially on SMP? -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From mattdm at mattdm.org Wed May 18 12:37:01 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Wed, 18 May 2005 08:37:01 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <428AF284.5070104@Utel.no> References: <000001c5570b$b3f1c430$14aaa8c0@utelsystems.local> <428AF284.5070104@Utel.no> Message-ID: <20050518123701.GA17975@jadzia.bu.edu> On Wed, May 18, 2005 at 09:45:08AM +0200, "Nils O. Sel?sdal" wrote: > >/etc/profile.d/tmpdir.sh: > Why not try to make use of bind mounts ? > mount --bind /home/user/tmp /tmp 'cause: mount: only root can do that -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 72 degrees Fahrenheit. From rdieter at math.unl.edu Wed May 18 14:08:40 2005 From: rdieter at math.unl.edu (Rex Dieter) Date: Wed, 18 May 2005 09:08:40 -0500 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <428AF284.5070104@Utel.no> References: <000001c5570b$b3f1c430$14aaa8c0@utelsystems.local> <428AF284.5070104@Utel.no> Message-ID: Nils O. Sel?sdal wrote: > Matthew Miller wrote: > >> Just a thought: >> >> >> /etc/profile.d/tmpdir.sh: > > Why not try to make use of bind mounts ? > mount --bind /home/user/tmp /tmp Wouldn't that (obviously) fail for multi-user systems? -- Rex From buildsys at redhat.com Wed May 18 15:16:45 2005 From: buildsys at redhat.com (Build System) Date: Wed, 18 May 2005 11:16:45 -0400 Subject: rawhide report: 20050518 changes Message-ID: <200505181516.j4IFGjcs001754@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.6-20050517.141233.FC4.0 ----------------------------------------- * Mon May 16 2005 Chris Feist - Add depmod in %post scriplet. anaconda-10.1.0.2-1 ------------------- * Thu Oct 28 2004 Jeremy Katz - 10.1.0.2-1 - Bring in firefox on upgrade if mozilla/netscape were previously installed (#137244) * Wed Oct 27 2004 Jeremy Katz - 10.1.0.1-1 - Punjabi shouldn't try to do text mode (#137030) - Fix traceback on upgrade (#137345) * Wed Oct 20 2004 Jeremy Katz - 10.1.0.0-1 - Lowercase OSA addresses from a parm file too (karsten) - Turn off beta nag for FC3 release anaconda-10.2.0.63-1 -------------------- * Tue May 17 2005 Jeremy Katz - 10.2.0.63-1 - add arch to buildstamp (notting, #151927) - Fix am.po format strings * Tue May 17 2005 Jeremy Katz - 10.2.0.62-1 - Fix execcon used for anaconda (pjones) - Fix traceback on tui netstg2.img install (#157709) - Fix various splittree bugs (clumens, #157722, #157721, #157723) - Blacklist perl.i386 on x86_64 to be removed on upgrade (pnasrat, #156658) - Fix drive sorting (clumens) - Remove %installclass support for kickstart since it's never worked (#149690) - Fix name.arch in packages (pnasrat) - Remove bogus pre-existing RAID info on kickstart installs (clumens, #88359) - Pretend to have nano in the rescue environment - Don't load stage2.img into RAM for rescue mode if booted with 'linux text' (#155398) apr-0.9.6-3 ----------- * Tue May 17 2005 Joe Orton 0.9.6-3 - fix apr_procattr_child_*_set error handling * Tue Mar 01 2005 Joe Orton 0.9.6-2 - have apr-devel depend on specific version of gcc - add NOTICE to docdir * Wed Feb 09 2005 Joe Orton 0.9.6-1 - update to 0.9.6 cman-kernel-2.6.11.4-20050517.141233.FC4.0 ------------------------------------------ * Mon May 16 2005 Chris Feist - Add depmod in %post scriptlet. coreutils-5.2.1-45 ------------------ * Mon May 16 2005 Tim Waugh 5.2.1-45 - Applied Russell Coker's selinux changes (bug #157856). * Fri Apr 08 2005 Tim Waugh - Fixed pam patch from Steve Grubb (bug #154946). - Use better upstream patch for "stale utmp". cpio-2.6-7 ---------- * Tue May 17 2005 Peter Vrabec 2.6-7 - fix #156314 (CAN-2005-1229) cpio directory traversal issue - fix some gcc warnings dlm-kernel-2.6.11.4-20050517.141233.FC4.0 ----------------------------------------- * Mon May 16 2005 Chris Feist - Add depmod in %post scriplet. eclipse-1:3.1.0_fc-0.M6.18 -------------------------- * Tue May 17 2005 Ben Konrath 3.1.0_fc-0.M6.18 - Add Epoch to jsch requires. * Mon May 16 2005 Ben Konrath 3.1.0_fc-0.M6.17 - Update libswt-mozilla patches and require mozilla 1.7.8. * Fri May 13 2005 Andrew Overholt - Use %{ix86} macro in ExclusiveArch rather than i386 (jorton). elinks-0.10.3-3 --------------- * Tue May 17 2005 Karel Zak 0.10.3-3 - fix #157300 - Strange behavior on ppc64 (patch by Miloslav Trmac) emacs-21.4-4 ------------ * Mon Apr 25 2005 Jens Petersen - 21.4-4 - don't accidently exclude emacsclient from common package (Jonathan Kamens, #157808) - traditional Chinese desktop file translation (Wei-Lun Chao, #157287) fonts-indic-1.10-2 ------------------ * Tue May 17 2005 Leon Ho 1.10-2 - fix Requires: fontconfig (#151771) * Mon May 16 2005 Leon Ho 1.10-1 - New punjabi fonts to fix RH#151630 gcc-4.0.0-6 ----------- * Tue May 17 2005 Jakub Jelinek 4.0.0-6 - update from CVS - PRs target/18655, bootstrap/21230, fortran/17143, fortran/17432, java/21519, libfortran/21324, libgcj/20504, libgcj/21557, libgcj/21606, libstdc++/21526, middle-end/21237, target/21551, target/21556, tree-optimization/21532 - fixed WHERE inside FORALL (PR fortran/15080) - fix a tree sharing bug (#157792, PR tree-optimization/21610) - avoid using hash tables recursively (#157308) - fix ppc64 libgcj (Andrew Haley, #154684, #142611) gdm-1:2.6.0.8-14 ---------------- * Thu May 12 2005 Ray Strode 1:2.6.0.8-14 - Fix processing of new-line characters that got broken in 2.6.0.8-11 (bug 157442). gnbd-kernel-2.6.11.2-20050420.133124.FC4.23 ------------------------------------------- * Mon May 16 2005 Chris Feist - Add depmod in %post scriplet. gphoto2-2.1.5-9 --------------- * Sat May 14 2005 Tim Waugh 2.1.5-9 - Fixed buffer overrun in ricoh/g3 (bug #157739). gthumb-2.6.5-1 -------------- * Mon May 16 2005 John (J5) Palmieri - 2.6.5-1 - Minor updated that fixes a couple of bugs and adds some translations jessie-0:1.0.0-7 ---------------- * Mon May 16 2005 Thomas Fitzsimmons - 0:1.0.0-7 - Require jpackage-utils for post and postun. - Run rebuild-security-providers with full path. kbd-1.12-9 ---------- * Tue May 17 2005 Miloslav Trmac - 1.12-9 - Fix another violation of C aliasing rules (#157720, patch by Jan Kratochvil) kernel-2.6.11-1.1315_FC4 ------------------------ * Mon May 16 2005 Rik van Riel - enable Xen again (not tested yet) - fix a typo in the EXPORT_SYMBOL patch libsepol-1.5.9-2 ---------------- * Tue May 17 2005 Dan Walsh 1.5.9-2 - Upgrade to latest from NSA * Added sepol_genbools_policydb and sepol_genusers_policydb for audit2why. * Mon May 16 2005 Dan Walsh 1.5.8-2 - export sepol_context_to_sid netpbm-10.27-3 -------------- * Mon May 16 2005 Jindrich Novy 10.27-3 - fix ppmdither leak caused by bug in security patch (#157757) - drop gcc34 patch perl-3:5.8.6-13 --------------- * Sun May 15 2005 Warren Togami - 3:5.8.6-13 - Better patch for FindBin.pm (#127023#c37) * Sat May 14 2005 Jose Pedro Oliveira - 3:5.8.6-12 - New findbin-selinux patch: it now passes the FindBin.t tests (patch28 replaces patch23). #118877 #127023 - Remove 5.8.2 ABI compat (#154295 comments 6 and 7). * Thu Apr 28 2005 Ville Skytt?? - 3:5.8.6-10 - Apply fixes for CAN-2004-0452, CAN-2005-0155 and CAN-2005-0156 (#156128). perl-SGMLSpm-1.03ii-16 ---------------------- * Sat Apr 30 2005 Jose Pedro Oliveira - 1.03ii-15 - Specfile cleanup. (#156483) policycoreutils-1.23.9-1 ------------------------ * Mon May 16 2005 Dan Walsh 1.23.9-1 - Update to match NSA * Updated audit2why for sepol_ prefixes on Flask types to avoid namespace collision with libselinux, and to include now. * Fri May 13 2005 Dan Walsh 1.23.8-1 - Fix fixfiles to accept -f - Update to match NSA * Added audit2why utility. rhpl-0.164-1 ------------ * Mon May 16 2005 Chris Lumens 0.164-1 - Remove directory name from dvorak so it'll show up in anaconda's lists (#157763, #157840). * Mon May 16 2005 Chris Lumens 0.163-1 - Change resolution sorting to match X's behavior (Jef Spaleta, #157596). - Don't pad IP address information with unnecessary zeros (#157657). selinux-policy-strict-1.23.16-1 ------------------------------- * Mon May 16 2005 Dan Walsh 1.23.16-1 - Update from NSA * Added rdisc policy from Russell Coker. * Merged minor fix to named policy by Petre Rodan. * Merged minor fixes to policy from Russell Coker for kudzu, named, screen, setfiles, telnet, and xdm. * Merged minor fix to Makefile from Russell Coker. * Thu May 12 2005 Dan Walsh 1.23.15-5 - Remove incorrect user_*_t from te files - Add secadmfile attribute - Allow secadm to relabel secadmfile - Add wine file_context * Wed May 11 2005 Dan Walsh 1.23.15-4 - Allow smbd to communicate with cups - fix some net_conf contexts - Add a bunch of / files file_context selinux-policy-targeted-1.23.16-1 --------------------------------- * Mon May 16 2005 Dan Walsh 1.23.16-1 - Update from NSA * Added rdisc policy from Russell Coker. * Merged minor fix to named policy by Petre Rodan. * Merged minor fixes to policy from Russell Coker for kudzu, named, screen, setfiles, telnet, and xdm. * Merged minor fix to Makefile from Russell Coker. * Thu May 12 2005 Dan Walsh 1.23.15-5 - Remove incorrect user_*_t from te files - Add secadmfile attribute - Allow secadm to relabel secadmfile - Add wine file_context * Wed May 11 2005 Dan Walsh 1.23.15-4 - Allow smbd to communicate with cups - fix some net_conf contexts - Add a bunch of / files file_context system-config-printer-0.6.131-1 ------------------------------- * Sun May 15 2005 Tim Waugh 0.6.131-1 - 0.6.131: - Fixes for page margins (bug #134260). xchat-1:2.4.3-3 --------------- * Sun May 15 2005 Warren Togami 1:2.4.3-3 - Prevent interception of down arrow during Input Method (#144588 tagoh) From enrico.scholz at informatik.tu-chemnitz.de Wed May 18 18:15:48 2005 From: enrico.scholz at informatik.tu-chemnitz.de (Enrico Scholz) Date: Wed, 18 May 2005 20:15:48 +0200 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1115913808.3424.5.camel@nexus.verbum.private> (Colin Walters's message of "Thu, 12 May 2005 12:03:28 -0400") References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> Message-ID: <87is1gjt7v.fsf@kosh.bigo.ensc.de> walters at redhat.com (Colin Walters) writes: > There's actually been some work going on on giving each user their > own /tmp namespace via the kernel's CLONE_NEWNS capability and a PAM > module, AIUI. To the system administrator this could appear as > /tmp/. I think the problem is in getting later mounts to > actually appear in the cloned namespace. This CLONE_NEWNS and (related) 'mount --bind' operations are not very well supported by the kernel: * there does not exist a way to enter an already existing namespace; so, e.g. two different ssh sessions would have different /tmp directories * namespaces are causing problems with automounters * 'mount --bind' does not accept/honor options like 'noatime' or 'noexec' (which could be usefully e.g. to mount $HOME/tmp as /tmp). Patches are existing but responsible kernel maintainer refuses to apply them :( * CLONE_NEWNS + 'mount --bind' are not very well documented and it is often unclear whether strange behavior is expected or not. E.g. it may happen that '/' and '/..' are pointing to different inodes; dunno if this is wanted or not. Enrico -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 480 bytes Desc: not available URL: From kwade at redhat.com Wed May 18 22:23:26 2005 From: kwade at redhat.com (Karsten Wade) Date: Wed, 18 May 2005 15:23:26 -0700 Subject: call for release notes for FC4 Message-ID: <1116455007.6897.473.camel@erato.phig.org> Now is your final time to get your release note requests in for Fedora Core 4. Current version of the release notes is available here: http://fedora.redhat.com/docs/beta/release-notes-en/fc4-relnotes This version is in plain text. Here is how you can send in release notes: * Use this pre-filled bugzilla template (preferred method): http://tinyurl.com/exov3 -- OR -- * Send email with details to relnotes at fedoraproject.org . The latter will likely have a bug created to track discussion and resolution. Please file a bug instead in the first place. Release notes are now being handled by a team through the Fedora Documentation Project. Individuals or small groups handle a particular relnotes beat. Your note will be assigned to a writer on your beat. All release note requests are available through the tracking bugzilla number: https://bugzilla.redhat.com/beta/showdependencytree.cgi?id=151189 You can read more about the beats, which are subject to change as we develop this process. Note that many of the current writers are temporary assignments only. http://fedoraproject.org/wiki/DocsProject/ReleaseNotes/Beats We are looking for ongoing relnotes beat writers. If you or anyone else is interested, please visit: http://www.fedoraproject.org/wiki/DocsProject/NewWriters Thanks - Karsten -- Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/ gpg fingerprint: 2680 DBFD D968 3141 0115 5F1B D992 0E06 AD0E 0C41 Red Hat SELinux Guide http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/selinux-guide/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From enrico.scholz at informatik.tu-chemnitz.de Wed May 18 22:25:11 2005 From: enrico.scholz at informatik.tu-chemnitz.de (Enrico Scholz) Date: Thu, 19 May 2005 00:25:11 +0200 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <20050512155504.GA25624@jadzia.bu.edu> (Matthew Miller's message of "Thu, 12 May 2005 11:55:04 -0400") References: <20050512155504.GA25624@jadzia.bu.edu> Message-ID: <8764xgjho8.fsf@kosh.bigo.ensc.de> mattdm at mattdm.org (Matthew Miller) writes: > Just a thought: > > > /etc/profile.d/tmpdir.sh: FWIW... I use | #! /bin/bash | | T=/var/tmp/sessiondir-$USER | | dir= | for i in $T.*; do | test -d "$i" -a -O "$i" -a ! -L "$i" -a "$i" -nt /.autofsck || continue | dir="$i" | done | | test -n "$dir" || { | dir=$(mktemp -d "$T.$(date +%s).XXXXXX") && \ | mkdir -p $dir/{tmp,redhat/{SOURCES,RPMS/{i{3,4,5,6}86,noarch,x86_64},SPECS,SRPMS,BUILD}} \ | $dir/cvsextras && \ | ln -s ../SRPMS $dir/redhat/RPMS/ | } || { | echo "Failed to create tempdir" >&2 | exit 1 | } | | tmp="export ENSC_SESSIONDIR=$dir" | eval $tmp | echo "$tmp" to create a session directory. It assigns one temporary directory per machine cycle (detected by age of /.autofsck). So, e.g. two parallel ssh logins will have the same sessiondir. Adding the date to its name make 'ls -l ...' show the recent directory last. Enrico -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 480 bytes Desc: not available URL: From walters at redhat.com Wed May 18 22:39:20 2005 From: walters at redhat.com (Colin Walters) Date: Wed, 18 May 2005 18:39:20 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <87is1gjt7v.fsf@kosh.bigo.ensc.de> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> Message-ID: <1116455960.3552.42.camel@nexus.verbum.private> On Wed, 2005-05-18 at 20:15 +0200, Enrico Scholz wrote: > This CLONE_NEWNS and (related) 'mount --bind' operations are not very > well supported by the kernel: > > * there does not exist a way to enter an already existing namespace; so, > e.g. two different ssh sessions would have different /tmp directories Right, but that shouldn't be a problem since you can share data via your home directory or a specially-designated scratch area, etc. > * namespaces are causing problems with automounters Sounds like a regular bug; I don't think automounters would come into play for /tmp anyways? > * 'mount --bind' does not accept/honor options like 'noatime' or 'noexec' > (which could be usefully e.g. to mount $HOME/tmp as /tmp). Patches are > existing but responsible kernel maintainer refuses to apply them :( noexec's always been virtually useless. noatime is useful, but not so much that it would be a showstopper for CLONE_NEWNS, in my opinion. > * CLONE_NEWNS + 'mount --bind' are not very well documented and it is > often unclear whether strange behavior is expected or not. E.g. it may > happen that '/' and '/..' are pointing to different inodes; dunno if > this is wanted or not. Hm, so it might confuse tools? I'd imagine most tools out there recurse downwards into a path and so won't hit that issue, but it is something to watch out for. From notting at redhat.com Wed May 18 22:48:34 2005 From: notting at redhat.com (Bill Nottingham) Date: Wed, 18 May 2005 18:48:34 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1116455960.3552.42.camel@nexus.verbum.private> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> Message-ID: <20050518224834.GB25468@nostromo.devel.redhat.com> Colin Walters (walters at redhat.com) said: > On Wed, 2005-05-18 at 20:15 +0200, Enrico Scholz wrote: > > > This CLONE_NEWNS and (related) 'mount --bind' operations are not very > > well supported by the kernel: > > > > * there does not exist a way to enter an already existing namespace; so, > > e.g. two different ssh sessions would have different /tmp directories > > Right, but that shouldn't be a problem since you can share data via your > home directory or a specially-designated scratch area, etc. Well, there's agent sockets and the like in your tmp dir. Bill From nmiell at comcast.net Wed May 18 22:55:48 2005 From: nmiell at comcast.net (Nicholas Miell) Date: Wed, 18 May 2005 15:55:48 -0700 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <20050518224834.GB25468@nostromo.devel.redhat.com> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> Message-ID: <1116456948.4956.0.camel@localhost.localdomain> On Wed, 2005-05-18 at 18:48 -0400, Bill Nottingham wrote: > Colin Walters (walters at redhat.com) said: > > On Wed, 2005-05-18 at 20:15 +0200, Enrico Scholz wrote: > > > > > This CLONE_NEWNS and (related) 'mount --bind' operations are not very > > > well supported by the kernel: > > > > > > * there does not exist a way to enter an already existing namespace; so, > > > e.g. two different ssh sessions would have different /tmp directories > > > > Right, but that shouldn't be a problem since you can share data via your > > home directory or a specially-designated scratch area, etc. > > Well, there's agent sockets and the like in your tmp dir. Yes, but if all namespaces bind mount the same tmp dir, it doesn't matter that processes are running in different namespaces. -- Nicholas Miell From kwade at redhat.com Wed May 18 22:23:26 2005 From: kwade at redhat.com (Karsten Wade) Date: Wed, 18 May 2005 15:23:26 -0700 Subject: call for release notes for FC4 Message-ID: <1116455007.6897.473.camel@erato.phig.org> Now is your final time to get your release note requests in for Fedora Core 4. Current version of the release notes is available here: http://fedora.redhat.com/docs/beta/release-notes-en/fc4-relnotes This version is in plain text. Here is how you can send in release notes: * Use this pre-filled bugzilla template (preferred method): http://tinyurl.com/exov3 -- OR -- * Send email with details to relnotes at fedoraproject.org . The latter will likely have a bug created to track discussion and resolution. Please file a bug instead in the first place. Release notes are now being handled by a team through the Fedora Documentation Project. Individuals or small groups handle a particular relnotes beat. Your note will be assigned to a writer on your beat. All release note requests are available through the tracking bugzilla number: https://bugzilla.redhat.com/beta/showdependencytree.cgi?id=151189 You can read more about the beats, which are subject to change as we develop this process. Note that many of the current writers are temporary assignments only. http://fedoraproject.org/wiki/DocsProject/ReleaseNotes/Beats We are looking for ongoing relnotes beat writers. If you or anyone else is interested, please visit: http://www.fedoraproject.org/wiki/DocsProject/NewWriters Thanks - Karsten -- Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/ gpg fingerprint: 2680 DBFD D968 3141 0115 5F1B D992 0E06 AD0E 0C41 Red Hat SELinux Guide http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/selinux-guide/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From rgorosito at comarb.gov.ar Wed May 18 22:55:25 2005 From: rgorosito at comarb.gov.ar (Ricardo Ariel Gorosito) Date: Wed, 18 May 2005 19:55:25 -0300 Subject: PPC and RS/6000 (PReP) Message-ID: <428BC7DD.6010708@comarb.gov.ar> Hi, will RS/6000 be supported in FC4? rawhide's changelog from 20050514 say "Enable CONFIG_ISA on ppc32 to make the RS/6000 user happy" in kernel package and AFAIK anaconda has some support for PReP machines (changelog in anaconda 2005-Apr-26 talk about PReP in iSeries). I have downloades FC4T3 and when I try to boot 1st CD, the host freeze and in led display show F4D ('Bit-map file read into memory, start processing') CD1/images/README say "Images for installing on an iSeries machine can be found in the ppc/iSeries directory." but ppc/iSeries does not exist I'm testing with RS/6000 7043-240 (UP) processor is ppc604e (pSeries PReP, not CHRP). It is running linux today. R.- From ivazquez at ivazquez.net Wed May 18 23:12:34 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Wed, 18 May 2005 19:12:34 -0400 Subject: PPC and RS/6000 (PReP) In-Reply-To: <428BC7DD.6010708@comarb.gov.ar> References: <428BC7DD.6010708@comarb.gov.ar> Message-ID: <1116457954.7161.37.camel@ignacio.ignacio.lan> On Wed, 2005-05-18 at 19:55 -0300, Ricardo Ariel Gorosito wrote: > I have downloades FC4T3 and when I try to boot 1st CD, the host freeze > and in led display show F4D ('Bit-map file read into memory, start > processing') AFAIK CD 1 uses the Mac/32 boot image, so it will only work on 32-bit PowerMacs, G3 or better. -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From rodd at clarkson.id.au Thu May 19 00:32:47 2005 From: rodd at clarkson.id.au (Rodd Clarkson) Date: Thu, 19 May 2005 10:32:47 +1000 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <20050518224834.GB25468@nostromo.devel.redhat.com> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> Message-ID: <1116462767.3483.19.camel@jellyfish.redfishdemo.com> On Wed, 2005-05-18 at 18:48 -0400, Bill Nottingham wrote: > Colin Walters (walters at redhat.com) said: > > On Wed, 2005-05-18 at 20:15 +0200, Enrico Scholz wrote: > > > > > This CLONE_NEWNS and (related) 'mount --bind' operations are not very > > > well supported by the kernel: > > > > > > * there does not exist a way to enter an already existing namespace; so, > > > e.g. two different ssh sessions would have different /tmp directories > > > > Right, but that shouldn't be a problem since you can share data via your > > home directory or a specially-designated scratch area, etc. > > Well, there's agent sockets and the like in your tmp dir. Not sure if this is related but.... With regard to tmp directories, I'd like to see two things as default: 1. Each user should have there own ~/tmp space which only they can access. This could be used for the users agent sockets, but also just for general files that they would put in /tmp. This would give a better level of privacy (for whatever reason, but maybe simply so they don't have to think about the privacy implications of putting things in a publically accessible /tmp folder) 2. The system should have a general 'shared' folder that appears as a folder in each users home directory, but where any files placed there can be accessed by anyone else. This folder shouldn't delete files after a period like /tmp does, and if shouldn't cause problems with ownership (a security issue maybe). My father, who worked at the Bureau of Meteorology in Melbourne and who has use Unix as part of that always comments that one of the biggest issues he had was being able to simply share files with others without having to contact a sysadmin just to get a 'shared folder' set up. My wife concurs with him, thinking it's mad that she has to put important files in /tmp just to be able to share them, and while I could do something about this, the reality is that they would both like it to be done without having to ask anyone. (It's a small sample size, but I'm sure others have heard similar comments. 8-] ) Thoughts (and beratings ;-] ) R. -- "It's a fine line between denial and faith. It's much better on my side" From walters at redhat.com Thu May 19 02:12:02 2005 From: walters at redhat.com (Colin Walters) Date: Wed, 18 May 2005 22:12:02 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <20050518224834.GB25468@nostromo.devel.redhat.com> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> Message-ID: <1116468722.3642.34.camel@nexus.verbum.private> On Wed, 2005-05-18 at 18:48 -0400, Bill Nottingham wrote: > Colin Walters (walters at redhat.com) said: > > On Wed, 2005-05-18 at 20:15 +0200, Enrico Scholz wrote: > > > > > This CLONE_NEWNS and (related) 'mount --bind' operations are not very > > > well supported by the kernel: > > > > > > * there does not exist a way to enter an already existing namespace; so, > > > e.g. two different ssh sessions would have different /tmp directories > > > > Right, but that shouldn't be a problem since you can share data via your > > home directory or a specially-designated scratch area, etc. > > Well, there's agent sockets and the like in your tmp dir. Sure. But you weren't expecting to share an agent between separate ssh logins, were you? Defining a per-machine "session" gets terribly hackish, as the Gentoo keychain program shows. You get into this whole mess of trying to get some way of communicating data between independent logins, which gets a lot more difficult than "stuff it in ~/.foo" with fun things like NFS /home in the mix. /tmp is problematic since you can't use well-known filenames. You basically end up having to trawl /tmp looking for an active socket or something, or try ~/.foo/$hostname/bar, and $hostname has its own problems... From notting at redhat.com Thu May 19 02:20:40 2005 From: notting at redhat.com (Bill Nottingham) Date: Wed, 18 May 2005 22:20:40 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1116468722.3642.34.camel@nexus.verbum.private> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> <1116468722.3642.34.camel@nexus.verbum.private> Message-ID: <20050519022040.GB28670@nostromo.devel.redhat.com> Colin Walters (walters at redhat.com) said: > Sure. But you weren't expecting to share an agent between separate ssh > logins, were you? Consdering it's worked *fine* for me, including NFS, for about 5 years, yes, actually. :) Bill From enrico.scholz at informatik.tu-chemnitz.de Thu May 19 06:04:22 2005 From: enrico.scholz at informatik.tu-chemnitz.de (Enrico Scholz) Date: Thu, 19 May 2005 08:04:22 +0200 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1116455960.3552.42.camel@nexus.verbum.private> (Colin Walters's message of "Wed, 18 May 2005 18:39:20 -0400") References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> Message-ID: <871x83kazd.fsf@kosh.bigo.ensc.de> walters at redhat.com (Colin Walters) writes: >> This CLONE_NEWNS and (related) 'mount --bind' operations are not very >> well supported by the kernel: >> >> * there does not exist a way to enter an already existing namespace; so, >> e.g. two different ssh sessions would have different /tmp directories > > Right, but that shouldn't be a problem since you can share data via your > home directory or a specially-designated scratch area, etc. It will cause lot of problems when two regular sessions have a different view of the filesystem. E.g. when session A mounts /media/cdrom, this will not be available in session B. When B mounts it also, the unmounting will become funny as the kernel but not processes (detectable with 'fuser -m') locks this device. Or, when administrator mounts additional devices in the root-NS, this will not be reflected in the NS. Or the /etc/mtab designflaw of 'mount'... it is not NS aware, and although it causes other problems e.g. in read-only / it was impossible to eradicate it in all the years. >> * namespaces are causing problems with automounters > > Sounds like a regular bug; I don't think automounters would come into > play for /tmp anyways? Think of automounting /home/foo: In root-namespace (where the automounter was started), nobody accessed this dir and is is not mounted yet. Now, 'foo' opens a session A which creates a new namespace and tries to access his homedir. automounter sees that and mounts it; either in the root-NS (which keeps /home/foo still inaccessible for session A) or in the NS of session A. In fact, current automounters will do only the first option, are NS unaware and will be confused by them. All this is more than I could write in a bugreport; design flaws of current automounters are well known and documented e.g. in ftp://ftp-eng.cobalt.com/pub/whitepapers/autofs/towards_a_modern_autofs.txt >> * 'mount --bind' does not accept/honor options like 'noatime' or 'noexec' >> (which could be usefully e.g. to mount $HOME/tmp as /tmp). Patches are >> existing but responsible kernel maintainer refuses to apply them :( > > noexec's always been virtually useless. Why? The '/lib/ld-2... /tmp/foo' trick does not work anymore with recent kernels. >> * CLONE_NEWNS + 'mount --bind' are not very well documented and it is >> often unclear whether strange behavior is expected or not. E.g. it may >> happen that '/' and '/..' are pointing to different inodes; dunno if >> this is wanted or not. > > Hm, so it might confuse tools? Yes, it breaks e.g. chroot operations of yum. Enrico -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 480 bytes Desc: not available URL: From dwmw2 at infradead.org Thu May 19 07:09:37 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Thu, 19 May 2005 08:09:37 +0100 Subject: PPC and RS/6000 (PReP) In-Reply-To: <428BC7DD.6010708@comarb.gov.ar> References: <428BC7DD.6010708@comarb.gov.ar> Message-ID: <1116486578.3329.20.camel@localhost.localdomain> On Wed, 2005-05-18 at 19:55 -0300, Ricardo Ariel Gorosito wrote: > will RS/6000 be supported in FC4? Maybe. There isn't a lot which prevents the kernel from working -- CONFIG_ISA being a large part of it. I'm not sure how far people got with that kernel. > CD1/images/README say "Images for installing on an iSeries machine can > be found in the ppc/iSeries directory." but ppc/iSeries does not exist The image directories are somewhat misnamed. We have only two images -- the 64-bit kernel which runs on pSeries and Mac G5, and the 32-bit kernel which runs on 32-bit Macs (and might work on 32-bit PReP/CHRP too). We have 3 CDs -- the 64-bit kernel on a CHRP-bootable CD for pSeries, the 64-bit kernel on a Mac-bootable CD for G5s, and the 32-bit kernel on a Mac-bootable CD. These can theoretically be condensed into a single CD. For now, you'll have to netboot the RS/6000 from the images, if it's going to work. Since you're running Linux already, the best thing you can do is test the current rawhide kernel. If it doesn't work, it's probably due to configuration options which are missing; make it work and tell me which they were. If you join #fedora-ppc on irc.freenode.net, we can discuss it in more detail there. -- dwmw2 From aoliva at redhat.com Thu May 19 14:03:46 2005 From: aoliva at redhat.com (Alexandre Oliva) Date: 19 May 2005 11:03:46 -0300 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1116468722.3642.34.camel@nexus.verbum.private> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> <1116468722.3642.34.camel@nexus.verbum.private> Message-ID: On May 18, 2005, Colin Walters wrote: > Sure. But you weren't expecting to share an agent between separate ssh > logins, were you? That would be quite useful. I often leave GNU screen sessions behind, whose SSH agent sockets are removed when I disconnect, and then, when I reconnect, all I have to do to be able to ssh into other boxes again is to set SSH_AUTH_SOCK to the newly-created agent. With separate namespaces, there's just no way to get to that agent from the running screen session. -- Alexandre Oliva http://www.ic.unicamp.br/~oliva/ Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org} Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org} From buildsys at redhat.com Thu May 19 15:36:04 2005 From: buildsys at redhat.com (Build System) Date: Thu, 19 May 2005 11:36:04 -0400 Subject: rawhide report: 20050519 changes Message-ID: <200505191536.j4JFa4wR025454@porkchop.devel.redhat.com> Updated Packages: alsa-lib-1.0.9rc2-5 ------------------- * Wed May 18 2005 Martin Stransky 1.0.9rc2-5 - fix for #130593 - new ainit (dmix/dsnoop is default only for cards which really need it) - fix dsnoop - add fix for mixer (from https://bugs.gentoo.org/attachment.cgi?id=58918) audit-0.8.1-1 ------------- * Tue May 17 2005 Steve Grubb 0.8.1-1 - Fix code to "or" uid & gid checks for ausearch -ua & -ga - Change msg() to audit_msg() to avoid conflicts - Parse socket messages for hostname in ausearch * Thu May 12 2005 Steve Grubb 0.8-1 - ausearch fix bugs related to -f & -x - Parse messages using new types - Properly unescape filenames - Update interface for sending userspace messages to use more types beecrypt-4.1.2-8 ---------------- * Tue May 17 2005 Miloslav Trmac - 4.1.2-8 - Remove dependencies on private symbols not present in Python 2.4 from beecrypt-python * Tue May 17 2005 Miloslav Trmac - 4.1.2-7 - Doh, actually apply the patch * Tue May 17 2005 Miloslav Trmac - 4.1.2-6 - Fix b64encode() for data starting with NUL (#123650) bind-24:9.3.1-4 --------------- * Sat Apr 16 2005 Jason Vas Dias - 24:9.2.1-4 - Fix bug 157601: give named.init a configtest function - Fix bug 156797: named.init should check SELinux booleans.local before booleans - Fix bug 154335: if no controls in named.conf, stop named with -TERM sig, not rndc - Fix bug 155848: add NOTES section to named.8 man-page with info on all Red Hat BIND quirks and SELinux DDNS / slave zone file configuration - D-BUS patches NOT applied until dhcdbd is in FC * Fri Apr 15 2005 Jason Vas Dias - 24:9.3.1-4_dbus - Enhancement to allow dynamic forwarder table management and - DHCP forwarder auto-configuration with D-BUS cman-1.0-0.pre33.15 ------------------- * Tue May 17 2005 Chris Feist - Require cman-kernel-modules. * Thu May 05 2005 Chris Feist - Added patch to disable starting up the init scripts. * Mon Dec 20 2004 Chris Feist - Rebuild with new sources. cman-kernel-2.6.11.4-20050517.141233.FC4.3 ------------------------------------------ * Wed May 18 2005 Chris Feist - Fixed xen building with ARCH=xen & providing cman-kernel-modules. device-mapper-1.01.02-1.0 ------------------------- * Tue May 17 2005 Alasdair Kergon - 1.01.02-1.0 - Add --exec and --target to dmsetup for initrd convenience. - Automatically call dm_lib_exit when exiting. * Tue Mar 29 2005 Alasdair Kergon - 1.01.01-1.0 - Update dmsetup man page. - Replicate devmap_name output with dmsetup info -c --noheadings -o name. * Fri Mar 04 2005 Alasdair Kergon - 1.01.00-1.1 - Rebuild requested for gcc update. dlm-1.0-0.pre21.10 ------------------ * Tue May 17 2005 Chris Feist - Requires dlm-kernel-modules. * Sun May 08 2005 Florian La Roche - fix -devel requires * Fri May 06 2005 Chris Feist - Cleaned up .spec file. dlm-kernel-2.6.11.4-20050517.141233.FC4.2 ----------------------------------------- * Tue May 17 2005 Chris Feist - Provide dlm-kernel-modules & ARCH=xen on xen makes. eclipse-1:3.1.0_fc-0.M6.19 -------------------------- * Tue May 17 2005 Andrew Overholt 3.1.0_fc-0.M6.19 - Add Epoch on eclipse-platform. - Use %{_bindir} in post and postun scripts. emacs-21.4-5 ------------ * Wed May 18 2005 Jens Petersen - 21.4-5 - update cc-mode to 5.30.9 stable release to address font-lock problems (126165,148977,150197,155292,158044) * Mon May 16 2005 Jens Petersen - 21.4-4 - don't accidently exclude emacsclient from common package (Jonathan Kamens, #157808) - traditional Chinese desktop file translation (Wei-Lun Chao, #157287) * Wed Apr 20 2005 Jens Petersen - 21.4-3 - add igrep.el and init file evolution-connector-2.2.2-5 --------------------------- * Wed May 18 2005 David Malcolm - 2.2.2-5 - add Aaron Gaudio's patch to fix PDA syncronization (#139393) * Tue May 17 2005 David Malcolm - 2.2.2-4 - Install the debug utilities from the "lib" subdirectory; renumber patches accordingly; regenerate the generated patch evolution-data-server-1.2.2-3 ----------------------------- * Wed May 18 2005 David Malcolm - 1.2.2-3 - bumped libsoup requirement to 2.2.3; removed mozilla_build_version, using pkg-config instead for locating NSPRS and NSS headers/libraries (#158085) gail-1.8.3-2 ------------ * Wed May 18 2005 Matthias Clasen 1.8.3-2 - Fix the tamil translations gnbd-1.0-0.pre14.6 ------------------ * Tue May 17 2005 Chris Feist - Require cman-kernel-modules. * Fri May 06 2005 Chris Feist - Cleanup .spec file, don't glob /usr/share/man. * Mon Dec 20 2004 Chris Feist - Rebuild with new sources. gnbd-kernel-2.6.11.2-20050420.133124.FC4.25 ------------------------------------------- * Tue May 17 2005 Chris Feist - Provides gnbd-kernel-modules & added ARCH=xen to build. gnome-libs-1:1.4.1.2.90-46 -------------------------- * Thu May 12 2005 Ray Strode - 1:1.4.1.2.90-46 - add UTF-8 encoded ta.po (bug #135349) - make compile gnome-media-2.10.2-2 -------------------- * Tue May 17 2005 John (J5) Palmieri 2.10.2-2 - remove gnome-sound-recorder because it hasn't worked for some time gnome-utils-1:2.10.0-3 ---------------------- * Tue May 17 2005 Ray Strode 1:2.10.0-3 - call more autofoo to allow recompilation on x86_64 (bug 150627) gnu-crypto-0:2.0.1-1jpp_4fc --------------------------- * Mon May 16 2005 Thomas Fitzsimmons - 0:2.0.1-1jpp_4fc - Require jpackage-utils for post and postun. - Run rebuild-security-providers with full path. httpd-2.0.54-9 -------------- * Wed May 18 2005 Joe Orton 2.0.54-9 - add piped logger fixes (w/Jeff Trawick) hwbrowser-0.21-1 ---------------- * Wed May 18 2005 Nils Philippsen 0.21 - use True, False instead of gtk.TRUE, gtk.FALSE - set window manager icon (#155866) java-1.4.2-gcj-compat-0:1.4.2.0-11jpp ------------------------------------- * Thu Oct 07 2004 Gary Benson 0:1.4.2.0-11jpp - Don't override $CLASSPATH when we add ecj.jar in javac. * Wed Oct 06 2004 Thomas Fitzsimmons 0:1.4.2.0-10jpp - Bump gcc version. (Fix #133898) * Fri Oct 01 2004 Thomas Fitzsimmons 0:1.4.2.0-9jpp - Change /usr/lib/jvm directory to java-1.4.2-gcj from java-1.4.2-gcj-compat. java-1.4.2-gcj-compat-0:1.4.2.0-40jpp_22rh ------------------------------------------ * Wed May 18 2005 Thomas Fitzsimmons - 0:1.4.2.0-40jpp_22rh - Move gcc-java requirement from base to -devel. * Wed May 18 2005 Thomas Fitzsimmons - 0:1.4.2.0-40jpp_21rh - Comment out bouncy castle stuff. * Tue May 17 2005 Thomas Fitzsimmons - 0:1.4.2.0-40jpp_20rh - Require jpackage-utils for post and postun. - Run rebuild-security-providers with full path. kernel-2.6.11-1.1319_FC4 ------------------------ * Wed May 18 2005 Dave Jones - Fix up some warnings in the IDE patches. - 2.6.12-rc4-git2 * Tue May 17 2005 Dave Jones - 2.6.12-rc4-git1 ARM, ioctl security fixes, mmc driver update, ibm_emac & tulip netdriver fixes, serial updates ELF loader security fix. kudzu-1.1.116.2-1 ----------------- * Wed May 18 2005 Bill Nottingham 1.1.116.2-1 - tweak synaptics detection for ALPS (#158062, ) mkinitrd-4.2.15-1 ----------------- * Tue May 17 2005 Peter Jones - 4.2.15-1 - Better init argument handling (don't mess up with serial console) perl-3:5.8.6-15 --------------- * Wed May 18 2005 Warren Togami - 3:5.8.6-15 - remove unused /tmp/MANIFEST.all (#151801) * Tue May 17 2005 Warren Togami - 3:5.8.6-14 - CGI.pm 3.10 fixes mod_perl problems (#158036) * Sun May 15 2005 Warren Togami - 3:5.8.6-13 - Better patch for FindBin.pm (#127023#c37) qt-1:3.3.4-14 ------------- * Wed May 18 2005 Than Ngo 1:3.3.4-14 - apply patch to use ecvt, fcvt (thanks to Jakub) - fix a bug in printing of postscript #156977 * Wed May 18 2005 Than Ngo 1:3.3.4-13 - rebuild rsync-2.6.4-3 ------------- * Tue May 17 2005 Jay Fenlason 2.6.4-3 - Include the -address patch from upstream, to close bz#154752 Unable to use --address in client mode spamassassin-3.0.3-4.fc4 ------------------------ * Tue May 17 2005 Warren Togami - 3.0.3-4 - allow user-level disabling of subject rewriting pref (#147464) system-config-securitylevel-1.5.8-1 ----------------------------------- * Wed May 18 2005 Chris Lumens 1.5.8-1 - Revert .desktop file changes since they weren't properly marked for translation. system-config-soundcard-1.2.11-3 -------------------------------- * Wed May 18 2005 Martin Stransky 1.2.11-3 - fixed #158064 totem-1.0.2-1 ------------- * Tue May 17 2005 John (J5) Palmieri - 1.0.2-1 - Update to upstream version 1.0.2 to fix minor bugs - Register the thumbnail and handlers schemas From tarjei.knapstad at predichem.com Thu May 19 16:01:54 2005 From: tarjei.knapstad at predichem.com (Tarjei Knapstad) Date: Thu, 19 May 2005 18:01:54 +0200 Subject: workaround gcc-4 assembly constraint bug In-Reply-To: <1116253847.23731.19.camel@Madison.badger.com> References: <1116253847.23731.19.camel@Madison.badger.com> Message-ID: <1116518514.24785.4.camel@tarjei.predichem.nett> On Mon, 2005-05-16 at 16:30, Toshio wrote: > > I'm not too familiar with gcc's constraints, though. Would anyone with > a bit more experience be willing to tell me if that looks right or not? > Toshio, I think you'd get a lot more feedback on this from the gcc list if you didn't post there already. They're the guys with intimate knowledge of gcc code generation (even though I'm sure there's a couple of gcc developers at redhat too :)) Cheers, -- Tarjei From jpo at di.uminho.pt Thu May 19 17:43:38 2005 From: jpo at di.uminho.pt (=?ISO-8859-1?Q?Jos=E9_Pedro_Oliveira?=) Date: Thu, 19 May 2005 18:43:38 +0100 Subject: fedora-cvs-commits mailling list problem? Message-ID: <428CD04A.3080306@di.uminho.pt> The latest Fedora CVS commits aren't being forward to the fedora-cvs-commits mailling list (or not being archived). Archive: https://www.redhat.com/archives/fedora-cvs-commits/2005-May/thread.html jpo -- Jos? Pedro Oliveira * mailto: jpo at di.uminho.pt * http://gsd.di.uminho.pt/~jpo * * gpg fingerprint = F9B6 8D87 859D 1C94 48F0 84C0 9749 9EB5 91BD 851B * -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From ivazquez at ivazquez.net Thu May 19 17:56:53 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Thu, 19 May 2005 13:56:53 -0400 Subject: fedora-cvs-commits mailling list problem? In-Reply-To: <428CD04A.3080306@di.uminho.pt> References: <428CD04A.3080306@di.uminho.pt> Message-ID: <1116525413.7161.66.camel@ignacio.ignacio.lan> On Thu, 2005-05-19 at 18:43 +0100, Jos? Pedro Oliveira wrote: > The latest Fedora CVS commits aren't being forward to the > fedora-cvs-commits mailling list (or not being archived). > > Archive: > https://www.redhat.com/archives/fedora-cvs-commits/2005-May/thread.html How many commits have happened since bzip2-1.0.2-bomb.patch? -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From jpo at di.uminho.pt Thu May 19 18:03:10 2005 From: jpo at di.uminho.pt (=?UTF-8?B?Sm9zw6kgUGVkcm8gT2xpdmVpcmE=?=) Date: Thu, 19 May 2005 19:03:10 +0100 Subject: fedora-cvs-commits mailling list problem? In-Reply-To: <1116525413.7161.66.camel@ignacio.ignacio.lan> References: <428CD04A.3080306@di.uminho.pt> <1116525413.7161.66.camel@ignacio.ignacio.lan> Message-ID: <428CD4DE.9030106@di.uminho.pt> Ignacio Vazquez-Abrams wrote: > On Thu, 2005-05-19 at 18:43 +0100, Jos? Pedro Oliveira wrote: > >>The latest Fedora CVS commits aren't being forward to the >>fedora-cvs-commits mailling list (or not being archived). >> >>Archive: >>https://www.redhat.com/archives/fedora-cvs-commits/2005-May/thread.html > > > How many commits have happened since bzip2-1.0.2-bomb.patch? Almost everything that appeared today in rawhide (no devel branch commits? one commit in the FC4 branch?) Example: I don't see the latest perl package commits. jpo -- Jos? Pedro Oliveira * mailto: jpo at di.uminho.pt * http://gsd.di.uminho.pt/~jpo * * gpg fingerprint = F9B6 8D87 859D 1C94 48F0 84C0 9749 9EB5 91BD 851B * -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From walters at redhat.com Thu May 19 18:28:50 2005 From: walters at redhat.com (Colin Walters) Date: Thu, 19 May 2005 14:28:50 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <871x83kazd.fsf@kosh.bigo.ensc.de> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <871x83kazd.fsf@kosh.bigo.ensc.de> Message-ID: <1116527330.3585.27.camel@nexus.verbum.private> On Thu, 2005-05-19 at 08:04 +0200, Enrico Scholz wrote: > It will cause lot of problems when two regular sessions have a different > view of the filesystem. E.g. when session A mounts /media/cdrom, this > will not be available in session B. Work is going on on making "shared subtrees" as I think they're called working upstream. > Or the /etc/mtab designflaw of 'mount'... it is not NS aware, and although > it causes other problems e.g. in read-only / it was impossible to eradicate > it in all the years. Yeah...that's a can of worms. > Think of automounting /home/foo: In root-namespace (where the automounter > was started), nobody accessed this dir and is is not mounted yet. I think this is essentially the same issue as the first one; if we have shared subtrees then this should work. > > noexec's always been virtually useless. > > Why? The '/lib/ld-2... /tmp/foo' trick does not work anymore with recent > kernels. That's cool that that hole is closed, although I wonder if there are other ones. Besides that, there's a few other facets of "useless": first, that lots of things break on noexec /tmp, and second, you need to ensure that there aren't any other writable areas that aren't mounted noexec, and third, SELinux gives you a much more fine-grained control over execute permissions anyways. > >> * CLONE_NEWNS + 'mount --bind' are not very well documented and it is > >> often unclear whether strange behavior is expected or not. E.g. it may > >> happen that '/' and '/..' are pointing to different inodes; dunno if > >> this is wanted or not. > > > > Hm, so it might confuse tools? > > Yes, it breaks e.g. chroot operations of yum. Even just for a new /tmp namespace? Seems surprising. From Matt_Domsch at dell.com Fri May 20 04:09:24 2005 From: Matt_Domsch at dell.com (Matt Domsch) Date: Thu, 19 May 2005 23:09:24 -0500 Subject: [PATCH 1.4.11 0/2] OpenIPMI initscript and config file, RPM spec patch Message-ID: <20050520040924.GA25048@lists.us.dell.com> The following messages will contain a patch implementing an initscript and related config file for the OpenIPMI library, and a patch against Fedora Core 4 devel OpenIPMI-1.4.11-5 spec file to enable them. The initscript works for me on RHEL3 and RHEL4 releases. I'd appreciate feedback and/or inclusion directly into both FC4-devel and the OpenIPMI project. Thanks, Matt -- Matt Domsch Software Architect Dell Linux Solutions linux.dell.com & www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com From Matt_Domsch at dell.com Fri May 20 04:10:31 2005 From: Matt_Domsch at dell.com (Matt Domsch) Date: Thu, 19 May 2005 23:10:31 -0500 Subject: [PATCH 1.4.11 1/2] OpenIPMI initscript and config file In-Reply-To: <20050520040924.GA25048@lists.us.dell.com> References: <20050520040924.GA25048@lists.us.dell.com> Message-ID: <20050520041031.GB25048@lists.us.dell.com> config file and initscript -- Matt Domsch Software Architect Dell Linux Solutions linux.dell.com & www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com --- /dev/null 2005-05-08 18:50:01.506062864 -0500 +++ OpenIPMI/ipmi.sysconf 2005-05-19 16:43:13.000000000 -0500 @@ -0,0 +1,30 @@ +# Enable standard hardware interfaces (KCS, BT, SMIC) +# You probably want this enabled. +IPMI_SI=1 + +# Enable nonstandard interfaces (SMB via i2c) +# IPMI_SMB=1 + +# Enable /dev/ipmi0 interface, used by ipmitool, ipmicmd, +# and other userspace IPMI-using applications. +# You probably want this enabled. +DEV_IPMI=1 + +# Enable IPMI_WATCHDOG if you want the IPMI watchdog +# to reboot the system if it hangs +# IPMI_WATCHDOG=1 + +# Enable IPMI_POWEROFF if you want the IPMI +# poweroff module to be loaded. +# IPMI_POWEROFF=1 + +# Enable IPMI_POWERCYCLE if you want the system to be power-cycled (power +# down, delay briefly, power on) rather than power off, on systems +# that support such. IPMI_POWEROFF=1 is also required. +# IPMI_POWERCYCLE=1 + +# Enable "legacy" interfaces for applications +# Intel IMB driver interface +# IPMI_IMB=1 +# Radisys driver interface +# IPMI_RADISYS=1 --- /dev/null 2005-05-08 18:50:01.506062864 -0500 +++ OpenIPMI/ipmi.init 2005-05-19 16:42:58.000000000 -0500 @@ -0,0 +1,196 @@ +#!/bin/sh +############################################################################# +# +# ipmi: OpenIPMI Driver init script +# +# Authors: Matt Domsch +# Chris Poblete +# +# chkconfig: - 04 96 +# description: OpenIPMI Driver init script +# +### BEGIN INIT INFO +# Provides: ipmidrv +# Required-Start: $localfs $remotefs $syslog +# Required-Stop: $localfs $remotefs $syslog +# Default-Start: +# Default-Stop: +# Short-Description: OpenIPMI Driver init script +# Description: OpenIPMI Driver init script +### END INIT INFO +# +############################################################################# +# for log_success_msg and friends +[ -r /lib/lsb/init-functions ] && . /lib/lsb/init-functions +# source config info +[ -r /etc/sysconfig/ipmi ] && . /etc/sysconfig/ipmi + +############################################################################# +# GLOBALS +############################################################################# +MODULE_NAME="ipmi" +INTF_NUM=0 + +IPMI_SMB_MODULE_NAME="ipmi_smb" +IPMI_SI_MODULE_NAME="ipmi_si" +kernel=`uname -r | cut -d. -f1-2` +if [ "${kernel}" == "2.4" ]; then + IPMI_SMB_MODULE_NAME="ipmi_smb_intf" + IPMI_SI_MODULE_NAME="ipmi_si_drv" +fi + +MODULES="ipmi_radisys ipmi_imb ipmi_poweroff ipmi_watchdog \ + ipmi_devintf ${IPMI_SMB_MODULE_NAME} ${IPMI_SI_MODULE_NAME} \ + ipmi_msghandler" + + +RETVAL=0 +LOCKFILE=/var/lock/subsys/ipmi + +############################################################################# +start_watchdog() +{ + if [ "${IPMI_WATCHDOG}" = "1" ]; then + modprobe ipmi_watchdog || RETVAL=2 + fi +} + +stop_watchdog() +{ + modprobe -r ipmi_watchdog +} + +start_powercontrol() +{ + local poweroff_opts="" + if [ "${IPMI_POWEROFF}" = "1" ]; then + [ "${IPMI_POWERCYCLE}" == "1" ] && poweroff_opts="chassis_ctrl_cmd_param=2" + modprobe ipmi_poweroff "${poweroff_opts}" || RETVAL=2 + fi +} + +stop_powercontrol() +{ + modprobe -r ipmi_poweroff +} + +############################################################################# +load_ipmi_modules () +{ + modprobe ipmi_msghandler || RETVAL=1 + if [ "${IPMI_SI}" = "1" ]; then + modprobe ${IPMI_SI_MODULE_NAME} || RETVAL=1 + fi + if [ "${IPMI_SMB}" = "1" ]; then + modprobe ${IPMI_SMB_MODULE_NAME} || RETVAL=1 + fi + [ "${RETVAL}" = "1" ] && return + + if [ "${DEV_IPMI}" = "1" ]; then + modprobe ipmi_devintf || RETVAL=2 + if [ "${RETVAL}" != "2" ]; then + # Note, this really should be done by udev on 2.6 + DEVMAJOR=`cat /proc/devices | awk '/ipmidev/{print $1}'` + mknod -m 0600 /dev/ipmi${INTF_NUM} c ${DEVMAJOR} 0 || RETVAL=2 + ln -sf /dev/ipmi${INTF_NUM} /dev/ipmi || RETVAL=2 + fi + fi + + start_watchdog + start_powercontrol + if [ "${IPMI_IMB}" = "1" ]; then + modprobe ipmi_imb || RETVAL=2 + # FIXME create canonical /dev/foo entry + fi + if [ "${IPMI_RADISYS}" = "1" ]; then + modprobe ipmi_radisys || RETVAL=2 + # FIXME create canonical /dev/foo entry + fi + return +} + +############################################################################# +unload_ipmi_modules() +{ + # Note, deleting these /dev files really should be done by udev + # so this function will change soon as the driver changes + # to allow such to happen automatically. + rm -f "/dev/ipmi${INTF_NUM}" + rm -f "/dev/ipmi" + for m in ${MODULES}; do + modprobe -q -r ${m} + done +} + +############################################################################# +start() +{ + echo -n $"Starting ${MODULE_NAME} drivers: " + load_ipmi_modules + [ "${RETVAL}" = "1" ] && log_failure_msg && return + [ "${RETVAL}" = "2" ] && touch ${LOCKFILE} && log_warning_msg + [ "${RETVAL}" = "0" ] && touch ${LOCKFILE} && log_success_msg +} + +############################################################################# +stop() +{ + echo -n $"Stopping ${MODULE_NAME} drivers: " + unload_ipmi_modules + rm -f ${LOCKFILE} + log_success_msg +} + +############################################################################# +restart() +{ + stop + start +} + +############################################################################# +status () +{ + for m in ${MODULES}; do + if /sbin/lsmod | grep $m >/dev/null 2>&1 ; then + echo "$m module loaded" + else + echo "$m module not loaded" + fi + done +} + +usage () +{ + echo $"Usage: $0 {start|stop|status|restart|condrestart|" + echo $" start-watchdog|stop-watchdog|start-powercontrol|stop-powercontrol}" 1>&2 + RETVAL=1 +} + +condrestart () +{ + [ -e ${LOCKFILE} ] && restart +} + +############################################################################# +# MAIN +############################################################################# +case "$1" in + start) start ;; + stop) stop ;; + restart) restart ;; + status) status ;; + condrestart) condrestart ;; + start-watchdog) start_watchdog ;; + stop-watchdog) stop_watchdog ;; + start-powercontrol) start_powercontrol ;; + stop-powercontrol) stop_powercontrol ;; + *) usage ;; +esac + +exit ${RETVAL} + +############################################################################# +# end of file +############################################################################# + From Matt_Domsch at dell.com Fri May 20 04:11:21 2005 From: Matt_Domsch at dell.com (Matt Domsch) Date: Thu, 19 May 2005 23:11:21 -0500 Subject: [PATCH 1.4.11 2/2] OpenIPMI RPM spec patch In-Reply-To: <20050520040924.GA25048@lists.us.dell.com> References: <20050520040924.GA25048@lists.us.dell.com> Message-ID: <20050520041121.GC25048@lists.us.dell.com> spec file -- Matt Domsch Software Architect Dell Linux Solutions linux.dell.com & www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com --- OpenIPMI.spec.orig 2005-05-19 16:47:03.000000000 -0500 +++ OpenIPMI.spec 2005-05-19 22:53:59.000000000 -0500 @@ -1,13 +1,15 @@ Summary: OpenIPMI (Intelligent Platform Management Interface) library and tools Name: OpenIPMI Version: 1.4.11 -Release: 5 +Release: 6 License: GPL Group: System Environment/Base Source: http://prdownloads.sourceforge.net/openipmi/%{name}-%{version}.tar.gz Patch1: OpenIPMI-1.4.11-gcc4.patch +Patch2: initscripts.patch BuildRoot: /var/tmp/%{name}-%{version}-root BuildPrereq: gdbm-devel swig +Requires: lsb %description The Open IPMI project aims to develop an open code base to allow access to @@ -26,6 +28,7 @@ %prep %setup -q %patch1 -p1 -b .gcc4 +%patch2 -p1 %build %configure @@ -35,9 +38,34 @@ %makeinstall rm -rf $RPM_BUILD_ROOT/%{_libdir}/*.la +# install SYSV init stuff +mkdir -p $RPM_BUILD_ROOT/etc/init.d +install -m755 $RPM_BUILD_DIR/%{name}-%{version}/ipmi.init \ + $RPM_BUILD_ROOT/etc/init.d/ipmi +mkdir -p $RPM_BUILD_ROOT/etc/sysconfig +install -m644 $RPM_BUILD_DIR/%{name}-%{version}/ipmi.sysconf \ + $RPM_BUILD_ROOT/etc/sysconfig/ipmi + + + %clean rm -rf $RPM_BUILD_ROOT +%post +if [ -x /usr/lib/lsb/install_initd ]; then + /usr/lib/lsb/install_initd ipmi +elif [ -x /sbin/chkconfig ]; then + /sbin/chkconfig --add ipmi +fi + +%preun +if [ -x /usr/lib/lsb/remove_initd ]; then + /usr/lib/lsb/remove_initd ipmi +elif [ -x /sbin/chkconfig ]; then + /sbin/chkconfig --del ipmi +fi + + %files %defattr(-,root,root) %{_bindir}/* @@ -47,6 +75,8 @@ %{_libdir}/perl5/site_perl/*/*-linux-thread-multi/OpenIPMI.pm %dir %{_libdir}/perl5/site_perl/*/*-linux-thread-multi/auto/OpenIPMI %{_libdir}/perl5/site_perl/*/*-linux-thread-multi/auto/OpenIPMI/* +%config(noreplace) %{_sysconfdir}/init.d/ipmi +%config(noreplace) %{_sysconfdir}/sysconfig/ipmi %files devel %defattr(-,root,root) @@ -55,6 +85,9 @@ %{_libdir}/*.a %changelog +* Thu May 19 2005 Matt Domsch 1.4.11-6 +- add initscript and config file + * Wed Mar 30 2005 Phil Knirsch 1.4.11-5 - Correctly put libs in the proper packages From notting at redhat.com Fri May 20 04:19:24 2005 From: notting at redhat.com (Bill Nottingham) Date: Fri, 20 May 2005 00:19:24 -0400 Subject: [PATCH 1.4.11 1/2] OpenIPMI initscript and config file In-Reply-To: <20050520041031.GB25048@lists.us.dell.com> References: <20050520040924.GA25048@lists.us.dell.com> <20050520041031.GB25048@lists.us.dell.com> Message-ID: <20050520041924.GA24328@nostromo.devel.redhat.com> Matt Domsch (Matt_Domsch at dell.com) said: > + if [ "${DEV_IPMI}" = "1" ]; then > + modprobe ipmi_devintf || RETVAL=2 > + if [ "${RETVAL}" != "2" ]; then > + # Note, this really should be done by udev on 2.6 > + DEVMAJOR=`cat /proc/devices | awk '/ipmidev/{print $1}'` > + mknod -m 0600 /dev/ipmi${INTF_NUM} c ${DEVMAJOR} 0 || RETVAL=2 > + ln -sf /dev/ipmi${INTF_NUM} /dev/ipmi || RETVAL=2 > + fi > + fi Why isn't the driver fixed to populate sysfs properly? > + > + start_watchdog > + start_powercontrol > + if [ "${IPMI_IMB}" = "1" ]; then > + modprobe ipmi_imb || RETVAL=2 > + # FIXME create canonical /dev/foo entry > + fi > + if [ "${IPMI_RADISYS}" = "1" ]; then > + modprobe ipmi_radisys || RETVAL=2 > + # FIXME create canonical /dev/foo entry > + fi > + return How many, if any, of these modules can be detected for the proper hardware? Bill From Matt_Domsch at dell.com Fri May 20 09:54:57 2005 From: Matt_Domsch at dell.com (Matt Domsch) Date: Fri, 20 May 2005 04:54:57 -0500 Subject: [PATCH 1.4.11 1/2] OpenIPMI initscript and config file In-Reply-To: <20050520041924.GA24328@nostromo.devel.redhat.com> References: <20050520040924.GA25048@lists.us.dell.com> <20050520041031.GB25048@lists.us.dell.com> <20050520041924.GA24328@nostromo.devel.redhat.com> Message-ID: <20050520095457.GA32316@lists.us.dell.com> On Fri, May 20, 2005 at 12:19:24AM -0400, Bill Nottingham wrote: > Matt Domsch (Matt_Domsch at dell.com) said: > > + if [ "${DEV_IPMI}" = "1" ]; then > > + modprobe ipmi_devintf || RETVAL=2 > > + if [ "${RETVAL}" != "2" ]; then > > + # Note, this really should be done by udev on 2.6 > > + DEVMAJOR=`cat /proc/devices | awk '/ipmidev/{print $1}'` > > + mknod -m 0600 /dev/ipmi${INTF_NUM} c ${DEVMAJOR} 0 || RETVAL=2 > > + ln -sf /dev/ipmi${INTF_NUM} /dev/ipmi || RETVAL=2 > > + fi > > + fi > > Why isn't the driver fixed to populate sysfs properly? Just-in-time programming. Corey submitted class_simple support this evening. http://marc.theaimsgroup.com/?l=linux-kernel&m=111654581209165& That'll handle the kernels where it exists, still need to handle the cases where it doesn't. I'll add a test for existance of the dev file, and fall back to creating the devnode here if it doesn't exist. > > + > > + start_watchdog > > + start_powercontrol > > + if [ "${IPMI_IMB}" = "1" ]; then > > + modprobe ipmi_imb || RETVAL=2 > > + # FIXME create canonical /dev/foo entry > > + fi > > + if [ "${IPMI_RADISYS}" = "1" ]; then > > + modprobe ipmi_radisys || RETVAL=2 > > + # FIXME create canonical /dev/foo entry > > + fi > > + return > > How many, if any, of these modules can be detected for the > proper hardware? You're thinking of using kudzu instead? Possible, though the data is scattered throughout ACPI and SMBIOS tables, when it is known. That's more than I was up for this week. Future incremental change? Thanks, Matt -- Matt Domsch Software Architect Dell Linux Solutions linux.dell.com & www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com From buildsys at redhat.com Fri May 20 11:51:35 2005 From: buildsys at redhat.com (Build System) Date: Fri, 20 May 2005 07:51:35 -0400 Subject: rawhide report: 20050520 changes Message-ID: <200505201151.j4KBpZVK028658@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.6-20050517.141233.FC4.8 ----------------------------------------- * Tue May 17 2005 Chris Feist - Provides GFS-kernel-modules and add ARCH=xen to build. anaconda-10.2.0.64-1 -------------------- * Thu May 19 2005 Jeremy Katz - 10.2.0.64-1 - Handle longer arch strings (notting) - Fix traceback in network screen (#158134) - Include synaptics for X config (pnasrat) - Magic boot for mac vs mac64 on disc1/dvd (pnasrat) - Bump point at which we use graphical stage2 for http/ftp (#157274) - Use uuid in mdadm.conf, stop using copy of md.h (#136051) - Support deletion of bootloader entries in text mode (#125358) - Support RAID /boot on pSeries along with handling of multiple PReP partitions (Dustin Kirkland) bzip2-1.0.2-16 -------------- * Thu May 19 2005 Jiri Ryska - fixed permission setting for decompressed files #155742 - fixed decompression bomb (DoS) #157548 cman-kernel-2.6.11.4-20050517.141233.FC4.4 ------------------------------------------ devhelp-0.10-1 -------------- * Thu May 19 2005 Ray Strode 0.10.0-1 - Update to 0.10.0 (bug #157753) dlm-kernel-2.6.11.4-20050517.141233.FC4.4 ----------------------------------------- dmraid-1.0.0.rc8-FC4_1 ---------------------- * Thu May 19 2005 Heinz Mauelshagen 1.0.0.rc8-FC4_1 - nv.c: fixed stripe size - sil.c: avoid incarnation_no in name creation, because the Windows driver changes it every time - added --ignorelocking option to avoid taking out locks in early boot where no read/write access to /var is given eclipse-bugzilla-1:0.1.0_fc-16 ------------------------------ * Wed May 18 2005 Jeff Pound 0.1.0_fc-16 - Fix bug in query option parser (see comment 1 from bz#151443). - Parse icons from bugzilla sites (bz#151441). gcc-4.0.0-7 ----------- * Wed May 18 2005 Jakub Jelinek 4.0.0-7 - update from CVS - PRs fortran/20954, libgcj/18220, libgcj/19729, libgcj/21140 - fix IA-64 ICE on sdata/sbss variable references with offsets (Richard Henderson, #158001, PR target/21632) - add gjnih and gjnih.1 to gcc-java subpackage - fix german, spanish, turkish and chinese translations (#157879, PR translation/21364) - configure with --enable-libgcj-multifile to avoid make -jN related multilib failures gdm-1:2.6.0.8-15 ---------------- * Thu May 19 2005 Ray Strode 1:2.6.0.8-15 - Take out some syslog spew (bug 157711). gnbd-kernel-2.6.11.2-20050420.133124.FC4.26 ------------------------------------------- gnome-applets-1:2.10.1-8 ------------------------ * Thu May 19 2005 Ray Strode 1:2.10.1-8 - Install modemlights gconf schema (bug 157764). gnome-games-1:2.10.0-4 ---------------------- * Thu May 19 2005 Ray Strode 1:2.10.0-4 - Change some sounds around grub-0.95-13 ------------ * Thu May 19 2005 Peter Jones 0.95-13 - Make the spec work with gcc3 and gcc4, so people can test on existing installations. - don't treat i2o like a cciss device, since its partition names aren't done that way. (#158158) kernel-2.6.11-1.1323_FC4 ------------------------ * Thu May 19 2005 Dave Jones - Fix up missing symbols in ipw2200 driver. - Reenable debugfs / usbmon. SELinux seems to cope ok now. (Needs selinux-targeted-policy >= 1.23.16-1) kudzu-1.1.116.2-2 ----------------- * Thu May 19 2005 Bill Nottingham 1.1.116.2-2 - rebuild against new libpci logwatch-6.0.1-2 ---------------- * Thu May 19 2005 Jiri Ryska 6.0.1-2 - fixed temp dir creation #155795 ncpfs-2.2.4-8 ------------- * Thu May 19 2005 Jiri Ryska - fixed possible buffer overflow CAN-2005-0014 nfs-utils-1.0.7-7 ----------------- * Thu May 19 2005 Steve Dickson 1.0.7-7 - Fixed buffer overflow in rpc.svcgssd (bz 114288) openssl-0.9.7f-7 ---------------- * Thu May 19 2005 Tomas Mraz 0.9.7f-7 - fix CAN-2005-0109 - use constant time/memory access mod_exp so bits of private key aren't leaked by cache eviction (#157631) - a few more fixes from upstream 0.9.7g pam_krb5-2.1.7-3 ---------------- * Wed May 18 2005 Nalin Dahyabhai - 2.1.7-3 - rebuild * Wed May 18 2005 Nalin Dahyabhai - 2.1.7-2 - rebuild * Wed May 18 2005 Nalin Dahyabhai - 2.1.7-1 - update to 2.1.7 - install afs5log and its man page - fix detection of the realm of foreign cells with newer OpenAFS - also try AFS principal = afs at REALM for foreign cells in foreign realms pciutils-2.1.99.test8-10 ------------------------ * Thu May 19 2005 Bill Nottingham - 2.1.99.test8-10 - allow 64-bit addresses on x86_64 (#158217, ) readahead-1:1.1-1.14 -------------------- * Wed May 18 2005 Bill Nottingham - new readahead.c from Ziga Mahkovec - optimizes read access for more throughput - regenerate file lists (#128444) - fix lack of newlines (#146744) - fix lists so that they are architecture-neutral - move check for > 384MB into the init scripts, not the %post rhpl-0.165-1 ------------ * Thu May 19 2005 Paul Nasrat 0.165-1 - ALPS configuration (#155291) rpm-4.4.1-19 ------------ * Tue May 17 2005 Paul Nasrat - 4.4.1-19 - Check for symlinks in check-files (#108778) - Move zh_CN (#154623) - Test fix for signing old rpms (#127113) selinux-policy-strict-1.23.16-5 ------------------------------- * Thu May 19 2005 Dan Walsh 1.23.16-5 - Fix slapd and cups for targeted policy * Wed May 18 2005 Dan Walsh 1.23.16-4 - Add anaconda back in - Fix for nvidia - Fixes for acpi - Fix several ":file read" -> ":file { getattr read }: selinux-policy-targeted-1.23.16-5 --------------------------------- * Thu May 19 2005 Dan Walsh 1.23.16-5 - Fix slapd and cups for targeted policy * Wed May 18 2005 Dan Walsh 1.23.16-4 - Add anaconda back in - Fix for nvidia - Fixes for acpi - Fix several ":file read" -> ":file { getattr read }: sysklogd-1.4.1-30 ----------------- * Thu May 19 2005 Jason Vas Dias 1.4.1rh-30 - fix bug 158205: fix MARK message generation Patch contributed by Ray Van Tassle of Motorola * Mon Mar 28 2005 Jason Vas Dias 1.4.1rh-28 - Fix bug 152319: potential ctime() deadlock in domark() when called - from signal handler * Fri Jan 14 2005 Jason Vas Dias 1.4.1rh-26 - Final fixup of '@host' name checking code: remove possible - duplicates properly syslinux-3.08-2 --------------- * Thu May 19 2005 Peter Jones - 3.08-2 - update to 3.08 xfig-3.2.4-11 ------------- * Thu May 19 2005 Than Ngo 3.2.4-11 - apply patch to fix buffer oveeflow #158088 yelp-2.10.0-1 ------------- * Thu May 19 2005 Ray Strode 2.10.0-1 - Update to 2.10.0 (bug 157752, 146862). * Thu May 19 2005 Christopher Aillon 2.9.3-7 - Depend on mozilla 1.7.8 From Axel.Thimm at ATrpms.net Fri May 20 13:20:23 2005 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Fri, 20 May 2005 15:20:23 +0200 Subject: kpartx & dm-multipath kernel module: Where/when should they be called (wrt FC4)? Message-ID: <200505201626.j4KGQrnj012380@neu.nirvana> Hi, I successfully managed to get multipath support on an EVA 3000 (HSV100). BTW shouldn't the policy for HSV100 be group_by_serial (former group_by_tur)? --- multipath.conf~ 2005-04-01 22:56:49.000000000 +0200 +++ multipath.conf 2005-05-20 14:38:14.000000000 +0200 @@ -60,7 +60,7 @@ device { vendor "HP " product "HSV100 " - path_grouping_policy multibus + path_grouping_policy group_by_serial path_checker tur } device { Anyway now I have my /dev/dm-2 block device and I had to get to the partitions on it. After some search I found that kpartx -a /dev/dm-2 did the right job and extracted "/dev/dm-2p1" onto /dev/dm-3 (is that the indented bahaviour? Getting a new device name?). What I wonder now, is where and how kpartx should be embedded into the distribution's init scripts. It needs to come before any lvm parts as this is an lvm partition. OTOH dm-multipath needs to be modprobed even earlier and currently I'm using a new interface in the init scripts (/etc/sysconfig/modules) which is called after lvm setup. :( Any chance this could still hit FC4? <8-) Thanks! -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From notting at redhat.com Fri May 20 16:38:57 2005 From: notting at redhat.com (Bill Nottingham) Date: Fri, 20 May 2005 12:38:57 -0400 Subject: [PATCH 1.4.11 1/2] OpenIPMI initscript and config file In-Reply-To: <20050520095457.GA32316@lists.us.dell.com> References: <20050520040924.GA25048@lists.us.dell.com> <20050520041031.GB25048@lists.us.dell.com> <20050520041924.GA24328@nostromo.devel.redhat.com> <20050520095457.GA32316@lists.us.dell.com> Message-ID: <20050520163857.GA610@nostromo.devel.redhat.com> Matt Domsch (Matt_Domsch at dell.com) said: > > Why isn't the driver fixed to populate sysfs properly? > > Just-in-time programming. Corey submitted class_simple support this > evening. > http://marc.theaimsgroup.com/?l=linux-kernel&m=111654581209165& > > That'll handle the kernels where it exists, still need to handle the > cases where it doesn't. I'll add a test for existance of the dev > file, and fall back to creating the devnode here if it doesn't exist. Well, if we're talking about current releases, I'd just assume the kernel is fixed; the FC package doesn't really need 2.4 support, for example. > > How many, if any, of these modules can be detected for the > > proper hardware? > > You're thinking of using kudzu instead? Possible, though the data is > scattered throughout ACPI and SMBIOS tables, when it is known. That's > more than I was up for this week. Future incremental change? I'm thinking of using *something*; we (in HEAD) have some ACPI-based PnP code, so grokking in other places could be used. It's better than making people edit a config file. Bill From invest at juun.com Fri May 20 20:10:47 2005 From: invest at juun.com (Philip George) Date: Fri, 20 May 2005 15:10:47 -0500 Subject: how to tell a socket to route thru a specific interface... Message-ID: How can I specify that a socket use a specific interface? If I have a wi-fi connection to one network and a cat5 connection to another, how can i tell my socket to route through a specific one (and not even try the other)? To be clear, I'm talking about an outgoing connection. For example, I use send() a few lines after what is shown here to send a web request and recv() to get the requested page. The catch being that I need to specify that the request be routed through a specific interface, not just any old interface (assuming that at runtime there are 2 or more available connections from the computer out to the internet). I thought it was bind() that did this, but it doesn't seem like that's what it's doing at all. I have a book that led me to believe bind() would tell a socket to route through a specific interface (thus BINDing it to the socket), but I think it's more likely that it's a dns-related call. unfortunately, the man page is strangely ambiguous about what it actually does, using the word 'name' a lot, but never actually saying 'hostname'. In any case, here's the code i've got that isn't working (although I'm pretty sure the setsockopt() line is okay) : // setup local address stuff... struct sockaddr_in LOCaddr; LOCaddr.sin_family = AF_INET; LOCaddr.sin_addr.s_addr = inet_addr(local_ip); // set to true for SO_DONTROUTE... int val = 1; // bind to specific interface and then set SO_DONTROUTE... if (bind(sockethandle, (struct sockaddr *)&LOCaddr, sizeof(LOCaddr))==0) { setsockopt(sockethandle, SOL_SOCKET, SO_DONTROUTE, &val, sizeof(val)); } What do I need to do here instead of bind()? Thanks. - Philip From kyrre at solution-forge.net Fri May 20 20:11:11 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Fri, 20 May 2005 22:11:11 +0200 Subject: What happened to pup? Message-ID: <1116619871.4491.1.camel@localhost.localdomain> What happened to pup - the yum GUI frontend that was so much talked about? Its not in FC4-t3... Kyrre Ness Sj?b?k From pnasrat at redhat.com Fri May 20 21:48:58 2005 From: pnasrat at redhat.com (Paul Nasrat) Date: Fri, 20 May 2005 17:48:58 -0400 Subject: What happened to pup? In-Reply-To: <1116619871.4491.1.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> Message-ID: <1116625739.3467.9.camel@enki.eridu> On Fri, 2005-05-20 at 22:11 +0200, Kyrre Ness Sjobak wrote: > What happened to pup - the yum GUI frontend that was so much talked > about? > > Its not in FC4-t3... It's merely set back. I'll send an update out next week Paul From sundaram at redhat.com Sat May 21 05:53:16 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Sat, 21 May 2005 11:23:16 +0530 Subject: how to tell a socket to route thru a specific interface... In-Reply-To: References: Message-ID: <428ECCCC.50905@redhat.com> Philip George wrote: > How can I specify that a socket use a specific interface? > > If I have a wi-fi connection to one network and a cat5 connection to > another, how can i tell my socket to route through a specific one (and > not even try the other)? This is a mailing list for the development of Fedora itself and not for general development questions. Please ask in the Fedora users mailing list instead http://fedoraproject.org/wiki/PostIsOffTopic regards Rahul From dragoran at feuerpokemon.de Sat May 21 07:41:35 2005 From: dragoran at feuerpokemon.de (dragoran) Date: Sat, 21 May 2005 09:41:35 +0200 Subject: What happened to pup? In-Reply-To: <1116625739.3467.9.camel@enki.eridu> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> Message-ID: <428EE62F.6020707@feuerpokemon.de> Paul Nasrat wrote: >On Fri, 2005-05-20 at 22:11 +0200, Kyrre Ness Sjobak wrote: > > >>What happened to pup - the yum GUI frontend that was so much talked >>about? >> >>Its not in FC4-t3... >> >> > >It's merely set back. I'll send an update out next week > >Paul > > > will it be in fc4 final or will we have to wait until fc5 for it? From kyrre at solution-forge.net Sat May 21 09:43:53 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sat, 21 May 2005 11:43:53 +0200 Subject: What happened to pup? In-Reply-To: <428EE62F.6020707@feuerpokemon.de> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <428EE62F.6020707@feuerpokemon.de> Message-ID: <1116668632.3361.2.camel@localhost.localdomain> l?r, 21.05.2005 kl. 09.41 skrev dragoran: > Paul Nasrat wrote: > > >On Fri, 2005-05-20 at 22:11 +0200, Kyrre Ness Sjobak wrote: > > > > > >>What happened to pup - the yum GUI frontend that was so much talked > >>about? > >> > >>Its not in FC4-t3... > >> > >> > > > >It's merely set back. I'll send an update out next week > > > >Paul > > > > > > > will it be in fc4 final or will we have to wait until fc5 for it? And if not in fc4 core - can it be added to extras? From pnasrat at redhat.com Sat May 21 14:09:34 2005 From: pnasrat at redhat.com (Paul Nasrat) Date: Sat, 21 May 2005 10:09:34 -0400 Subject: What happened to pup? In-Reply-To: <1116668632.3361.2.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <428EE62F.6020707@feuerpokemon.de> <1116668632.3361.2.camel@localhost.localdomain> Message-ID: <1116684574.3582.0.camel@enki.eridu> On Sat, 2005-05-21 at 11:43 +0200, Kyrre Ness Sjobak wrote: > l?r, 21.05.2005 kl. 09.41 skrev dragoran: > > Paul Nasrat wrote: > > >On Fri, 2005-05-20 at 22:11 +0200, Kyrre Ness Sjobak wrote: > > >>What happened to pup - the yum GUI frontend that was so much talked > > >>about? > > >> > > >>Its not in FC4-t3... > > >> > ill it be in fc4 final or will we have to wait until fc5 for it? > And if not in fc4 core - can it be added to extras? Yes it'll go to extras when it's done, and then into next core release. Paul From liste-p.alain at wanadoo.fr Sat May 21 20:13:45 2005 From: liste-p.alain at wanadoo.fr (Aph) Date: Sat, 21 May 2005 22:13:45 +0200 Subject: rpmbuild --sign Message-ID: <200505212213.45684@carola.nyarlathotep> After the last update of Fedora Core 4 Test 3, I can't sign my RPMS : [root at local SPECS]# rpmbuild -ba --sign * (build......) G?n?ration de la signature: 1005 *** glibc detected *** rpmbuild: double free or corruption (!prev): 0x0a151f70 *** ======= Backtrace: ========= /lib/libc.so.6[0x5711e4] /lib/libc.so.6(__libc_free+0x77)[0x57171f] /usr/lib/librpmdb-4.4.so[0x130662] /usr/lib/librpm-4.4.so[0x86133a] /usr/lib/librpm-4.4.so(rpmAddSignature+0x1ce)[0x86195c] /usr/lib/librpmbuild-4.4.so(writeRPM+0x9ab)[0xc44ada] /usr/lib/librpmbuild-4.4.so(packageSources+0x1d2)[0xc45491] /usr/lib/librpmbuild-4.4.so(buildSpec+0x3dc)[0xc3bf37] rpmbuild[0x804a794] rpmbuild[0x804ab64] rpmbuild[0x804b91c] /lib/libc.so.6(__libc_start_main+0xc6)[0x522de6] rpmbuild[0x8049ea1] ======= Memory map: ======== 00111000-00218000 r-xp 00000000 03:0a 2454511 /usr/lib/librpmdb-4.4.so 00218000-0021d000 rwxp 00106000 03:0a 2454511 /usr/lib/librpmdb-4.4.so 0021d000-00264000 r-xp 00000000 03:0a 2443563 /usr/lib/libbeecrypt.so.6.4.0 00264000-00267000 rwxp 00046000 03:0a 2443563 /usr/lib/libbeecrypt.so.6.4.0 00267000-0035f000 r-xp 00000000 03:0a 845330 /lib/libcrypto.so.0.9.7f 0035f000-00371000 rwxp 000f8000 03:0a 845330 /lib/libcrypto.so.0.9.7f 00371000-00374000 rwxp 00371000 00:00 0 00374000-00376000 r-xp 00000000 03:0a 845441 /lib/libdl-2.3.5.so 00376000-00377000 r-xp 00001000 03:0a 845441 /lib/libdl-2.3.5.so 00377000-00378000 rwxp 00002000 03:0a 845441 /lib/libdl-2.3.5.so 00378000-0038a000 r-xp 00000000 03:0a 2453932 /usr/lib/libz.so.1.2.2.2 0038a000-0038b000 rwxp 00011000 03:0a 2453932 /usr/lib/libz.so.1.2.2.2 0038b000-003a1000 r-xp 00000000 03:0a 1367064 /usr/lib/libgssapi_krb5.so.2.2 003a1000-003a2000 rwxp 00016000 03:0a 1367064 /usr/lib/libgssapi_krb5.so.2.2 003a2000-00411000 r-xp 00000000 03:0a 1367061 /usr/lib/libkrb5.so.3.2 00411000-00414000 rwxp 0006e000 03:0a 1367061 /usr/lib/libkrb5.so.3.2 00414000-00416000 r-xp 00000000 03:0a 845336 /lib/libcom_err.so.2.1 00416000-00417000 rwxp 00001000 03:0a 845336 /lib/libcom_err.so.2.1 00417000-0041f000 r-xp 00000000 03:0a 845559 /lib/librt-2.3.5.so 0041f000-00420000 r-xp 00007000 03:0a 845559 /lib/librt-2.3.5.so 00420000-00421000 rwxp 00008000 03:0a 845559 /lib/librt-2.3.5.so 00421000-0042b000 rwxp 00421000 00:00 0 0042b000-0043c000 r-xp 00000000 03:0a 2443189 /usr/lib/libbz2.so.1.0.2 0043c000-0043d000 rwxp 00010000 03:0a 2443189 /usr/lib/libbz2.so.1.0.2 0045a000-0045c000 r-xp 00000000 03:0a 2443118 /usr/lib/gconv/ISO8859-15.so 0045c000-0045e000 rwxp 00001000 03:0a 2443118 /usr/lib/gconv/ISO8859-15.so 004be000-004c7000 r-xp 00000000 03:0a 845488 /lib/libnss_files-2.3.5.so 004c7000-004c8000 r-xp 00008000 03:0a 845488 /lib/libnss_files-2.3.5.so 004c8000-004c9000 rwxp 00009000 03:0a 845488 /lib/libnss_files-2.3.5.so 004d0000-004ea000 r-xp 00000000 03:0a 845401 /lib/ld-2.3.5.so 004ea000-004eb000 r-xp 00019000 03:0a 845401 /lib/ld-2.3.5.so 004eb000-004ec000 rwxp 0001a000 03:0a 845401 /lib/ld-2.3.5.so 0050e000-00632000 r-xp 00000000 03:0a 845324 /lib/libc-2.3.5.so 00632000-00634000 r-xp 00124000 03:0a 845324 /lib/libc-2.3.5.so 00634000-00636000 rwxp 00126000 03:0a 845324 /lib/libc-2.3.5.so 00636000-00638000 rwxp 00636000 00:00 0 00694000-0069d000 r-xp 00000000 03:0a 845314 /lib/libgcc_s-4.0.0-20050518.so.1 0069d000-0069e000 rwxp 00009000 03:0a 845314 /lib/libgcc_s-4.0.0-20050518.so.1 0069e000-0077a000 r-xp 00000000 03:0a 2443399 /usr/lib/libstdc++.so.6.0.4 0077a000-0077f000 rwxp 000dc000 03:0a 2443399 /usr/lib/libstdc++.so.6.0.4 0077f000-00784000 rwxp 0077f000 00:00 0 007d6000-007dd000 r-xp 00000000 03:0a 2447186 /usr/lib/libpopt.so.0.0.0 007dd000-007de000 rwxp 00006000 03:0a 2447186 /usr/lib/libpopt.so.0.0.0 0081d000-00875000 r-xp 00000000 03:0a 2447859 /usr/lib/librpm-4.4.so 00875000-00878000 rwxp 00058000 03:0a 2447859 /usr/lib/librpm-4.4.so 00878000-008ab000 rwxp 00878000 00:00 0 008d2000-008e2000 r-xp 00000000 03:0a 845705 /lib/libselinux.so.1 008e2000-008e3000 rwxp 00010000 03:0a 845705 /lib/libselinux.so.1 0090a000-0091c000 r-xp 00000000 03:0a 2443395 /usr/lib/libelf-0.108.so 0091c000-0091d000 rwxp 00012000 03:0a 2443395 /Abandon [root at local SPECS]# -- > 2 J'en sais s?rement plus que tu ne crois...sur vous tous... -+- W in WeC : Shir Rabin Duval -+- From dcbw at redhat.com Sat May 21 20:28:49 2005 From: dcbw at redhat.com (Dan Williams) Date: Sat, 21 May 2005 16:28:49 -0400 (EDT) Subject: rpmbuild --sign In-Reply-To: <200505212213.45684@carola.nyarlathotep> References: <200505212213.45684@carola.nyarlathotep> Message-ID: On Sat, 21 May 2005, Aph wrote: > After the last update of Fedora Core 4 Test 3, I can't sign my RPMS : > > [root at local SPECS]# rpmbuild -ba --sign * > > (build......) > > G?n?ration de la signature: 1005 > *** glibc detected *** rpmbuild: double free or corruption (!prev): 0x0a151f70 > *** > ======= Backtrace: ========= > /lib/libc.so.6[0x5711e4] > /lib/libc.so.6(__libc_free+0x77)[0x57171f] > /usr/lib/librpmdb-4.4.so[0x130662] > /usr/lib/librpm-4.4.so[0x86133a] > /usr/lib/librpm-4.4.so(rpmAddSignature+0x1ce)[0x86195c] > /usr/lib/librpmbuild-4.4.so(writeRPM+0x9ab)[0xc44ada] > /usr/lib/librpmbuild-4.4.so(packageSources+0x1d2)[0xc45491] > /usr/lib/librpmbuild-4.4.so(buildSpec+0x3dc)[0xc3bf37] > rpmbuild[0x804a794] > rpmbuild[0x804ab64] > rpmbuild[0x804b91c] > /lib/libc.so.6(__libc_start_main+0xc6)[0x522de6] > rpmbuild[0x8049ea1] File it in bugzilla. Dan From nicolas.mailhot at laposte.net Sat May 21 21:58:57 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Sat, 21 May 2005 23:58:57 +0200 Subject: rpmbuild --sign In-Reply-To: References: <200505212213.45684@carola.nyarlathotep> Message-ID: <428FAF21.6050209@laposte.net> Dan Williams a ?crit : > On Sat, 21 May 2005, Aph wrote: > >>After the last update of Fedora Core 4 Test 3, I can't sign my RPMS : >> >>[root at local SPECS]# rpmbuild -ba --sign * >> >>(build......) >> >>G?n?ration de la signature: 1005 >>*** glibc detected *** rpmbuild: double free or corruption (!prev): 0x0a151f70 Already done https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=158382 And it's not only filed but being taken care of (thanks Paul) Regards, -- Nicolas Mailhot -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From buildsys at redhat.com Sun May 22 11:59:04 2005 From: buildsys at redhat.com (Build System) Date: Sun, 22 May 2005 07:59:04 -0400 Subject: rawhide report: 20050522 changes Message-ID: <200505221159.j4MBx4q7003407@porkchop.devel.redhat.com> Updated Packages: (none) From cfk at pacbell.net Sun May 22 12:53:41 2005 From: cfk at pacbell.net (cfk) Date: Sun, 22 May 2005 05:53:41 -0700 Subject: adding an item to menu In-Reply-To: <200505221159.j4MBx4q7003407@porkchop.devel.redhat.com> References: <200505221159.j4MBx4q7003407@porkchop.devel.redhat.com> Message-ID: <200505220553.41902.cfk@pacbell.net> How does one go about adding an executable program to the menu in Fedora Core. I know how to add an item from the menu to the taskbar, or to create a launcher by right clicking the desktop, but I havent figured out how to add an item the the menu yet. This is the menu accessible by clicking the icon of the redhat in either gnome or KDE. I would appreciate guidance applicable to both. Charles From paul at city-fan.org Sun May 22 14:12:02 2005 From: paul at city-fan.org (Paul Howarth) Date: Sun, 22 May 2005 15:12:02 +0100 Subject: adding an item to menu In-Reply-To: <200505220553.41902.cfk@pacbell.net> References: <200505221159.j4MBx4q7003407@porkchop.devel.redhat.com> <200505220553.41902.cfk@pacbell.net> Message-ID: <1116771123.4628.199.camel@laurel.intra.city-fan.org> On Sun, 2005-05-22 at 05:53 -0700, cfk wrote: > How does one go about adding an executable program to the menu in Fedora Core. > > I know how to add an item from the menu to the taskbar, or to create a > launcher by right clicking the desktop, but I havent figured out how to add > an item the the menu yet. > > This is the menu accessible by clicking the icon of the redhat in either gnome > or KDE. I would appreciate guidance applicable to both. You can create a "desktop entry" file such as the ones you'll find in /usr/share/applications. A spec for the desktop entry file can be found at: http://standards.freedesktop.org/desktop-entry-spec/latest/ Paul. -- Paul Howarth From mike at navi.cx Sun May 22 14:27:15 2005 From: mike at navi.cx (Mike Hearn) Date: Sun, 22 May 2005 15:27:15 +0100 Subject: What happened to pup? References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> Message-ID: On Fri, 20 May 2005 17:48:58 -0400, Paul Nasrat wrote: > It's merely set back. I'll send an update out next week Pup is this new "non threatening" yum front-end, right? Could we see screenshots of its current state at least? At the moment I'm not really convinced you can make a non-threatening yet useful frontend over a system like apt or yum ... short of asking people to embed Fedora specific information straight into web pages (which would suck massively) you can't get away from the huge list UI. Even if it's restricted to applications like what Ubuntu are doing, you still end up trying to give users the Ultimate Applications Menu. What I'm saying is, I'd really like to get more details on the UI thinking behind pup. thanks -mike From mattdm at mattdm.org Sun May 22 14:55:12 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Sun, 22 May 2005 10:55:12 -0400 Subject: What happened to pup? In-Reply-To: References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> Message-ID: <20050522145512.GA20689@jadzia.bu.edu> On Sun, May 22, 2005 at 03:27:15PM +0100, Mike Hearn wrote: > Pup is this new "non threatening" yum front-end, right? Could we see > screenshots of its current state at least? > At the moment I'm not really convinced you can make a non-threatening yet > useful frontend over a system like apt or yum ... short of asking people [...] > What I'm saying is, I'd really like to get more details on the UI thinking > behind pup. As I understand it, pup is a package updater front end only -- it doesn't attempt to do anything like let you select new packages to install, or provide a way to manage packages already on the system. For that, you may want to check out yumex: . -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 73 degrees Fahrenheit. From christey at csee.wvu.edu Sun May 22 15:15:14 2005 From: christey at csee.wvu.edu (Damian Christey) Date: Sun, 22 May 2005 11:15:14 -0400 Subject: What happened to pup? Message-ID: <1116774915.25744.18.camel@localhost.localdomain> > On Fri, 2005-05-20 at 22:11 +0200, Kyrre Ness Sjobak wrote: > > What happened to pup - the yum GUI frontend that was so much talked > > about? > > > > Its not in FC4-t3... > > It's merely set back. I'll send an update out next week > > Paul When last I checked out your cvs all I found was a small test app. Has development moved elsewhere? While I'm not suggesting that Pup be abandoned, you and others might be interested in this project I found: http://yumex.sourceforge.net/ Yumex aims to be a simple GUI to yum, written in Python and GLADE, develoment is quite active on the branch for FC4: http://linux.rasmil.dk/cms/modules/news/ At the moment it is quite functional, but could use some UI love in some areas. Cheers, Damian From avi at argo.co.il Sun May 22 15:23:30 2005 From: avi at argo.co.il (Avi Kivity) Date: Sun, 22 May 2005 18:23:30 +0300 Subject: adding an item to menu In-Reply-To: <200505220553.41902.cfk@pacbell.net> References: <200505221159.j4MBx4q7003407@porkchop.devel.redhat.com> <200505220553.41902.cfk@pacbell.net> Message-ID: <1116775410.7086.10.camel@blast.qumranet.com> On Sun, 2005-05-22 at 05:53 -0700, cfk wrote: > How does one go about adding an executable program to the menu in Fedora Core. > > I know how to add an item from the menu to the taskbar, or to create a > launcher by right clicking the desktop, but I havent figured out how to add > an item the the menu yet. > > This is the menu accessible by clicking the icon of the redhat in either gnome > or KDE. I would appreciate guidance applicable to both. > in kde, right click the K Menu button, and select 'Menu Editor' Avi From kyrre at solution-forge.net Sun May 22 15:31:35 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 22 May 2005 17:31:35 +0200 Subject: What happened to pup? In-Reply-To: References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> Message-ID: <1116775894.4322.31.camel@localhost.localdomain> s?n, 22.05.2005 kl. 16.27 skrev Mike Hearn: > On Fri, 20 May 2005 17:48:58 -0400, Paul Nasrat wrote: > > It's merely set back. I'll send an update out next week > > Pup is this new "non threatening" yum front-end, right? Could we see > screenshots of its current state at least? > > At the moment I'm not really convinced you can make a non-threatening yet > useful frontend over a system like apt or yum ... short of asking people > to embed Fedora specific information straight into web pages (which would > suck massively) you can't get away from the huge list UI. Even if it's > restricted to applications like what Ubuntu are doing, you still end up > trying to give users the Ultimate Applications Menu. > > What I'm saying is, I'd really like to get more details on the UI thinking > behind pup. > > thanks -mike While we are talking about the subject, just got a somewhat crazy idea for (3.party) package deployment: Lets say Adobe wanted to make a really, really simple-to-use installer for their reader. (the example could be used with any other 3.party software, F/OSS as well as non-F/OSS, but i'll use adobe in this example). Instead of presenting the user with a massive download page "pick your distro (version), platform, etc." - they gave you a file called AdobeReader.install. The user downloaded this file, and double-click'ed it. This would lauch some helper app, which would intepret the script - within a standard interface - every installer would be more or less the same. Some examples of "stages": 1. Welcome, a brief description of what the program does etc. 2. License agreement 3. which parts should be installed? (probably with one or more sets of standards) 4. some nice progress bars while installing (5. autoupdate on/off?) (6. settings) The install/autoupdate part should be integrated with yum - the installer should put a .repo-file in /etc/yum.repos.d, and yum in the neccesary parts from there - no need for the user to figure "wich rpm does what" - and yum would also pull in the neccessary deps. autoupdate would simply corespond to the "enablerepo" setting. Settings would correspond to the config file (but should not be neccessary). Ofcource, some scriptability (bash/perl perhaps?) would probably be welcome. The .install files should also be able to hold "if-sentences" for different arch'es/os'es (i.e. "if (distro == fedora && version == 2 && arch=i386) { use_repo("www.adobe.com/download/repo/adobereader_fedora2_i386.repo"); } elseif (distro == suse && etc. - but it should probably not be using c-style syntax :) ) In addition to "free download" software, this system should also handle to be put on a cdrom (therefore - either handle paths relatively, or be able to contain data such as rpm's within themselves). Some proprietary vendors would probably also love if yum did make it possible to put passwords on ftp's etc. But this installation method could also be expoited further within fedora itself. If we could create an on-line webpage where information about different packages would be organized, a user could simply visit this page, pick a package he/she would like to have, download the .install file (say, abiword.install). There would (ofcource) *normally* be no need to install - just check that the correct repo is there and enabled. This way, we could simply put a link in the menu, which opened "htmlview http://packages.download.fedora.redhat.com" - and have a nice webpage there. No need for speciality, list-view-type apps. The webpage should ofcource be localized (that is possible throug webagent strings, if i am not mistaken). Comments? Kyrre Ness Sj?b?k From kyrre at solution-forge.net Sun May 22 15:40:53 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 22 May 2005 17:40:53 +0200 Subject: What happened to pup? In-Reply-To: <1116774915.25744.18.camel@localhost.localdomain> References: <1116774915.25744.18.camel@localhost.localdomain> Message-ID: <1116776452.4322.33.camel@localhost.localdomain> s?n, 22.05.2005 kl. 17.15 skrev Damian Christey: > > On Fri, 2005-05-20 at 22:11 +0200, Kyrre Ness Sjobak wrote: > > > What happened to pup - the yum GUI frontend that was so much talked > > > about? > > > > > > Its not in FC4-t3... > > > > It's merely set back. I'll send an update out next week > > > > Paul > > When last I checked out your cvs all I found was a small test app. Has > development moved elsewhere? > > While I'm not suggesting that Pup be abandoned, you and others might be > interested in this project I found: http://yumex.sourceforge.net/ > > Yumex aims to be a simple GUI to yum, written in Python and GLADE, > develoment is quite active on the branch for FC4: > http://linux.rasmil.dk/cms/modules/news/ > > At the moment it is quite functional, but could use some UI love in some > areas. > Shame it's not in extras From mattdm at mattdm.org Sun May 22 16:06:25 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Sun, 22 May 2005 12:06:25 -0400 Subject: What happened to pup? In-Reply-To: <1116776452.4322.33.camel@localhost.localdomain> References: <1116774915.25744.18.camel@localhost.localdomain> <1116776452.4322.33.camel@localhost.localdomain> Message-ID: <20050522160625.GA22902@jadzia.bu.edu> On Sun, May 22, 2005 at 05:40:53PM +0200, Kyrre Ness Sjobak wrote: > > While I'm not suggesting that Pup be abandoned, you and others might be > > interested in this project I found: http://yumex.sourceforge.net/ > > Yumex aims to be a simple GUI to yum, written in Python and GLADE, > > develoment is quite active on the branch for FC4: > > http://linux.rasmil.dk/cms/modules/news/ > > At the moment it is quite functional, but could use some UI love in some > > areas. > Shame it's not in extras I think it'll get there after a few more . releases. -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 72 degrees Fahrenheit. From nicolas.mailhot at laposte.net Sun May 22 16:05:36 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Sun, 22 May 2005 18:05:36 +0200 Subject: What happened to pup? In-Reply-To: <1116775894.4322.31.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> Message-ID: <4290ADD0.3000303@laposte.net> Kyrre Ness Sjobak a ?crit : > Lets say Adobe wanted to make a really, really simple-to-use installer > for their reader ... ROFL What you've described is the windows way and it's nothing but simple. In fact it's so convoluted no one will ever read all the screens you've described (it's nothing but a click ok pipeline - if you really think users process them just add one with the proceed button moved and we'll have massive user consternation) The really, really simple-to-use installer is a web page that uses your browser id to suggest the right repo file to dump into /etc/yum.repos.d with a short example like : su wget -O /etc/yum.repos.d/adobe.repo url yum install acrobat-reader exit (you can add a .repo gui handler if you like but it'll need to be as simple as these four lines) -- Nicolas Mailhot -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From kyrre at solution-forge.net Sun May 22 16:29:51 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 22 May 2005 18:29:51 +0200 Subject: What happened to pup? In-Reply-To: <4290ADD0.3000303@laposte.net> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> Message-ID: <1116779390.4322.41.camel@localhost.localdomain> s?n, 22.05.2005 kl. 18.05 skrev Nicolas Mailhot: > Kyrre Ness Sjobak a ?crit : > > > Lets say Adobe wanted to make a really, really simple-to-use installer > > for their reader > ... > > ROFL > > What you've described is the windows way and it's nothing but simple. In > fact it's so convoluted no one will ever read all the screens you've > described (it's nothing but a click ok pipeline - if you really think > users process them just add one with the proceed button moved and we'll > have massive user consternation) > So what? The user installs what the user needs withot crying "help" on every forum. The user is happy. Period. Yes, it is similar to a windows installer, at least on the scene. So what? It uses the package management system, present the user with some understandable options, and generally gets the job done. Btw. the reason i picked adobe was that it was used as an example on the test list a while ago. Doesn't remember what that was about. > The really, really simple-to-use installer is a web page that uses your > browser id to suggest the right repo file to dump into /etc/yum.repos.d > with a short example like : > > su > wget -O /etc/yum.repos.d/adobe.repo url > yum install acrobat-reader > exit > Something like that is presented at the bottom of the mail, but then in the form as a "userfriendly web synaptic". > (you can add a .repo gui handler if you like but it'll need to be as > simple as these four lines) One of the main points was that i should be able to download an install file once, and run on whatever distro suits me best. Kyrre From overholt at redhat.com Sun May 22 17:06:57 2005 From: overholt at redhat.com (Andrew Overholt) Date: Sun, 22 May 2005 13:06:57 -0400 Subject: rawhide report: 20050519 changes In-Reply-To: References: <200505191536.j4JFa4wR025454@porkchop.devel.redhat.com> Message-ID: <20050522170657.GA27042@redhat.com> * Justin Conover [2005-05-19 12:52]: > > Just curious of the M7 will be released/rebuilt before FC4 is out? We'd like to have it out, but there are a few issues that we're working out first. I hope the have everything sorted out today. Andrew From nicolas.mailhot at laposte.net Sun May 22 17:11:17 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Sun, 22 May 2005 19:11:17 +0200 Subject: What happened to pup? In-Reply-To: <1116779390.4322.41.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> Message-ID: <4290BD35.8070404@laposte.net> Kyrre Ness Sjobak a ?crit : > s?n, 22.05.2005 kl. 18.05 skrev Nicolas Mailhot: > >>Kyrre Ness Sjobak a ?crit : >> >> >>>Lets say Adobe wanted to make a really, really simple-to-use installer >>>for their reader >> >>... >> >>ROFL >> >>What you've described is the windows way and it's nothing but simple. In >>fact it's so convoluted no one will ever read all the screens you've >>described (it's nothing but a click ok pipeline - if you really think >>users process them just add one with the proceed button moved and we'll >>have massive user consternation) >> > So what? The user installs what the user needs withot crying "help" on > every forum. The user is happy. Period. He's not. Just because he cries on other forums you don't actually read does not mean the windows way "works" (a few weeks of windows helpdesk can be very enlightening) Windows installers are terrible but they sort-of work because they all have the same warts and people learn to work around them (click-ok pipeline). Their most crucial aspect is consistency - users will adapt to any UI as long as it does not change every year. Your proposal takes the worst of both worlds : - it inflicts meaningless Windows habits on Linux users - it can not even exploit this similarity since we have real security so run anything as any user won't work here - you'll have to add auth screens not found under windows - it ruins the consistency of the distro installation methods since your proposed installer will be app-specific not system-specific. Instead of learning one update method for all its needs each user will have to learn one method per app. >>The really, really simple-to-use installer is a web page that uses your >>browser id to suggest the right repo file to dump into /etc/yum.repos.d >>with a short example like : >> >>su >>wget -O /etc/yum.repos.d/adobe.repo url >>yum install acrobat-reader >>exit > Something like that is presented at the bottom of the mail, but then in > the form as a "userfriendly web synaptic". > >>(you can add a .repo gui handler if you like but it'll need to be as >>simple as these four lines) > > > One of the main points was that i should be able to download an install > file once, and run on whatever distro suits me best. You're attacking the problem from the wrong side. Who cares if the install method changes per OS? Vendors/developpers care about it because they want to have less work. End-users care about their own workload, which means less installation method variations within their own context, which is one (or two) operating systems, and lots of apps (_NOT_ the reverse). So the crucial part is to use a single method per OS, not a single method per app. That's why people like Dag are worshipped by hordes of end-users - they offer third-party stuff using the single well-known rpm paradigm. The benefits of a "better" installer (which your example is not) pale before the inconvenience of a "different" installer. If you want to change the installation experience you need to retool rpm/yum at the distribution level, not invent yet another marginal universal installer (this is the part the autopackage guys never understood). Regards, -- Nicolas Mailhot -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From alan at redhat.com Sun May 22 17:17:36 2005 From: alan at redhat.com (Alan Cox) Date: Sun, 22 May 2005 13:17:36 -0400 Subject: What happened to pup? In-Reply-To: <1116779390.4322.41.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> Message-ID: <20050522171736.GA19287@devserv.devel.redhat.com> On Sun, May 22, 2005 at 06:29:51PM +0200, Kyrre Ness Sjobak wrote: > > browser id to suggest the right repo file to dump into /etc/yum.repos.d > > with a short example like : > > > > su > > wget -O /etc/yum.repos.d/adobe.repo url > > yum install acrobat-reader > > exit > > > > Something like that is presented at the bottom of the mail, but then in > the form as a "userfriendly web synaptic". If you are going to do the scripts its better to have an XML descriptor for repositories and a helper tool for moz/firefox that knows that media type. You can then click and subscribe to repositories which is *not* the same thing as the windows click/download model because - Repositories have keys - You know *where* a package comes from - You get updates automagically From n3npq at nc.rr.com Sun May 22 17:31:35 2005 From: n3npq at nc.rr.com (Jeff Johnson) Date: Sun, 22 May 2005 13:31:35 -0400 Subject: What happened to pup? In-Reply-To: <20050522171736.GA19287@devserv.devel.redhat.com> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522171736.GA19287@devserv.devel.redhat.com> Message-ID: <4290C1F7.1040606@nc.rr.com> Alan Cox wrote: >On Sun, May 22, 2005 at 06:29:51PM +0200, Kyrre Ness Sjobak wrote: > > >>>browser id to suggest the right repo file to dump into /etc/yum.repos.d >>>with a short example like : >>> >>>su >>>wget -O /etc/yum.repos.d/adobe.repo url >>>yum install acrobat-reader >>>exit >>> >>> >>> >>Something like that is presented at the bottom of the mail, but then in >>the form as a "userfriendly web synaptic". >> >> > >If you are going to do the scripts its better to have an XML descriptor for >repositories and a helper tool for moz/firefox that knows that media type. >You can then click and subscribe to repositories which is *not* the same thing >as the windows click/download model because > >- Repositories have keys >- You know *where* a package comes from >- You get updates automagically > > There's more to package management than choosing a representation for metadata. But clearly the windows installer model is the strongest surviving paradigm. Reboot! Reboot! Reboot! A bit sad, really. 73 de Jeff From mwaite at redhat.com Sun May 22 17:42:15 2005 From: mwaite at redhat.com (Michael Waite) Date: Sun, 22 May 2005 13:42:15 -0400 Subject: netgearWG311 wireless card Message-ID: <1116783735.4109.3.camel@localhost.localdomain> anyone got an rpm for this card? I am trying to get FC3 (rawhide) online for a summer intern that is here this summer. Or, is there a card that is known to work with rawhide? Thanks. ------Mike -- Michael Waite 978-943-9042 mwaite at redhat.com 10 Technology Park Drive Westford, Ma 01876 Learn, Network and Experience Open Source. Red Hat Summit, New Orleans 2005 http://www.redhat.com/promo/summit/ From seyman at wanadoo.fr Sun May 22 17:06:56 2005 From: seyman at wanadoo.fr (Emmanuel Seyman) Date: Sun, 22 May 2005 19:06:56 +0200 Subject: What happened to pup? In-Reply-To: <1116779390.4322.41.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> Message-ID: <20050522170656.GA28937@orient.maison.moi> On Sun, May 22, 2005 at 06:29:51PM +0200, Kyrre Ness Sjobak wrote: > > So what? The user installs what the user needs withot crying "help" on > every forum. The user is happy. Period. I think you're over-estimating the capabilities of the average user here. > One of the main points was that i should be able to download an install > file once, and run on whatever distro suits me best. If the application is FOSS, the optimal way of installing it is by packaging it and putting the rpm in Fedora Extras. If it isn't, an rpm should be availible which installs on any LSB-compliant Linux distribution. Emmanuel From Axel.Thimm at ATrpms.net Sun May 22 17:59:06 2005 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Sun, 22 May 2005 19:59:06 +0200 Subject: netgearWG311 wireless card In-Reply-To: <1116783735.4109.3.camel@localhost.localdomain> References: <1116783735.4109.3.camel@localhost.localdomain> Message-ID: <20050522175906.GA9675@neu.nirvana> On Sun, May 22, 2005 at 01:42:15PM -0400, Michael Waite wrote: > anyone got an rpm for this card? > I am trying to get FC3 (rawhide) online for a summer intern that is here > this summer. Is this with an Atheros chip on it (lspci?)? If yes, then use http://atrpms.net/name/madwifi/ > Or, is there a card that is known to work with rawhide? -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From n3npq at nc.rr.com Sun May 22 18:31:25 2005 From: n3npq at nc.rr.com (Jeff Johnson) Date: Sun, 22 May 2005 14:31:25 -0400 Subject: What happened to pup? In-Reply-To: <20050522170656.GA28937@orient.maison.moi> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> Message-ID: <4290CFFD.6030205@nc.rr.com> Emmanuel Seyman wrote: >If it isn't, an rpm should be availible which installs on any >LSB-compliant Linux distribution. > > Surely you jest. The measure of the set of LSB-compliant packages that install on LSB-compliant Linux distributions is vanishingly small. And the LSB specification for compliant *.rpm packages is useless both theoretically and practically. 73 de Jeff From nutello at sweetness.com Sun May 22 18:49:23 2005 From: nutello at sweetness.com (Rudi Chiarito) Date: Sun, 22 May 2005 20:49:23 +0200 Subject: Standard keytab location Message-ID: <20050522184922.GB27447@plain.rackshack.net> Hi, in the wake of the recent discussion about the location for SSL certificates, I was wondering about the same regarding Kerberos keytabs. The only standard so far is /etc/krb5.keytab. That's the file meant to contain keys for the local machine. It is readable only by root for security reasons. Of course this is a problem for server applications that do not run as root, e.g. httpd. A number of applications provide means to specify an alternate location for the keytab - quite often through the KRB5_KTNAME environment variable. The other benefit of having separate keytab files is that it should reduce risks in case of a security breach and I think it should make it easier to enforce policies on keys with the help of SELinux. Where should these files reside, then? In the application's directory, when present, such as /etc/httpd/ or /etc/openldap/? Or something like /etc/httpd/keytab.d? Maybe /etc/keytabs/ or /etc/krb5.keytabs/? The last two would work for applications that do not use a directory of their own under /etc. What should the files be named? Should packages provide RPM ghost files? Should more than one keytab be supported for a single application? I'm thinking of Apache vhosts - I don't know yet if mod_auth_kerb will be able to handle that. Comments? -- Rudi From kyrre at solution-forge.net Sun May 22 19:25:06 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 22 May 2005 21:25:06 +0200 Subject: What happened to pup? In-Reply-To: <20050522170656.GA28937@orient.maison.moi> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> Message-ID: <1116789905.4322.60.camel@localhost.localdomain> s?n, 22.05.2005 kl. 19.06 skrev Emmanuel Seyman: > On Sun, May 22, 2005 at 06:29:51PM +0200, Kyrre Ness Sjobak wrote: > > > > So what? The user installs what the user needs withot crying "help" on > > every forum. The user is happy. Period. > > I think you're over-estimating the capabilities of the average user here. > > > One of the main points was that i should be able to download an install > > file once, and run on whatever distro suits me best. > > If the application is FOSS, the optimal way of installing it is by > packaging it and putting the rpm in Fedora Extras. > If it isn't, an rpm should be availible which installs on any > LSB-compliant Linux distribution. The point wasn't to replace yum or rpm or repositories - it was to create a user-friendly (read: not command-line. i *personally* love the command line, but then i am not an average user) frontend to yum install. Yum update would be handeled by pup. It is not like my idea for an interface is set in stone. It should probably be a standard format of metadata, which can be shared between many frontends. It was an idea about how to present theese things to the user, in a way the user understands: - Which rpm's do we need? - Fixing deps - Displaying license - Installing Today, this is done through the command line. Now, try explaining to somebody who hasn't got a clue about computers, how to install a piece of software today: - open a command line - login as root (su -) - find the correct package by yum search or other means - install it by yum install - possibly configure it - keep it up2date manually Or worse: - find and download an rpm - double click it, find that foo and bar is missing - go find foo and bar - install foo and bar - find that bar needs baz - find baz - install baz - install foo and bar - install the initial rpm Or EVEN worse: - find and download an rpm packaged within a shell archive to present the lisence - press enter the exact nr. of times to not overshoot the "Accept?: [y/N]" - install the rpm produced in the way descibed above. Now with this system: - Go to packages.download.redhat.com - click "extras" - find the correct package - look at at nice webpage describing the package - click the "install program" link - save the install file on the desktop, open in XXX/click "open in XXX" - enter root-password - See the license (possibly the GPL), accept it - select witch portions of the app you need if it is packaged into more than one rpm - install through yum - have pup update it automatically throug yum Which method would you describe to a newbie? Ofcource, in the ideal world, every computer would be delivered with a technican/secretary, hiding within the case when not in use... Or not. This is not in any way replacing yum/rpm - it is merely a frontend for them. Just trying to think outside of the synaptic box. Kyrre From pedro.lamarao at mndfck.org Sun May 22 19:44:07 2005 From: pedro.lamarao at mndfck.org (=?ISO-8859-15?Q?Pedro_Lamar=E3o?=) Date: Sun, 22 May 2005 16:44:07 -0300 Subject: What happened to pup? In-Reply-To: <1116789905.4322.60.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> Message-ID: <4290E107.6050404@mndfck.org> Kyrre Ness Sjobak wrote: [SNIP] > Today, this is done through the command line. Now, try explaining to > somebody who hasn't got a clue about computers, how to install a piece > of software today: > - open a command line > - login as root (su -) > - find the correct package by yum search or other means > - install it by yum install > - possibly configure it > - keep it up2date manually This usually works just fine in my experience as member of the #fedora channel. Generally people doesn't seem to have a problem understanding this procedure. Maybe the worst problem is finding a repository with a given package, but Core, Extras and Livna usually do it. > Or worse: > - find and download an rpm > - double click it, find that foo and bar is missing > - go find foo and bar > - install foo and bar > - find that bar needs baz > - find baz > - install baz > - install foo and bar > - install the initial rpm The above imply a situation in which the desired software has a dependency not present in the system. The Windows solution to this problem is to bundle every needed library with the application, effectively installing multiple copies of the same library in the system. Please don't argue that libraries can be installed in System32 as that is unmanageable in Windows and leads to DLL Hell, the reason why packages carry their own libraries in the first place. How does your solution deal with this situation? -- Pedro Lamar?o "Merda faz as flores crescerem e isso ? bonito." From alan at redhat.com Sun May 22 19:53:41 2005 From: alan at redhat.com (Alan Cox) Date: Sun, 22 May 2005 15:53:41 -0400 Subject: What happened to pup? In-Reply-To: <4290C1F7.1040606@nc.rr.com> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522171736.GA19287@devserv.devel.redhat.com> <4290C1F7.1040606@nc.rr.com> Message-ID: <20050522195341.GA16468@devserv.devel.redhat.com> On Sun, May 22, 2005 at 01:31:35PM -0400, Jeff Johnson wrote: > There's more to package management than choosing a representation for > metadata. Indeed there is but at the moment yum doesn't understand heirarchy rules for repositories - not that I'm sure it should. Also for packages going to Fedora/RHEL/related systems that are applications it probably is generally the case there are no complex dependancies (as opposed to shipping libraries) and you could if need be add release related info to the XML (in fact since yum has $foo macros it 'just works'). Solving the entire problem is hard, solving the bit that matters to most users is not. From nicolas.mailhot at laposte.net Sun May 22 19:54:03 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Sun, 22 May 2005 21:54:03 +0200 Subject: What happened to pup? In-Reply-To: <1116789905.4322.60.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> Message-ID: <4290E35B.70306@laposte.net> Kyrre Ness Sjobak a ?crit : > Which method would you describe to a newbie? Well just try to actually document what you propose (nice walkthrough with localised screenshots) and we'll laught a bit. Been there, done that, gui = documentationsize ? 5 If you had done this kind of work before you wouldn't even be proposing to add a license screen, as it adds nothing to the service and is a pain to document (like the rest, but the rest at least has some purpose) I posted a proposal that does what the user actually wants with minimal steps - easy to document and to follow. Alan wrote about how making it slightly better. You just pile up screens that won't make the user (or your) life any easier. Regards, -- Nicolas Mailhot -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From jspaleta at gmail.com Sun May 22 19:59:09 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Sun, 22 May 2005 15:59:09 -0400 Subject: What happened to pup? In-Reply-To: <1116789905.4322.60.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> Message-ID: <604aa7910505221259714d1d68@mail.gmail.com> On 5/22/05, Kyrre Ness Sjobak wrote: >- find the correct package thats a deceptively simple statement and hides the complexity of building a way to browse through packages so that novice users can find the 'correct' package among a technically diverse and yet generally similar set of software. Its not enough to browse by software name... since most project names do not connote functional or technical specfics. You certaintly can't expect 'novice' users to know which web browser is the 'correct' browser for them simply by looking package names. And its not enough to browse by general functionality.. because the 'correct' software for any role can be situationally dependant on what other people already have installed. Or it could be they are looking for a very specific feature instead of an easily described broad function. And lets not forget about license...the 'correct' package could very well be a BSD licensed application instead of a GPL licensed one. > - See the license (possibly the GPL), accept it And.... what if that one application needs several dependancies...each dependancy with a different license? Shall we loop over all dependancies as well.. a license review pane for each and every dependancy package? > Which method would you describe to a newbie? I don't think what you describe is inherently friendlier. More visual...but not friendlier. I personally think software installation/removal interface aimed at novice user/admins should mimic the menu structure novice users will be interacting with on a daily basis. Instead of thinking about 'packages' at all.. you build an interface that leads novice user/admins to interact with application menu items. Someone wants to install an application they fire up an install software program that presents the user/admin with the full applications menu showing all available applications just like they would appear in the gnome menu once installed. You can extend this with a useful tooltip that gives a brief summary of the package. You can extend this further and provided a properties item in a right click menu to see items like license or vendor. Packages without menu items. simply don't show up in this novice oriented tool.. reducing confusion. All is not lost, since you could easily extend the menu metaphor by providing an Install dialog similar to the Run application dialog for more advanced user/admins who know which packagenames they want to install. -jef From kyrre at solution-forge.net Sun May 22 20:42:00 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 22 May 2005 22:42:00 +0200 Subject: What happened to pup? In-Reply-To: <4290E107.6050404@mndfck.org> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> <4290E107.6050404@mndfck.org> Message-ID: <1116794520.4322.67.camel@localhost.localdomain> s?n, 22.05.2005 kl. 21.44 skrev Pedro Lamar?o: > Kyrre Ness Sjobak wrote: > > [SNIP] > > > Today, this is done through the command line. Now, try explaining to > > somebody who hasn't got a clue about computers, how to install a piece > > of software today: > > - open a command line > > - login as root (su -) > > - find the correct package by yum search or other means > > - install it by yum install > > - possibly configure it > > - keep it up2date manually > > This usually works just fine in my experience as member of the #fedora > channel. Generally people doesn't seem to have a problem understanding > this procedure. > > Maybe the worst problem is finding a repository with a given package, > but Core, Extras and Livna usually do it. > > > Or worse: > > - find and download an rpm > > - double click it, find that foo and bar is missing > > - go find foo and bar > > - install foo and bar > > - find that bar needs baz > > - find baz > > - install baz > > - install foo and bar > > - install the initial rpm > > The above imply a situation in which the desired software has a > dependency not present in the system. > > The Windows solution to this problem is to bundle every needed library > with the application, effectively installing multiple copies of the same > library in the system. Please don't argue that libraries can be > installed in System32 as that is unmanageable in Windows and leads to > DLL Hell, the reason why packages carry their own libraries in the first > place. > > How does your solution deal with this situation? By carrying their own .repo file. (one or more) yum then resolves the situation by pulling in the neccesary deps. This is not trying to solve problems related to program packaging itself - yum and rpm handles that, and handles it without problems. It should be noted that this solution (or a similar one) should probably be used to handle the gui-interaction with .rpm files from nautilus - i.e. when a user clicks a .rpm file, this program is started, and this program uses yum to handle the deps (as opposed to system-config-packages today). But the program cannot be kept up2date this way. From kyrre at solution-forge.net Sun May 22 20:52:41 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 22 May 2005 22:52:41 +0200 Subject: What happened to pup? In-Reply-To: <4290E35B.70306@laposte.net> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> <4290E35B.70306@laposte.net> Message-ID: <1116795160.4322.77.camel@localhost.localdomain> s?n, 22.05.2005 kl. 21.54 skrev Nicolas Mailhot: > Kyrre Ness Sjobak a ?crit : > > > Which method would you describe to a newbie? > > Well just try to actually document what you propose (nice walkthrough > with localised screenshots) and we'll laught a bit. > > Been there, done that, gui = documentationsize ? 5 > > If you had done this kind of work before you wouldn't even be proposing > to add a license screen, as it adds nothing to the service and is a pain > to document (like the rest, but the rest at least has some purpose) > > I posted a proposal that does what the user actually wants with minimal > steps - easy to document and to follow. Alan wrote about how making it > slightly better. You just pile up screens that won't make the user (or > your) life any easier. You mean this?: The really, really simple-to-use installer is a web page that uses your browser id to suggest the right repo file to dump into /etc/yum.repos.d with a short example like : su wget -O /etc/yum.repos.d/adobe.repo url yum install acrobat-reader exit (you can add a .repo gui handler if you like but it'll need to be as simple as these four lines) That is basically the same. Only thing is, that i want to push the responsibility to select the correct distro etc from the webserver to a special program on the client - as it is easier to do, and more reliable. Plus, no special config needed on the server (possibly except running a yum repo). And it works off removable storage. The licence screen - as i said: make it optional. Make it into a weblink ala "Licence: Gnu GPL" (and then the licence name isn't a common name, but the name of an individual object (i don't know the correct English term for that, sorry) - and doesn't need to be localized. You could possibly squeeze the installer into three, standard screens (followed by *every* installer, as the .install is just a bunch of metadata presented by the universal-gui-installer-frontend) (something i actually mentioned in the first mail, albeit probably not clear enough) Kyrre From nicolas.mailhot at laposte.net Sun May 22 21:11:03 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Sun, 22 May 2005 23:11:03 +0200 Subject: What happened to pup? In-Reply-To: <1116795160.4322.77.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> <4290E35B.70306@laposte.net> <1116795160.4322.77.camel@localhost.localdomain> Message-ID: <4290F567.8010101@laposte.net> Kyrre Ness Sjobak a ?crit : > That is basically the same. Only thing is, that i want to push the > responsibility to select the correct distro etc from the webserver to a > special program on the client - as it is easier to do, and more > reliable. Ok, as long as this program is provided by the distro itself, not a bunch of semi-broken scripts cooked by the third-party vendor (as it's usually the case). And it's rigid enough the software provider can not make it do fancy stuff to appease marketo?ds and legal vultures. ie no optional feel-good screens - if people want to be creative they can play on their own webserver. (not that the usual closed vendor won't abuse to death rpm scriptlets to hose the system anyway) As others noted, the problem usually is not the installation but the fact no clean package exists at all. Regards, -- Nicolas Mailhot -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From kyrre at solution-forge.net Sun May 22 21:19:10 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Sun, 22 May 2005 23:19:10 +0200 Subject: What happened to pup? In-Reply-To: <604aa7910505221259714d1d68@mail.gmail.com> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> <604aa7910505221259714d1d68@mail.gmail.com> Message-ID: <1116796750.4322.86.camel@localhost.localdomain> s?n, 22.05.2005 kl. 21.59 skrev Jeff Spaleta: > On 5/22/05, Kyrre Ness Sjobak wrote: > >- find the correct package > thats a deceptively simple statement and hides the complexity of > building a way to browse through packages so that novice users can > find the 'correct' package among a technically diverse and yet > generally similar set of software. Its not enough to browse by > software name... since most project names do not connote functional or > technical specfics. You certaintly can't expect 'novice' users to > know which web browser is the 'correct' browser for them simply by > looking package names. > > And its not enough to browse by general functionality.. because the > 'correct' software for any role can be situationally dependant on what > other people already have installed. Or it could be they are looking > for a very specific feature instead of an easily described broad > function. > And lets not forget about license...the 'correct' package could very > well be a BSD licensed application instead of a GPL licensed one. > > That is the job of a well-written web portal. > > - See the license (possibly the GPL), accept it > > And.... what if that one application needs several dependancies...each > dependancy with a different license? Shall we loop over all > dependancies as well.. a license review pane for each and every > dependancy package? > Auch. That i did not think about. But surely, some solution could be found. > > Which method would you describe to a newbie? > I don't think what you describe is inherently friendlier. More > visual...but not friendlier. > But more intuitive. Almost every user "discovers" up2date immediatly, and try to use that. They do not discover yum before someone points them to it. > I personally think software installation/removal interface aimed at > novice user/admins should mimic the menu structure novice users will > be interacting with on a daily basis. Instead of thinking about > 'packages' at all.. you build an interface that leads novice > user/admins to interact with application menu items. Someone wants to > install an application they fire up an install software program that > presents the user/admin with the full applications menu showing all > available applications just like they would appear in the gnome menu > once installed. > > You can extend this with a useful tooltip that gives a brief summary > of the package. You can extend this further and provided a properties > item in a right click menu to see items like license or vendor. > Packages without menu items. simply don't show up in this novice > oriented tool.. reducing confusion. All is not lost, since you could > easily extend the menu metaphor by providing an Install dialog similar > to the Run application dialog for more advanced user/admins who know > which packagenames they want to install. > > Also a good idea - but what about libs, such as gstreamer-mp3? BTW. doesn't Ubuntu use an interface similar to this one? Kyrre From mike at navi.cx Sun May 22 21:24:31 2005 From: mike at navi.cx (Mike Hearn) Date: Sun, 22 May 2005 22:24:31 +0100 Subject: What happened to pup? References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> Message-ID: On Sun, 22 May 2005 17:31:35 +0200, Kyrre Ness Sjobak wrote: > Lets say Adobe wanted to make a really, really simple-to-use installer > for their reader. You're starting out on the same line of thinking that led me to write autopackage. For what it's worth here's what we're thinking of for The Next Level in usability: http://autopackage.org/ui-vision.html All a long way off yet. Essentially, the idea is: * Add an extension to .desktop files which points to an autopackage or luau XML file (this is the part where you can hook native packages like RPMs and DEBs in as well) * Define a new XML namespace and element such that you can embed .desktop files into web pages, and the browser will render them as it would be rendered inside Nautilus, see the mockups if I'm not clear enough. For XHTML based web pages it can be embedded directly. For legacy pages, you could use an IFRAME that points to a piece of XHTML/XML. Gecko already supports this sort of thing: http://www.croczilla.com/xtf Why XTF and not a traditional Netscape style plugin? Well, you really want to be able to control the rendering precisely and defer certain things to Gecko - like rendering SVG icons which are blended on top of web page content. * Allow the user to click and/or drag-drop these embedded launcher icon things. So you can drag them into the panel for instance, or onto your desktop, or into an email, or just over the applications menu. These .desktop files resemble "advertised shortcuts" in MSI. They contain enough information to launch them and if the app is missing, install them. The desktop/shell can use this information to trigger automatic installation. * Because GNOME/KDE wish to remain package manager agnostic (which is right and good), you need some kind of abstraction so they can stay portable whilst still integrating deeply with the base OS. For hardware integration the HAL is being built and it seems to be working nicely. Nobody seems to have any hangups about using it and it adds value. For packaging, there could be an equivalent, one design of which is described here: http://live.gnome.org/PackagingAbstractionLayer None of this desktop integration need be autopackage specific. However, I think it may actually be better if it was from a holistic usability perspective because autopackage is distribution independent (to a large extent). The problem with doing this in the way people are talking about here, with adding repositories and stuff, is that it's got bad usability. Generally speaking, it doesn't matter how slick or streamlined your GUI is if it fails fundamental usability principles like: - Machine model should match users mental model - System should be safe, and allow exploration - System should be predictable and reliable It's the last one that concerns me here. If you go about relying on repositories or Fedora Extras you can get these situations: Bob goes to neat-app.org on his desktop computer and clicks the icon. After confirming, it automatically installs and runs. He is happy. Bob switches on his laptop, and does the same thing. He clicks the icon and gets an error message he doesn't understand. What happened? His laptop was running Fedora Core 3 and his desktop was running Fedora Core 4. There are no RPMs that work on Fedora Core 3, so it failed despite both desktop and laptop *appearing* to be pretty much the same. Alice finds out about GWhizz, goes to the website, clicks to install and it runs. She is happy. Because she thinks it's so cool she gets onto her instant messaging network and tells Sarah about this amazing new GWhizz and sends her the URL. Sarah clicks the icon and nothing happens. She is not happy. Neither her nor Alice can figure out why it works in one place but not the other. What happened? Alice was using Fedora, but Sarah was using Ubuntu. The desktop and theme are the same though so they, at first glance, appear to be identical. In both cases the systems are unusable not because they failed (no system is perfect) but because they are *unpredictable*. Eventually users will learn that this system (and computers in general) cannot be trusted and will flake out just when you need them most. Bad news. Now imagine only universal packages work with this UI. Maybe they're LSB RPMs, maybe they're autopackages, the important thing is they work on any reasonably sane Linux distribution. FreeBSD and other non-Linux systems aren't really an issue here because the type of people that need very simple, straightforward UI are not the type who will be using an OS outside of {Linux, MacOS, Windows}. Now it works whenever you click the icon, or at least gets a lot further. The system is more predictable, more reliable, and therefore more usable. People may raise security as an issue but this is orthogonal to usability: making things "secure" by being hard to use is just security through obscurity, which doesn't work very well. Better solutions would be things like sandboxed installs, whitelisting networks and so on. It's discussed in the FAQ section of the autopackage website. thanks -mike From mike at navi.cx Sun May 22 21:30:12 2005 From: mike at navi.cx (Mike Hearn) Date: Sun, 22 May 2005 22:30:12 +0100 Subject: What happened to pup? References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> Message-ID: On Sun, 22 May 2005 22:24:31 +0100, Mike Hearn wrote: > Why XTF and not a traditional Netscape style plugin? Well, you really > want to be able to control the rendering precisely and defer certain > things to Gecko - like rendering SVG icons which are blended on top of > web page content. Oh, I forgot to mention some other advantages of using XTF: - Browsers that don't support it will cleanly ignore the tags and so that CSS block will just disappear. That makes web design a *lot* easier as you don't need hacky JavaScript to figure out whether to add the icon or not - Semantically cleaner: a (or whatever) tag is so much nicer than random goop - Makes it (reasonably) implementation independent - Easier to extend with extra attributes or tags later instead of the system thanks -mike From seyman at wanadoo.fr Sun May 22 22:24:04 2005 From: seyman at wanadoo.fr (Emmanuel Seyman) Date: Mon, 23 May 2005 00:24:04 +0200 Subject: What happened to pup? In-Reply-To: <4290CFFD.6030205@nc.rr.com> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <4290CFFD.6030205@nc.rr.com> Message-ID: <20050522222404.GA29620@orient.maison.moi> On Sun, May 22, 2005 at 02:31:25PM -0400, Jeff Johnson wrote: > > Surely you jest. The measure of the set of LSB-compliant packages that > install on LSB-compliant Linux distributions is vanishingly small. Probably (I don't care enough about non-FOSS packages to install them). > And the LSB specification for compliant *.rpm packages is useless both > theoretically and practically. Will this be fixed in future versions of LSB? Emmanuel From naheemzaffar at gmail.com Mon May 23 00:59:50 2005 From: naheemzaffar at gmail.com (Naheem Zaffar) Date: Mon, 23 May 2005 01:59:50 +0100 Subject: User Friendly Install In-Reply-To: <20050522195546.620937364D@hormel.redhat.com> References: <20050522195546.620937364D@hormel.redhat.com> Message-ID: <3adc77210505221759509be784@mail.gmail.com> Hi This was RE: what happened to pup? I have been reading the discussion regarding how we should handle installs that benefits an end user. So far the discussion has been regarding web interfaces etc. that randomly invoke yum, and add a repo file (If I understand correctly). What I as an end user would probably prefer would be something slightly different. A local scripted folder which is also referenced in yum. How this could work: 1. There is a local.repo file linking to a local folder. 2. I the end user search Fedora for an RPM, but do not find one. 3. I find an rpm file. (there is a trust issue here, just like there is with a website automatically creating a .repo file). 4. I save it onto my computer, and drag it onto the localrepo folder. 5. Yum is automatically invoked by the scripted folder, calculating any dependencies. 6. To uninstall, just drag the rpm out of the localrepo folder, automatically invoking 'yum remove'... Ofcourse this method will not automatically update files (as the localrepo is only updated by the individual, not the rpm owner). I do not know if Nautilus can have scriptable folders, or yum install from a local folder, so these could be issues that need to be worked on. This method takes the mac approach and IMO beats it hands down, also having any benefits of the windows approach without most of the drawbacks. PS I am an End User, not a developer, so developers may disagree with me. -------------- next part -------------- An HTML attachment was scrubbed... URL: From seandarcy2 at gmail.com Mon May 23 01:45:11 2005 From: seandarcy2 at gmail.com (sean) Date: Sun, 22 May 2005 21:45:11 -0400 Subject: Any reason we're using libdv-0.103? Message-ID: libdv-0.104 was announced in November, 2004. We're using 0.103. I could hack the spec file from 0.103 to build 0.104, but I figured I'd better ask if there's a reason to stick with 0.103. sean From wtogami at redhat.com Mon May 23 01:51:25 2005 From: wtogami at redhat.com (Warren Togami) Date: Sun, 22 May 2005 15:51:25 -1000 Subject: Any reason we're using libdv-0.103? In-Reply-To: References: Message-ID: <4291371D.7050806@redhat.com> sean wrote: > libdv-0.104 was announced in November, 2004. We're using 0.103. I could > hack the spec file from 0.103 to build 0.104, but I figured I'd better > ask if there's a reason to stick with 0.103. > > sean > 0.104 wouldn't even build on x86_64. Upstream clearly didn't know what they were doing when they added x86_64 specific asm. There is little real benefit to 0.104. We'll get there eventually for FC5, hopefully after upstream makes a real release that works on all archs. FC4 Updates maybe... if there is a real reason to do so. Warren Togami wtogami at redhat.com From ivg2 at cornell.edu Mon May 23 02:22:14 2005 From: ivg2 at cornell.edu (Ivan Gyurdiev) Date: Sun, 22 May 2005 22:22:14 -0400 Subject: netgearWG311 wireless card In-Reply-To: <20050522175906.GA9675@neu.nirvana> References: <1116783735.4109.3.camel@localhost.localdomain> <20050522175906.GA9675@neu.nirvana> Message-ID: <1116814934.8110.0.camel@localhost.localdomain> On Sun, 2005-05-22 at 19:59 +0200, Axel Thimm wrote: > On Sun, May 22, 2005 at 01:42:15PM -0400, Michael Waite wrote: > > anyone got an rpm for this card? > > I am trying to get FC3 (rawhide) online for a summer intern that is here > > this summer. > > Is this with an Atheros chip on it (lspci?)? If yes, then use > http://atrpms.net/name/madwifi/ Thanks.. I'm set now. The card in question did not have an Atheros chip, but I bought another one that does. Madwifi seems to work. -- Ivan Gyurdiev Cornell University From mattdm at mattdm.org Mon May 23 03:24:35 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Sun, 22 May 2005 23:24:35 -0400 Subject: User Friendly Install In-Reply-To: <3adc77210505221759509be784@mail.gmail.com> References: <20050522195546.620937364D@hormel.redhat.com> <3adc77210505221759509be784@mail.gmail.com> Message-ID: <20050523032434.GA8134@jadzia.bu.edu> On Mon, May 23, 2005 at 01:59:50AM +0100, Naheem Zaffar wrote: > I do not know if Nautilus can have scriptable folders, or yum install from > a local folder, so these could be issues that need to be worked on. Recent versions of yum can indeed install from local files (and solve dependencies using any enabled repositories), so something like what you're proposing might be workable. I'm not convinced it's the best way, but it wouldn't be terribly hard. -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 72 degrees Fahrenheit. From russell at coker.com.au Mon May 23 09:17:54 2005 From: russell at coker.com.au (Russell Coker) Date: Mon, 23 May 2005 19:17:54 +1000 Subject: talks about Fedora Message-ID: <200505231917.58860.russell@coker.com.au> I want to give a talk about the new features in FC4, the aim is for a 30 min talk on the day of the release. Anyone working on notes for such a talk? If not does anyone have suggestions about what new features should be mentioned? Also I am considering offering a talk on Fedora on PPC, however not knowing much about PPC I would need to have the notes mostly written for me. Is there anyone who has such notes? The talk would have to go for 90 mins. -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From sundaram at redhat.com Mon May 23 09:24:03 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Mon, 23 May 2005 14:54:03 +0530 Subject: talks about Fedora In-Reply-To: <200505231917.58860.russell@coker.com.au> References: <200505231917.58860.russell@coker.com.au> Message-ID: <4291A133.50807@redhat.com> Russell Coker wrote: >I want to give a talk about the new features in FC4, the aim is for a 30 min >talk on the day of the release. Anyone working on notes for such a talk? If >not does anyone have suggestions about what new features should be mentioned? > > The following is a draft copy of the document that I am working on . This is meant to be appealing to non-technical end users http://people.redhat.com/sundaram/fedora_notes.html You might also want to mention the boot optimisation work using bootchart and improvements in readahead depending on the audience regards Rahul From sundaram at redhat.com Mon May 23 09:28:12 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Mon, 23 May 2005 14:58:12 +0530 Subject: talks about Fedora In-Reply-To: <4291A133.50807@redhat.com> References: <200505231917.58860.russell@coker.com.au> <4291A133.50807@redhat.com> Message-ID: <4291A22C.2030708@redhat.com> Rahul Sundaram wrote: > Russell Coker wrote: > >> I want to give a talk about the new features in FC4, the aim is for a >> 30 min talk on the day of the release. Anyone working on notes for >> such a talk? If not does anyone have suggestions about what new >> features should be mentioned? >> >> > > The following is a draft copy of the document that I am working on . > This is meant to be appealing to non-technical end users > > http://people.redhat.com/sundaram/fedora_notes.html > > You might also want to mention the boot optimisation work using > bootchart and improvements in readahead depending on the audience and also that Fedora Extras has around 1000+ packages maintained by the community that are in sync with Fedora Core and is enabled by default. regards Rahul From sundaram at redhat.com Mon May 23 09:38:59 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Mon, 23 May 2005 15:08:59 +0530 Subject: talks about Fedora In-Reply-To: <200505231917.58860.russell@coker.com.au> References: <200505231917.58860.russell@coker.com.au> Message-ID: <4291A4B3.20109@redhat.com> Hi >Also I am considering offering a talk on Fedora on PPC, however not knowing >much about PPC I would need to have the notes mostly written for me. Is >there anyone who has such notes? The talk would have to go for 90 mins. > > > This might also be useful for PPC http://www.redhat.com/magazine/007may05/features/mac-mini/ regards Rahul From buildsys at redhat.com Mon May 23 11:49:12 2005 From: buildsys at redhat.com (Build System) Date: Mon, 23 May 2005 07:49:12 -0400 Subject: rawhide report: 20050523 changes Message-ID: <200505231149.j4NBnCHP001934@porkchop.devel.redhat.com> Updated Packages: audit-0.8.2-1 ------------- * Fri May 20 2005 Steve Grubb 0.8.2-1 - Update documentation - Handle user space audit events in more uniform way - Update all parsers for more robustness with new kernel changes - Create quiet mode for error messages - Make rotated logs readonly eclipse-1:3.1.0_fc-0.M6.22 -------------------------- * Sun May 22 2005 Ben Konrath 3.1.0_fc-0.M6.22 - Bump required version of java-gcj-compat to the latest (-40jpp_24rh). - Add patch to make swt use libgcjawt instead of libjawt for gcj. * Fri May 20 2005 Ben Konrath 3.1.0_fc-0.M6.21 - Add ecj-options patch to bootstrap source. - Make embedded browser widget work (Robin Green). gnome-desktop-2.10.0-4 ---------------------- * Mon May 16 2005 Ray Strode - 2.10.0-4 - run gettext initialization routines on startup (bug 155659) (use right patch). gnome-media-2.10.2-4 -------------------- * Fri May 20 2005 John (J5) Palmieri 2.10.2-4 - patch that avoids a deadlock when the last track on the cd is done (bug #151093) kernel-2.6.11-1.1340_FC4 ------------------------ * Sat May 21 2005 Dave Jones - Fix divide by zero in ipw2100 driver. (#158406) - 2.6.12-rc4-git5 More x86-64 updates, Further pktcdvd frobbing, yet more dvb updates, x86(64) ioremap fixes, ppc updates, IPMI sysfs support (reverted for now due to breakage), various SCSI fixes (aix7xxx, spi transport), vmalloc improvements * Sat May 21 2005 David Woodhouse - Fix oops in avc_audit() (#158377) - Include serial numbers in non-syscall audit messages * Sat May 21 2005 Bill Nottingham - bump ipw2200 conflict nss_ldap-234-4 -------------- * Fri May 20 2005 Nalin Dahyabhai 234-4 - override glibc version detection so that mismatches between the versions of 32- and 64-bit glibc don't result in our %install installing the module with a different name than the 'make install' target uses * Fri May 20 2005 Nalin Dahyabhai 234-3 - fix type mismatch bug in patch for using non-blocking start_tls in preference to the blocking version when it's available (#156582) redhat-artwork-0.122-10 ----------------------- * Sun May 22 2005 John (J5) Palmieri 0.122-10 - Add bluecurve weather icons for the weather applet rpm-4.4.1-20 ------------ * Sat May 21 2005 Paul Nasrat - 4.4.1-20 - Drop signature patch - dangling unpackaged symlinks udev-058-1 ---------- * Fri May 20 2005 Bill Nottingham - 058-1 - update to 058, fixes conflict with newer kernels (#158371) * Thu May 12 2005 Harald Hoyer - 057-6 - polished persistent scripts yum-2.3.2-3 ----------- * Fri May 20 2005 Jeremy Katz - 2.3.2-3 - add fixes from Seth for the shell to run depsolve and to clean up output somewhat (#158267) From naheemzaffar at gmail.com Mon May 23 11:50:43 2005 From: naheemzaffar at gmail.com (Naheem Zaffar) Date: Mon, 23 May 2005 12:50:43 +0100 Subject: User Friendly Install In-Reply-To: <20050523092412.367AA733AE@hormel.redhat.com> References: <20050523092412.367AA733AE@hormel.redhat.com> Message-ID: <3adc772105052304502a8c57fc@mail.gmail.com> From: Matthew Miller Subject: Re: User Friendly Install To: Development discussions related to Fedora Core Message-ID: < 20050523032434.GA8134 at jadzia.bu.edu> Content-Type: text/plain; charset=us-ascii On Mon, May 23, 2005 at 01:59:50AM +0100, Naheem Zaffar wrote: >> I do not know if Nautilus can have scriptable folders, or yum install from >> a local folder, so these could be issues that need to be worked on. >Recent versions of yum can indeed install from local files (and solve >dependencies using any enabled repositories), so something like what >you'reproposing might be workable. I'm not convinced it's the best way, but it >wouldn't be terribly hard. Its not the most elegant solution, but it solves the problem at hand without forcing package providers to provide a complete web architecture. The problem is allowing people to install packages without going into 'DLL Hell'. The other solutions are far more elegant, but also far more work, as they provide a complete new architecture. Instead of having an install folder, you can just use yum to install instead of rpm when you (right)click on an rpm to install. -------------- next part -------------- An HTML attachment was scrubbed... URL: From byte at aeon.com.my Mon May 23 12:34:28 2005 From: byte at aeon.com.my (Colin Charles) Date: Mon, 23 May 2005 22:34:28 +1000 Subject: talks about Fedora In-Reply-To: <200505231917.58860.russell@coker.com.au> References: <200505231917.58860.russell@coker.com.au> Message-ID: <1116851668.4564.794.camel@arena.soho.bytebot.net> On Mon, 2005-05-23 at 19:17 +1000, Russell Coker wrote: > I want to give a talk about the new features in FC4, the aim is for a 30 min > talk on the day of the release. Anyone working on notes for such a talk? If > not does anyone have suggestions about what new features should be mentioned? I am working on notes for such a talk; in fact, I'd probably have slides that are available in due time > Also I am considering offering a talk on Fedora on PPC, however not knowing > much about PPC I would need to have the notes mostly written for me. Is > there anyone who has such notes? The talk would have to go for 90 mins. Consider it done, if this is for LUV[1], we can take this discussion offline. [1] - yes, russell and I share a similar linux user group ;-) -- Colin Charles, http://www.bytebot.net/ From alan at redhat.com Mon May 23 12:38:21 2005 From: alan at redhat.com (Alan Cox) Date: Mon, 23 May 2005 08:38:21 -0400 Subject: What happened to pup? In-Reply-To: <20050522222404.GA29620@orient.maison.moi> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <4290CFFD.6030205@nc.rr.com> <20050522222404.GA29620@orient.maison.moi> Message-ID: <20050523123821.GA7339@devserv.devel.redhat.com> On Mon, May 23, 2005 at 12:24:04AM +0200, Emmanuel Seyman wrote: > > And the LSB specification for compliant *.rpm packages is useless both > > theoretically and practically. > > Will this be fixed in future versions of LSB? I don't believe the LSB currently agrees with Jeff on the state of play. From n3npq at nc.rr.com Mon May 23 12:46:44 2005 From: n3npq at nc.rr.com (Jeff Johnson) Date: Mon, 23 May 2005 08:46:44 -0400 Subject: What happened to pup? In-Reply-To: <20050523123821.GA7339@devserv.devel.redhat.com> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <4290CFFD.6030205@nc.rr.com> <20050522222404.GA29620@orient.maison.moi> <20050523123821.GA7339@devserv.devel.redhat.com> Message-ID: <1D6D9B08-14AD-4C81-9802-1D982796E5A4@nc.rr.com> On May 23, 2005, at 8:38 AM, Alan Cox wrote: > On Mon, May 23, 2005 at 12:24:04AM +0200, Emmanuel Seyman wrote: > >>> And the LSB specification for compliant *.rpm packages is useless >>> both >>> theoretically and practically. >>> >> >> Will this be fixed in future versions of LSB? >> > > I don't believe the LSB currently agrees with Jeff on the state of > play. > Yep. LSB prohibits all dependencies save one in *.rpm packages and does not have a testable and objective meaning for Requires: lsb except Whatever LSB says or will say in the future. OTOH, the benefit of that is that *.tar and *.rpm become functionally equivalent when LSB compliant. In fact, that was one of the stated goals of the LSB packaging standard. 73 de Jeff From martin at faltesek.net Mon May 23 14:13:27 2005 From: martin at faltesek.net (Martin) Date: Mon, 23 May 2005 09:13:27 -0500 Subject: talks about Fedora In-Reply-To: <4291A133.50807@redhat.com> References: <200505231917.58860.russell@coker.com.au> <4291A133.50807@redhat.com> Message-ID: <1116857607.5239.11.camel@desktop> On Mon, 2005-05-23 at 14:54 +0530, Rahul Sundaram wrote: > You might also want to mention the boot optimisation work using > bootchart and improvements in readahead depending on the audience Isn't that a work in progress for the future? or did it integrate (last time I checked boot still took 2.5 minutes from power on). > > > > http://people.redhat.com/sundaram/fedora_notes.html You might want to make it more clear what apps you refer to are new, and which were present in FC3 but updated significantly. Marty From sundaram at redhat.com Mon May 23 14:16:31 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Mon, 23 May 2005 19:46:31 +0530 Subject: talks about Fedora In-Reply-To: <1116857607.5239.11.camel@desktop> References: <200505231917.58860.russell@coker.com.au> <4291A133.50807@redhat.com> <1116857607.5239.11.camel@desktop> Message-ID: <4291E5BF.8000802@redhat.com> Hi >http://people.redhat.com/sundaram/fedora_notes.html > > > >You might want to make it more clear what apps you refer to are new, >and which were present in FC3 but updated significantly. > > > You would have to read the classical release notes linked from the document to get that information. regards Rahul From kyrre at solution-forge.net Mon May 23 14:43:54 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Mon, 23 May 2005 16:43:54 +0200 Subject: User Friendly Install In-Reply-To: <3adc772105052304502a8c57fc@mail.gmail.com> References: <20050523092412.367AA733AE@hormel.redhat.com> <3adc772105052304502a8c57fc@mail.gmail.com> Message-ID: <1116859434.3356.14.camel@localhost.localdomain> man, 23.05.2005 kl. 13.50 skrev Naheem Zaffar: > From: Matthew Miller > Subject: Re: User Friendly Install > To: Development discussions related to Fedora Core > > Message-ID: <20050523032434.GA8134 at jadzia.bu.edu> > Content-Type: text/plain; charset=us-ascii > > On Mon, May 23, 2005 at 01:59:50AM +0100, Naheem Zaffar wrote: > >> I do not know if Nautilus can have scriptable folders, or yum > install from > >> a local folder, so these could be issues that need to be worked on. > > >Recent versions of yum can indeed install from local files (and solve > >dependencies using any enabled repositories), so something like what > >you'reproposing might be workable. I'm not convinced it's the best > way, but it > >wouldn't be terribly hard. > > Its not the most elegant solution, but it solves the problem at hand > without forcing package providers to provide a complete web > architecture. > > The problem is allowing people to install packages without going into > 'DLL Hell'. The other solutions are far more elegant, but also far > more work, as they provide a complete new architecture. > > Instead of having an install folder, you can just use yum to install > instead of rpm when you (right)click on an rpm to install. > > > ______________________________________________________________________ That is also a workable solution, but it doesn't solve two problems, which frontend reading from a special metadata file doesn't solve: - Autmatic updates - Wich of the 10 rpm's openoffice is built up of do i need? I do also not see the the great security benefit of having such a folder instead of a metadata file - with the metadata file, you download this file, open it, type in the root password, and ansver to a couple of standard questions. Install. Done. Now we only need a solution to search and delete old software that is no longer needed :) Kyrre From jkeating at j2solutions.net Mon May 23 15:18:56 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Mon, 23 May 2005 08:18:56 -0700 Subject: What happened to pup? In-Reply-To: <1D6D9B08-14AD-4C81-9802-1D982796E5A4@nc.rr.com> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <4290CFFD.6030205@nc.rr.com> <20050522222404.GA29620@orient.maison.moi> <20050523123821.GA7339@devserv.devel.redhat.com> <1D6D9B08-14AD-4C81-9802-1D982796E5A4@nc.rr.com> Message-ID: <1116861536.2962.3.camel@yoda.loki.me> On Mon, 2005-05-23 at 08:46 -0400, Jeff Johnson wrote: > OTOH, the benefit of that is that *.tar and *.rpm become > functionally > equivalent > when LSB compliant. In fact, that was one of the stated goals of the > LSB packaging standard. > Gee that sounds helpful -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From ronny-vlug at vlugnet.org Mon May 23 15:42:50 2005 From: ronny-vlug at vlugnet.org (Ronny Buchmann) Date: Mon, 23 May 2005 17:42:50 +0200 Subject: What happened to pup? In-Reply-To: <4290F567.8010101@laposte.net> References: <1116619871.4491.1.camel@localhost.localdomain> <1116795160.4322.77.camel@localhost.localdomain> <4290F567.8010101@laposte.net> Message-ID: <200505231742.50703.ronny-vlug@vlugnet.org> On Sunday 22 May 2005 23:11, Nicolas Mailhot wrote: > (not that the usual closed vendor won't abuse to death rpm scriptlets to > hose the system anyway) > > As others noted, the problem usually is not the installation but the > fact no clean package exists at all. That's true unfortunately. I usually end up repackaging all the non-FOSS stuff internally to have a sane way of upgrading. (IBM TSM client for example is such a biest) -- http://LinuxWiki.org/RonnyBuchmann From jspaleta at gmail.com Mon May 23 15:55:07 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Mon, 23 May 2005 11:55:07 -0400 Subject: What happened to pup? In-Reply-To: <1116796750.4322.86.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> <604aa7910505221259714d1d68@mail.gmail.com> <1116796750.4322.86.camel@localhost.localdomain> Message-ID: <604aa791050523085572649ca2@mail.gmail.com> On 5/22/05, Kyrre Ness Sjobak wrote: > That is the job of a well-written web portal. web portals are nice.. but surely a different website portal per repository is an inherently redudant and complex way of accessing information that should be available on the client computer via metadata. I don't think relying on web portals is the best solution long term. A client side application that reads metadata either from repositories on the network or repositories defined locally is the best primary way to 'find' packages. Unless of course, you are willing to jettison any hope of building a solution that allows people to use repositories on local media and will require everyone to have network access. Fire up an install application, that allows for some ability to browse the entire configured package catelog available on that client. > Auch. That i did not think about. But surely, some solution could be > found. Isn't license review as part of 'finding' a package good enough? I don't think license review is worth imposing on people as part of the install step for the vast majority of software. For the small percentage of proprietary software that exists for linux... provide a post package install license review mechanism... if those restrictive propretary eula's require a click-through before using the software. > But more intuitive. Almost every user "discovers" up2date immediatly, > and try to use that. They do not discover yum before someone points them > to it. And I say build something to replace up2date that is a client side app instead of relying on browsers and multiple website trolling everytime you want to find a package. If up2date is a poor implementation of an intuitive approach..then i say we re-implement that client side application approach.. instead of moving to a website portal download approach. > Also a good idea - but what about libs, such as gstreamer-mp3? BTW. > doesn't Ubuntu use an interface similar to this one? Plugins and kernel modules... are more difficult.. and go right back to the subtle but more difficult question of how users are expected to discover or find applications at all. For items that don't fit intuitively as menu items, you can extend the menu metaphor to include a "Search for Software" dialog. So for example, users who are looking for mp3 support would fire up the "Search For Software" dialog from the menu layout and search for mp3. The resulting list of packages and short descriptions should be enough for users to use to install the plugin package they want. Take a good look at the UI for the gnome "Search for Files" dialog from the gnome menu. I think a lot of the UI from that dialog could be mapped over to software installation functionality. Repos instead of folders, and things like Tag "whatever" contains the text "this text" with a pull down menu for package tags, so you can do very refined searches by adding multiple tag related alpha-numeric searches. This should be as intuitive to any user who is use to using the gnome menu as a primary way of interacting with their system. Can it be more intuitive than that? -jef From mike at navi.cx Mon May 23 17:47:54 2005 From: mike at navi.cx (Mike Hearn) Date: Mon, 23 May 2005 18:47:54 +0100 Subject: User Friendly Install References: <20050523092412.367AA733AE@hormel.redhat.com> <3adc772105052304502a8c57fc@mail.gmail.com> <1116859434.3356.14.camel@localhost.localdomain> Message-ID: On Mon, 23 May 2005 16:43:54 +0200, Kyrre Ness Sjobak wrote: > Now we only need a solution to search and delete old software that is no > longer needed :) In the autopackage UI document I linked to a garbage collection scheme for packages is described. It might interest you. thanks -mike From commonlaw at vtlink.net Mon May 23 18:22:22 2005 From: commonlaw at vtlink.net (James Sylvester) Date: Mon, 23 May 2005 14:22:22 -0400 (EDT) Subject: What happened to pup? Message-ID: <41893.64.30.50.185.1116872542.squirrel@mail.vtlink.net> Yesterday, at about this time, I flawlessly installed 5 CDs containining a week old distro of Fedora Core 3.9, aka Fedora Core 4 test 3. Having installed Red Hat from 7.1 to 9 and Fedora 1, 2, 3, I thought the install was flawless until I tried to use Redhat Package Manager, up2date and then yum to update the many changes that occur in development in a week and a half's time. RPM up2date hung on the second panel, yum could get to extras but not to the development rpm repo at redhat or its mirrors. I being a user of some experience and a supporter of Fedora could see a potential enterprise user or newbie becoming real distressed. If Fedora Core 4 is released without better update resource managers, then we will not obtain the kudos that I believe F4 deserves. After 2 hours of frustration I began downloading single pkg updates, each time muttering, where's pup? Today, in a more refreshed mood, I looked at my dog Spot, who gleefully chases snakes, but had been unable to help, yesterday. The problem? A missing python library effecting both up2date and yum. Spot's to old to obtain a pup, but I am not. The need for a reliable "pup" may not become a part of Fedora 4 (due to devlp. time), but the need is sufficient enough to not have it relegated to extras. Many good ideas on the forum. Need core developer leadership on this one. Cheers to what I believe may be the best Fedora distro, yet. James Sylvester "There is one sense--and a very important one--in which it is determined quite independently of scientific laws, namely the sense that it will be what it will be" Bertrand Russell, Presidential address to the Aristotelian Society. (regarding causation) From nicolas.mailhot at laposte.net Mon May 23 19:19:09 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Mon, 23 May 2005 21:19:09 +0200 Subject: What happened to pup? In-Reply-To: <41893.64.30.50.185.1116872542.squirrel@mail.vtlink.net> References: <41893.64.30.50.185.1116872542.squirrel@mail.vtlink.net> Message-ID: <42922CAD.7080907@laposte.net> James Sylvester a ?crit : > Today, in a more refreshed mood, I looked at my dog Spot, who > gleefully chases snakes, but had been unable to help, yesterday. > The problem? A missing python library effecting both up2date and > yum. Spot's to old to obtain a pup, but I am not. So how would a pup have helped you ? It would have been as broken as up2date and yum by the missing lib. Something like pup would clearly help some people, but even if it's implemented perfectly it won't fix every unrelated problem there is under the sun. -- Nicolas Mailhot -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 251 bytes Desc: OpenPGP digital signature URL: From janina at rednote.net Mon May 23 19:30:26 2005 From: janina at rednote.net (Janina Sajka) Date: Mon, 23 May 2005 15:30:26 -0400 Subject: Bug 124248 Fixed--Can we have in FC4? Message-ID: <20050523193026.GA12493@rednote.net> Lee Bcc: Subject: Re: Raising the bar for FC4 Reply-To: In-Reply-To: X-Operating-System: Linux concerto.rednote.net 2.6.11-1.14_FC3spksmp Organization: Capital Accessibility LLC (http://www.CapitalAccessibility.com) X-PGP-Key: http://www.CapitalAccessibility.com/JaninaSajka_gpg_key.html I understand the telnet bug fix is now in CVS, so can this fix please make FC4? https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=124248 Those of us who really need telnet install, REALLY need it. For us it's critical. I may be mistaken, but I believe it does not impact anyone who isn't actually needing it. Thanks. Thanks also to Jeremy Katz for getting the fix in, and to Jim Cornette for his help reporting the problem in the first instance. Elliot Lee writes: > Hi all, > > Fedora Core 4 public release is currently targeted for Monday, June 6. To > hit that, we need some time in advance doing QA and final polishing. For > test releases that is usually about a week, but this is the final FC4 > release so we need a bit of extra time to do the final steps. > > This means that Monday, May 23 is the final final freeze for FC4. After > that, the bar for accepting changes into FC4 is even higher. Fixes will > need to be for showstopper bugs (data corruption, crashing programs, and > other things that impact a large percentage of users in a major way). I've > sent out reminder e-mails to people who currently own FC4Target and > FC4Blocker bugs in bugzilla, so you should know if you have specific bugs > to address. > > If you don't have any bugs to address, you can still help by doing test > installs of rawhide. I'll also try to get a few intermediate test trees > out for people to install. > > If the plan needs clarification, please let me know! > -- Elliot > > -- > fedora-test-list mailing list > fedora-test-list at redhat.com > To unsubscribe: > http://www.redhat.com/mailman/listinfo/fedora-test-list -- Chair, Accessibility Workgroup Free Standards Group (FSG) janina at freestandards.org http://a11y.org Janina Sajka Phone: +1.202.494.7040 Partner, Capital Accessibility LLC http://www.CapitalAccessibility.Com Bringing the Owasys 22C screenless cell phone to the U.S. and Canada. Go to http://www.ScreenlessPhone.Com to learn more. From janina at rednote.net Mon May 23 19:31:08 2005 From: janina at rednote.net (Janina Sajka) Date: Mon, 23 May 2005 15:31:08 -0400 Subject: Bug 124248 Fixed--Can we have in FC4? Message-ID: <20050523193026.GA12493@rednote.net> Lee Bcc: Subject: Re: Raising the bar for FC4 Reply-To: In-Reply-To: X-Operating-System: Linux concerto.rednote.net 2.6.11-1.14_FC3spksmp Organization: Capital Accessibility LLC (http://www.CapitalAccessibility.com) X-PGP-Key: http://www.CapitalAccessibility.com/JaninaSajka_gpg_key.html I understand the telnet bug fix is now in CVS, so can this fix please make FC4? https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=124248 Those of us who really need telnet install, REALLY need it. For us it's critical. I may be mistaken, but I believe it does not impact anyone who isn't actually needing it. Thanks. Thanks also to Jeremy Katz for getting the fix in, and to Jim Cornette for his help reporting the problem in the first instance. Elliot Lee writes: > Hi all, > > Fedora Core 4 public release is currently targeted for Monday, June 6. To > hit that, we need some time in advance doing QA and final polishing. For > test releases that is usually about a week, but this is the final FC4 > release so we need a bit of extra time to do the final steps. > > This means that Monday, May 23 is the final final freeze for FC4. After > that, the bar for accepting changes into FC4 is even higher. Fixes will > need to be for showstopper bugs (data corruption, crashing programs, and > other things that impact a large percentage of users in a major way). I've > sent out reminder e-mails to people who currently own FC4Target and > FC4Blocker bugs in bugzilla, so you should know if you have specific bugs > to address. > > If you don't have any bugs to address, you can still help by doing test > installs of rawhide. I'll also try to get a few intermediate test trees > out for people to install. > > If the plan needs clarification, please let me know! > -- Elliot > > -- > fedora-test-list mailing list > fedora-test-list at redhat.com > To unsubscribe: > http://www.redhat.com/mailman/listinfo/fedora-test-list -- Chair, Accessibility Workgroup Free Standards Group (FSG) janina at freestandards.org http://a11y.org Janina Sajka Phone: +1.202.494.7040 Partner, Capital Accessibility LLC http://www.CapitalAccessibility.Com Bringing the Owasys 22C screenless cell phone to the U.S. and Canada. Go to http://www.ScreenlessPhone.Com to learn more. From katzj at redhat.com Mon May 23 19:31:51 2005 From: katzj at redhat.com (Jeremy Katz) Date: Mon, 23 May 2005 15:31:51 -0400 Subject: Bug 124248 Fixed--Can we have in FC4? In-Reply-To: <20050523193026.GA12493@rednote.net> References: <20050523193026.GA12493@rednote.net> Message-ID: <1116876711.4292.28.camel@bree.local.net> On Mon, 2005-05-23 at 15:30 -0400, Janina Sajka wrote: > I understand the telnet bug fix is now in CVS, so can this fix please > make FC4? It should make it. anaconda has a few other changes which need to go in and since I haven't branched yet, this will get in as well Jeremy From paul at all-the-johnsons.co.uk Mon May 23 20:17:32 2005 From: paul at all-the-johnsons.co.uk (Paul) Date: Mon, 23 May 2005 21:17:32 +0100 Subject: Update / Install CDs Message-ID: <1116879453.28095.8.camel@localhost> Hi, Can we please have an option when updating (or installing fresh) not to install every language pack under the sun for OOo? I've just done an update and have spent the last 10 minutes or so ditching the lot. I have no problems including the $LANG language pack, but why in the UK would I want the rest?! TTFN Paul -- "Space", it says, "is big. Really big. You just won't believe how vastly, hugely, mind-bogglingly big space really is. I mean, you may think it's a long way down the road to the chemists, but that's just *peanuts* compared to space, listen" - Hitch Hikers Guide to the Galaxy -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From rodd at clarkson.id.au Mon May 23 23:15:03 2005 From: rodd at clarkson.id.au (Rodd Clarkson) Date: Tue, 24 May 2005 09:15:03 +1000 Subject: Update / Install CDs In-Reply-To: <1116879453.28095.8.camel@localhost> References: <1116879453.28095.8.camel@localhost> Message-ID: <1116890103.3405.5.camel@goose> On Mon, 2005-05-23 at 21:17 +0100, Paul wrote: > Can we please have an option when updating (or installing fresh) not to > install every language pack under the sun for OOo? I've just done an > update and have spent the last 10 minutes or so ditching the lot. I have > no problems including the $LANG language pack, but why in the UK would I > want the rest?! Paul, Bugzilla this, but it may not happen. I'm assuming that you've updated from FC3 to FC4test. OOo in FC3 included all the languages as part of a single rpm package. They've been split up in FC4's OOo, but the upgrade path isn't that clear given that you're going from all languages supported to ??? It might be that the installer could use languages selected during the upgrade to short list what OOo installs, but this wouldn't help for people upgrading using yum, or manually upgrading (neither is recommended, but it does happen a lot). I can assure you that installing a fresh copy of FC4test3 sees only the language packs you've selected installed, and not every language pack. Rodd -- "It's a fine line between denial and faith. It's much better on my side" From commonlaw at vtlink.net Mon May 23 23:42:07 2005 From: commonlaw at vtlink.net (James Sylvester) Date: Mon, 23 May 2005 19:42:07 -0400 (EDT) Subject: What happened to pup? In-Reply-To: <42922CAD.7080907@laposte.net> References: <41893.64.30.50.185.1116872542.squirrel@mail.vtlink.net> <42922CAD.7080907@laposte.net> Message-ID: <50966.64.30.33.80.1116891727.squirrel@mail.vtlink.net> > James Sylvester a ??crit : > >> Today, in a more refreshed mood, I looked at my dog Spot, who >> gleefully chases snakes, but had been unable to help, yesterday. The >> problem? A missing python library effecting both up2date and yum. >> Spot's to old to obtain a pup, but I am not. > > So how would a pup have helped you ? It would have been as broken as > up2date and yum by the missing lib. > > Something like pup would clearly help some people, but even if it's > implemented perfectly it won't fix every unrelated problem there is > under the sun. > > -- > Nicolas Mailhot Your point is well taken. However, the general focus of the discussions have focused upon the initial and continueing need for a workable solution to upgrading, maintaining, and modifying FC4. RHPM and yum are better than most other distros' methods, in my experience. Now we need to move forward and select a more integrated approach to the problem. One that is even more user friendly, without sacrificing reliability. The forum has a number of concrete suggestions. Winnowing may now be possible. James Sylvester From janina at rednote.net Tue May 24 01:29:31 2005 From: janina at rednote.net (Janina Sajka) Date: Mon, 23 May 2005 21:29:31 -0400 Subject: Bug 124248 Fixed--Can we have in FC4? In-Reply-To: <1116876711.4292.28.camel@bree.local.net> References: <20050523193026.GA12493@rednote.net> <1116876711.4292.28.camel@bree.local.net> Message-ID: <20050524012931.GC12493@rednote.net> Jeremy Katz writes: > On Mon, 2005-05-23 at 15:30 -0400, Janina Sajka wrote: > > I understand the telnet bug fix is now in CVS, so can this fix please > > make FC4? > > It should make it. anaconda has a few other changes which need to go in > and since I haven't branched yet, this will get in as well > Very cool. It will really help. Sorry for the messy post earlier today. I really shouldn't try to two three things at once. > Jeremy > > -- > fedora-devel-list mailing list > fedora-devel-list at redhat.com > http://www.redhat.com/mailman/listinfo/fedora-devel-list -- Chair, Accessibility Workgroup Free Standards Group (FSG) janina at freestandards.org http://a11y.org Janina Sajka Phone: +1.202.494.7040 Partner, Capital Accessibility LLC http://www.CapitalAccessibility.Com Bringing the Owasys 22C screenless cell phone to the U.S. and Canada. Go to http://www.ScreenlessPhone.Com to learn more. From buildsys at redhat.com Tue May 24 11:48:17 2005 From: buildsys at redhat.com (Build System) Date: Tue, 24 May 2005 07:48:17 -0400 Subject: rawhide report: 20050524 changes Message-ID: <200505241148.j4OBmHDC013267@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.7-20050519.153509.FC4.3 ----------------------------------------- * Mon May 23 2005 Chris Feist - Fixed depmod in %post scriptlet to use correct kernel. anaconda-10.2.1.0-1 ------------------- * Mon May 23 2005 Jeremy Katz - 10.2.1.0-1 - Branch for FC4, betanag is off - Hack around usb-storage slowness at finding devices that leads to the reload not occurring - Handle FC3 swap label format and convert to right format (#158195) - Only set things up to change the default kernel if we're booting us (#158195) - Fix deps on upgrades (#157754) - Try to keep install screen from moving with length of strings - Fix autopart problem leaving some freespace the size of where you started your partition growing - Allow excluding name.arch in kickstart (Dave Lehman, #158370) - Don't spew an error if essid or wepkey isn't set (#158223) - Add megaraid and other new drivers (#157420) - Left pad RAID uuid (clumens, #136051) - synaptics tweak (pnasrat) - Fix telnetd to use devpts instead of legacy ptys (#124248) ant-0:1.6.2-3jpp_8fc -------------------- * Mon May 23 2005 Gary Benson 0:1.6.2-3jpp_8fc - Use absolute paths for rebuild-gcj-db. * Mon May 23 2005 Gary Benson 0:1.6.2-3jpp_7fc - Build the javah task (#157750). - Add alpha to the list of build architectures (#157522). * Thu May 05 2005 Gary Benson 0:1.6.2-3jpp_6fc - Add dependencies for %post and %postun scriptlets (#156901). cman-kernel-2.6.11.4-20050517.141233.FC4.7 ------------------------------------------ * Mon May 23 2005 Chris Feist - Fixed depmod in %post scriptlet to use correct kernel. dlm-kernel-2.6.11.4-20050517.141233.FC4.7 ----------------------------------------- * Mon May 23 2005 Chris Feist - Fixed depmod in %post scriptlet to use correct kernel. filesystem-2.3.4-1 ------------------ * Mon May 23 2005 Bill Nottingham - 2.3.4-1 - ship /usr/share/games (#158433, ) * Thu May 05 2005 Peter Jones - 2.3.3-1 - remove /initrd, since mkinitrd doesn't use it anymore by default gnbd-kernel-2.6.11.2-20050420.133124.FC4.29 ------------------------------------------- * Mon May 23 2005 Chris Feist - Fixed depmod in %post scriptlet to use correct kernel. gnome-desktop-2.10.0-5 ---------------------- * Mon May 23 2005 Ray Strode - 2.10.0-5 - Let's try this gettext patch one more time--maybe, just maybe, I'll get it right this time (bug 155659). gulm-1.0-0.pre30.1 ------------------ hal-0.5.2-2 ----------- * Mon May 23 2005 David Zeuthen 0.5.2-2 - Fix doublefree when locking (#158474) - Never use the 'sync' mount option (#157674) - Update the fstab-sync man page (#158483) hotplug-3:2004_09_23-7 ---------------------- * Mon May 23 2005 Bill Nottingham 3:2004_09_23-7 - fix path to /sbin/ifrename (#158564) httpd-2.0.54-10 --------------- * Mon May 23 2005 Joe Orton 2.0.54-10 - remove broken symlink (Robert Scheck, #158404) iiimf-1:12.2-4 -------------- * Mon May 23 2005 Jens Petersen - 1:12.2-4 - really move gnome-im-properties.desktop to applications dir (#149226) - pass _target_platform instead of _host to update-gtk-immodules jakarta-commons-beanutils-0:1.7.0-1jpp_4fc ------------------------------------------ * Mon May 23 2005 Gary Benson - 0:1.7.0-1jpp_4fc - Add alpha to the list of build architectures (#157522). - Use absolute paths for rebuild-gcj-db. jakarta-commons-collections-0:3.1-1jpp_4fc ------------------------------------------ * Mon May 23 2005 Gary Benson - 0:3.1-1jpp_4fc - Add alpha to the list of build architectures (#157522). - Use absolute paths for rebuild-gcj-db. jakarta-commons-digester-0:1.6-2jpp_4fc --------------------------------------- * Mon May 23 2005 Gary Benson - 0:1.6-2jpp_4fc - Add alpha to the list of build architectures (#157522). - Use absolute paths for rebuild-gcj-db. jakarta-commons-el-0:1.0-2jpp_3fc --------------------------------- * Mon May 23 2005 Gary Benson - 0:1.0-2jpp_3fc - Add alpha to the list of build architectures (#157522). - Use absolute paths for rebuild-gcj-db. jakarta-commons-logging-0:1.0.4-2jpp_4fc ---------------------------------------- * Mon May 23 2005 Gary Benson - 0:1.0.4-2jpp_4fc - Add alpha to the list of build architectures (#157522). - Use absolute paths for rebuild-gcj-db. jakarta-commons-modeler-0:1.1-3jpp_4fc -------------------------------------- * Mon May 23 2005 Gary Benson - 0:1.1-3jpp_4fc - Add alpha to the list of build architectures (#157522). - Use absolute paths for rebuild-gcj-db. kernel-2.6.11-1.1341_FC4 ------------------------ * Sun May 22 2005 Dave Jones - 2.6.12-rc4-git6 MMC update, reiserfs fixes, AIO fix. libgal2-2:2.4.2-3 ----------------- * Mon May 23 2005 Elliot Lee - 2:2.4.2-3 - Remove static libraries to save space. libofx-0.7.0-3 -------------- * Mon May 23 2005 Bill Nottingham 0.7.0-3 - remove static libs xen-2-20050522 -------------- * Mon May 23 2005 Rik van Riel 2-20050522 - change default display method for VMX domains to SDL xpdf-1:3.00-19 -------------- * Mon May 23 2005 Than Ngo 3.00-19 - apply patch to fix texts in non-embedded cjk font disappear, (#158509) From kyrre at solution-forge.net Tue May 24 15:02:41 2005 From: kyrre at solution-forge.net (Kyrre Ness Sjobak) Date: Tue, 24 May 2005 17:02:41 +0200 Subject: What happened to pup? In-Reply-To: <604aa791050523085572649ca2@mail.gmail.com> References: <1116619871.4491.1.camel@localhost.localdomain> <1116625739.3467.9.camel@enki.eridu> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> <604aa7910505221259714d1d68@mail.gmail.com> <1116796750.4322.86.camel@localhost.localdomain> <604aa791050523085572649ca2@mail.gmail.com> Message-ID: <1116946960.3407.11.camel@localhost.localdomain> man, 23.05.2005 kl. 17.55 skrev Jeff Spaleta: > On 5/22/05, Kyrre Ness Sjobak wrote: > > That is the job of a well-written web portal. > > web portals are nice.. but surely a different website portal per > repository is an inherently redudant and complex way of accessing > information that should be available on the client computer via > metadata. I don't think relying on web portals is the best solution > long term. A client side application that reads metadata either from > repositories on the network or repositories defined locally is the > best primary way to 'find' packages. Unless of course, you are > willing to jettison any hope of building a solution that allows people > to use repositories on local media and will require everyone to have > network access. > Well, no - it does not require network access. Part of it was to make sure you could install an rpm from a local disk, and have a repo configured to update this sofware. About web portals - some migth want to use this functionality, such as happypenguin. They could now offer a link ("install this app"), which points to a .install file, which is handeled by this installer app. Same app would also install "naked" rpm's from nautilus, using yum to resolve deps. Hmm.. maybe what i am really thinking of could be "system-config-packages should use yum to resolve deps, instead of using rpm directly, when installing an rpm from nautilus"? The ".install" thing could then be a wrapper (a tar.gz perhaps), containing rpms and an xml file with metadata - so that a third party suplyer could avoid special scripts etc. to setup a repo in yum + be able to pick "i want this, this, but not that" without needing to more or less guess what is within foo-bar-component-baz.rpm - and wich order they need to be installed in (today, there are no option of specifying "rpm -ivh foo.rpm bar.rpm" without using the command line). > Fire up an install application, that allows for some ability to browse > the entire configured package catelog available on that client. > > > Auch. That i did not think about. But surely, some solution could be > > found. > > Isn't license review as part of 'finding' a package good enough? > I don't think license review is worth imposing on people as part of > the install step for the vast majority of software. For the small > percentage of proprietary software that exists for linux... provide a > post package install license review mechanism... if those restrictive > propretary eula's require a click-through before using the software. > Agreed. > > But more intuitive. Almost every user "discovers" up2date immediatly, > > and try to use that. They do not discover yum before someone points them > > to it. > > And I say build something to replace up2date that is a client side app > instead of relying on browsers and multiple website trolling everytime > you want to find a package. If up2date is a poor implementation of an > intuitive approach..then i say we re-implement that client side > application approach.. instead of moving to a website portal download > approach. > > > Also a good idea - but what about libs, such as gstreamer-mp3? BTW. > > doesn't Ubuntu use an interface similar to this one? > > Plugins and kernel modules... are more difficult.. and go right back > to the subtle but more difficult question of how users are expected to > discover or find applications at all. For items that don't fit > intuitively as menu items, you can extend the menu metaphor to include > a "Search for Software" dialog. So for example, users who are looking > for mp3 support would fire up the "Search For Software" dialog from > the menu layout and search for mp3. The resulting list of packages > and short descriptions should be enough for users to use to install > the plugin package they want. > Interesting, but that removes the "browse" aspect. There should be a description of what each package do as well. And the "yum groupinstall" thing should be closely mapped in the ui. > Take a good look at the UI for the gnome "Search for Files" dialog > from the gnome menu. I think a lot of the UI from that dialog could be > mapped over to software installation functionality. Repos instead of > folders, and things like Tag "whatever" contains the text "this text" > with a pull down menu for package tags, so you can do very refined > searches by adding multiple tag related alpha-numeric searches. > This should be as intuitive to any user who is use to using the gnome > menu as a primary way of interacting with their system. > Can it be more intuitive than that? > > -jef From seandarcy2 at gmail.com Tue May 24 15:12:28 2005 From: seandarcy2 at gmail.com (sean) Date: Tue, 24 May 2005 11:12:28 -0400 Subject: rawhide report: 20050524 changes In-Reply-To: <200505241148.j4OBmHDC013267@porkchop.devel.redhat.com> References: <200505241148.j4OBmHDC013267@porkchop.devel.redhat.com> Message-ID: yum upgrade --> Running transaction check --> Processing Dependency: libjawt.so.6 for package: openoffice.org-core --> Finished Dependency Resolution Error: Missing Dependency: libjawt.so.6 is needed by package openoffice.org-core I think it's because of: ---> Package libgcj.i386 0:4.0.0-8 set to be updated sean From byte at aeon.com.my Tue May 24 15:59:22 2005 From: byte at aeon.com.my (Colin Charles) Date: Wed, 25 May 2005 01:59:22 +1000 Subject: rawhide report: 20050524 changes In-Reply-To: References: <200505241148.j4OBmHDC013267@porkchop.devel.redhat.com> Message-ID: <1116950362.2773.72.camel@arena.soho.bytebot.net> On Tue, 2005-05-24 at 11:12 -0400, sean wrote: > Error: Missing Dependency: libjawt.so.6 is needed by package > openoffice.org-core > > > I think it's because of: > > ---> Package libgcj.i386 0:4.0.0-8 set to be updated yum --exclude \*4.0.0-8\* update -- Colin Charles, http://www.bytebot.net/ From jspaleta at gmail.com Tue May 24 16:05:42 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Tue, 24 May 2005 12:05:42 -0400 Subject: What happened to pup? In-Reply-To: <1116946960.3407.11.camel@localhost.localdomain> References: <1116619871.4491.1.camel@localhost.localdomain> <1116775894.4322.31.camel@localhost.localdomain> <4290ADD0.3000303@laposte.net> <1116779390.4322.41.camel@localhost.localdomain> <20050522170656.GA28937@orient.maison.moi> <1116789905.4322.60.camel@localhost.localdomain> <604aa7910505221259714d1d68@mail.gmail.com> <1116796750.4322.86.camel@localhost.localdomain> <604aa791050523085572649ca2@mail.gmail.com> <1116946960.3407.11.camel@localhost.localdomain> Message-ID: <604aa791050524090536e88df8@mail.gmail.com> On 5/24/05, Kyrre Ness Sjobak wrote: > Well, no - it does not require network access. Part of it was to make > sure you could install an rpm from a local disk, and have a repo > configured to update this sofware. But.. in your scheme web portals are primary way to FIND packages.. you said so yourself. > > About web portals - some migth want to use this functionality, such as > happypenguin. And I'm not saying do away with web portals.. im saying do not make them the primary way by which users of the system are expected to find software. Make the primary way a client application.. re-using as much of the existing desktop ui conventions as is feasible. > Hmm.. maybe what i am really thinking of could be > "system-config-packages should use yum to resolve deps, instead of using > rpm directly, when installing an rpm from nautilus"? And I'm pretty sure there is a long term master plan to make both anaconda and whatever replaces system-config-packages yum aware. > The ".install" > thing could then be a wrapper (a tar.gz perhaps), containing rpms and an > xml file with metadata - so that a third party suplyer could avoid > special scripts etc. to setup a repo in yum + be able to pick "i want > this, this, but not that" without needing to more or less guess what is > within foo-bar-component-baz.rpm - and wich order they need to be > installed in (today, there are no option of specifying "rpm -ivh foo.rpm > bar.rpm" without using the command line). This is overkill re-invention. repo structures exist and can be used by any 3rd party who wants to provide a collection of packages. You can even drop a repo structure into a tarball. The only piece missing is a mechanism for getting a repo definition added/removed to a configuration, everything else is something ui experts need to mull over. > Interesting, but that removes the "browse" aspect. Fine... shall we add a browse softwarespace to the find dialog as well? Look at the find dialog in the gnome menu... it has a browse button. The menu layout still holds as a ui to shoot for. The idea isn't to prevent people from browsing the full space of software, the point is to make sure there are intuitive mechanisms that make it less likely for users to need to do a full browse. browsing the full space is never..ever...ever going to be intuitive nor efficient. Just like using your filebrowser to hunt for random executables on your filesystem is something that can be done..and will be done...must be done.. in some cases, its not what you want novice users to be doing most of the time. For installed applications you want users to be able to quickly navigate a menu and then falling back to a usable find dialog, before resorting to a brute force browse. And I say, if the ui works for the main menu.. it will work for an software installation app. > There should be a description of what each package do as well. And the > "yum groupinstall" thing should be closely mapped in the ui. There is no technical reason why the visible comps.xml definitions can't be re-worked to closely align with the categorization of applications in the main menu. In fact i find the dissimilarity between Core's user visible comps group definitions and the main menu layout to be a glaring usability problem when it comes to post install usage by novice users. A yum enabled version of system-config-packages will still confuse novice users simply because the comps defined groupings are not aligned well with the application categorizations they use every day in the system menu. How to categorize packages, is a complicated issue. We might end up needing to provide different comps.xml files for different purposes or intended audiences. -jef From dcbw at redhat.com Tue May 24 16:13:57 2005 From: dcbw at redhat.com (Dan Williams) Date: Tue, 24 May 2005 12:13:57 -0400 Subject: rawhide report: 20050524 changes In-Reply-To: References: <200505241148.j4OBmHDC013267@porkchop.devel.redhat.com> Message-ID: <1116951237.1498.13.camel@dcbw.boston.redhat.com> On Tue, 2005-05-24 at 11:12 -0400, sean wrote: > yum upgrade > > --> Running transaction check > --> Processing Dependency: libjawt.so.6 for package: > openoffice.org-core > --> Finished Dependency Resolution > Error: Missing Dependency: libjawt.so.6 is needed by package > openoffice.org-core > > > I think it's because of: > > ---> Package libgcj.i386 0:4.0.0-8 set to be updated A new OOo finally got through the build system this morning to deal with this. It should show up tomorrow. Dan From antonio at apache.org Tue May 24 18:00:49 2005 From: antonio at apache.org (Antonio Gallardo) Date: Tue, 24 May 2005 13:00:49 -0500 (CDT) Subject: rawhide report: 20050524 changes In-Reply-To: <200505241148.j4OBmHDC013267@porkchop.devel.redhat.com> References: <200505241148.j4OBmHDC013267@porkchop.devel.redhat.com> Message-ID: <1671.10.0.0.2.1116957649.squirrel@ags01.agsoftware.dnsalias.com> On Mar, 24 de Mayo de 2005, 6:48, Build System dijo: > Updated Packages: > ant-0:1.6.2-3jpp_8fc > -------------------- > * Mon May 23 2005 Gary Benson 0:1.6.2-3jpp_8fc > - Use absolute paths for rebuild-gcj-db. > > * Mon May 23 2005 Gary Benson 0:1.6.2-3jpp_7fc > - Build the javah task (#157750). > - Add alpha to the list of build architectures (#157522). > > * Thu May 05 2005 Gary Benson 0:1.6.2-3jpp_6fc > - Add dependencies for %post and %postun scriptlets (#156901). > Ant few days ago released 1.6.4: http://ant.apache.org/ Best Regards, Antonio Gallardo From pjones at redhat.com Tue May 24 22:18:43 2005 From: pjones at redhat.com (Peter Jones) Date: Tue, 24 May 2005 18:18:43 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1116462767.3483.19.camel@jellyfish.redfishdemo.com> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> <1116462767.3483.19.camel@jellyfish.redfishdemo.com> Message-ID: <1116973123.7655.2.camel@localhost.localdomain> On Thu, 2005-05-19 at 10:32 +1000, Rodd Clarkson wrote: > 1. Each user should have there own ~/tmp space which only they can > access. This could be used for the users agent sockets, but also just > for general files that they would put in /tmp. This would give a better > level of privacy (for whatever reason, but maybe simply so they don't > have to think about the privacy implications of putting things in a > publically accessible /tmp folder) Of course, ~/tmp/ really sucks for this -- lots of places use nfs homedirs, and you'd rather not put the sort of stuff you use /tmp for on the network. -- Peter From mattdm at mattdm.org Tue May 24 22:24:45 2005 From: mattdm at mattdm.org (Matthew Miller) Date: Tue, 24 May 2005 18:24:45 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1116973123.7655.2.camel@localhost.localdomain> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> <1116462767.3483.19.camel@jellyfish.redfishdemo.com> <1116973123.7655.2.camel@localhost.localdomain> Message-ID: <20050524222445.GA23131@jadzia.bu.edu> On Tue, May 24, 2005 at 06:18:43PM -0400, Peter Jones wrote: > > 1. Each user should have there own ~/tmp space which only they can > > access. This could be used for the users agent sockets, but also just [...] > Of course, ~/tmp/ really sucks for this -- lots of places use nfs > homedirs, and you'd rather not put the sort of stuff you use /tmp for on > the network. Which goes back to my original post here, which cleverly (?) checks if ~/tmp is on a local filesystem and uses mktemp in /tmp otherwise. (PS: there's an obvious error in that first post, in that the variables should be actually exported.) -- Matthew Miller mattdm at mattdm.org Boston University Linux ------> Current office temperature: 73 degrees Fahrenheit. From buildsys at redhat.com Wed May 25 11:41:51 2005 From: buildsys at redhat.com (Build System) Date: Wed, 25 May 2005 07:41:51 -0400 Subject: rawhide report: 20050525 changes Message-ID: <200505251141.j4PBfpGQ023334@porkchop.devel.redhat.com> Updated Packages: alsa-lib-1.0.9rc4-1 ------------------- * Tue May 24 2005 Bill Nottingham 1.0.9rc4-1 - update to 1.0.9rc4 (#157180, #158547) boost-1.32.0-6 -------------- * Tue May 24 2005 Benjamin Kosnik 1.32.0-6 - (#153093: boost warns that gcc 4.0.0 is an unknown compiler) - (#152205: development .so symlinks should be in -devel subpackage) - (#154783: linker .so symlinks missing from boost-devel package) * Fri Mar 18 2005 Benjamin Kosnik 1.32.0-5 - Revert boost-base.patch to old behavior. - Use SONAMEVERSION instead of dllversion. * Wed Mar 16 2005 Benjamin Kosnik 1.32.0-4 - (#142612: Compiling Boost 1.32.0 Failed in RHEL 3.0 on Itanium2) - (#150069: libboost_python.so is missing) - (#141617: bad patch boost-base.patch) - (#122817: libboost_*.so symlinks missing) - Re-add boost-thread.patch. - Change boost-base.patch to show thread tags. - Change boost-gcc-tools.patch to use SOTAG, compile with dllversion. - Add symbolic links to files. - Sanity check can compile with gcc-3.3.x, gcc-3.4.2, gcc-4.0.x., gcc-4.1.x. control-center-1:2.10.1-6 ------------------------- * Mon May 23 2005 Bill Nottingham - 1:2.10.1-6 - don't ship static versions of nautilus extensions dbus-0.33-3 ----------- * Mon May 23 2005 Bill Nottingham - 0.33-3 - remove static libraries from python bindings eclipse-1:3.1.0_fc-0.M7.5 ------------------------- * Sun May 22 2005 Andrew Overholt 3.1.0_fc-0.M7.4 - Remove compilation of jdt.ui jar.so on ppc. * Sat May 21 2005 Ben Konrath 3.1.0_fc-0.M7.3 - Add ecj-options patch to bootstrap source. - Make embedded browser widget work (Robin Green). - Bump required version of java-gcj-compat to the latest (-40jpp_24rh). - Use -lgcjawt when building with gcj. * Wed May 18 2005 Ben Konrath 3.1.0_fc-0.M7.2 - Disable org.eclipse.osgi_3.1.0.jar.so. - Add ecj-options patch, remove ecj-extdirs patch. eclipse-cdt-1:3.0.0_fc-0.M6.7 ----------------------------- * Mon May 23 2005 Andrew Overholt 3.0.0_fc-0.M6.7 - Bring in new I-build to enable jump to Eclipse 3.1M7 and fix some critical issues. fedora-logos-1.1.31-1 --------------------- * Mon May 23 2005 Jeremy Katz - 1.1.31-1 - copyright date on anaconda splash (#153964) firstboot-1.3.41-4 ------------------ * Tue May 24 2005 Adrian Likins - 1.3.41 - fix #158095 - Subscription Alert (on first login) on non network installs * Thu May 19 2005 Adrian Likins - 1.3.40 - fix #154606 - First boot displays "please insert the red hat enterprise linux extras disk" ftp-0.17-26 ----------- * Tue May 24 2005 Miloslav Trmac - 0.17-26 - Fix passive mode with SELinux (#158234, patch by Nalin Dahyabhai) - Fix format string mismatch gamin-0.1.0-1 ------------- * Thu May 12 2005 Daniel Veillard 0.1.0-1 - Close inherited file descriptors on exec of gam_server - Cancelling a monitor send back a FAMAcknowledge - Fixed for big files > 2GB - Bug when monitoring a non existing directory - Make client side thread safe - Unreadable directory fixes - Better flow control handling - Updated to latest inotify version: 0.23-6 gdm-1:2.6.0.8-16 ---------------- * Mon May 23 2005 Ray Strode 1:2.6.0.8-16 - Make sure username/password incorrect message gets displayed (bug 158127). - reread system locale before starting gdm in early login mode (bug 158376). kernel-2.6.11-1.1353_FC4 ------------------------ * Tue May 24 2005 Dave Jones - Update various cpufreq drivers. * Mon May 23 2005 Dave Jones - Add extra id to SATA Sil driver. (#155748) - Fix oops on rmmod of lanai & ms558 drivers when no hardware present. * Mon May 23 2005 Dave Jones - Fix double unlock of spinlock on tulip. (#158522) libgnomeprint22-2.10.3-1 ------------------------ * Tue May 24 2005 David Zeuthen - 2.10.3-1 - Update to upstream release 2.10.3 (#158397). Temporarily disable the fix for #154939 since that may be fixed upstream. libxslt-1.1.14-2 ---------------- * Mon May 23 2005 Bill Nottingham - remove static library from python bindings log4j-0:1.2.8-7jpp_4fc ---------------------- * Mon May 23 2005 Gary Benson 0:1.2.8-7jpp_4fc - Work around chainsaw failure (#157585). lvm2-cluster-2.01.09-3.0 ------------------------ * Mon May 23 2005 Alasdair Kergon - 2.01.09-3.0 - Changed arch again, now: 'ExclusiveArch: i386 ppc ppc64 x86_64' - Fix clvmd init script. - Don't add it to any runlevels by default. (#157468) openoffice.org-1:1.9.104-2 -------------------------- * Tue May 24 2005 Dan Williams - 1:1.9.104-2 - Add -IN glyph fallbacks for Hindi, Gujarati, Bengali, Punjabi, and Tamil (penoffice.org-1.9.104.oooXXXXX.indic-font-fallbacks.patch) - Add Croatian 'hr' langpack - Package libportaudio and libsndfile libraries in -core package - drop workspace-db4.patch and workspace-db4-2.patch since it appears we can use db4 with gcc again - add openoffice.org-1.9.104-gcjawt.patch to deal with gcj AWT library change to libgcjawt.so - add openoffice.org-1.9.104-berkeleydb-jni-casting-misuse.patch to fix grievous misuse of C casting in berkeleydb's JNI glue rhpl-0.167-1 ------------ * Tue May 24 2005 Paul Nasrat - 0.167-1 - Fix for ALPS specific settings * Thu May 19 2005 Jeremy Katz - 0.166-1 - Retranslate keyboard names each time so that it works with anaconda's lang changing (#157733) rpm-4.4.1-21 ------------ * Tue May 24 2005 Paul Nasrat - 4.4.1-21 - Update translations (#154623) setools-2.1.0-5 --------------- * Mon May 23 2005 Bill Nottingham 2.1.0-5 - put libraries in the right place (also puts debuginfo in the right package) - add %defattr for -devel too * Thu May 12 2005 Dan Walsh 2.1.0-4 - Move sepcut to gui apps. * Fri May 06 2005 Dan Walsh 2.1.0-3 - Fix Missing return code. shadow-utils-2:4.0.7-9 ---------------------- * Mon May 23 2005 Peter Vrabec 2:4.0.7-9 - remove vigr binary * Mon May 23 2005 Peter Vrabec 2:4.0.7-8 - fix nscd socket path system-config-display-1.0.29-1 ------------------------------ * Mon May 23 2005 Jeremy Katz - 1.0.29-1 - put scriptlets from distcvs in this specfile * Mon May 23 2005 Jeremy Katz - 1.0.28-1 - fix typo in cleanup leading to traceback (#157423) * Thu Apr 28 2005 Soren Sandmann 1.0.26-1 - Clean up deprecation warnings (#153937) system-config-soundcard-1.2.11-5 -------------------------------- * Mon May 23 2005 Jeremy Katz - 1.2.11-5 - fix probing of mac audio (#157932) up2date-4.4.23-4 ---------------- * Tue May 24 2005 Bill Nottingham 4.4.23-4 - fix traceback in repomd support vte-0.11.13-2.fc4 ----------------- * Mon May 23 2005 BillNottingham 0.11.13-2.fc4 - fix removal of static libs from python bindings yum-2.3.2-6 ----------- * Tue May 24 2005 Paul Nasrat - 2.3.2-6 - Erase/remove reversing for yum cli (#158577) * Tue May 24 2005 Jeremy Katz - 2.3.2-5 - allow multiple packages _providing_ kernel-devel (or any installonlypkgs) to be installed (#155988) * Mon May 23 2005 Jeremy Katz - 2.3.2-4 - fix traceback on out of disk space error From ggw at wolves.durham.nc.us Wed May 25 22:22:49 2005 From: ggw at wolves.durham.nc.us (Gregory Woodbury) Date: Wed, 25 May 2005 18:22:49 -0400 Subject: What happened to the it821x IDE/PCI Raid driver? Message-ID: <20050525222249.GA23229@wolves.durham.nc.us> In the 2.6.10 series of kernels, Alan Cox' IT821x PCI/IDE-RAID driver was included in the kernel as a module. In 2.6.11 kernels, the module disappeared and the device is not recognized by the kernel. Researching some via Google and various kernel sites implied that the driver was rendered irrelevant by some changes to the main IDE driver, but the failure to be recognized seems to indicate that something has actually happened. I realize Alan is going on sabattical, but has anyone picked up the it821x driver? Or is there some magic kernal parameter incantation that will let FC3 and upward use this popular card? Thanks for the attention. -- G.Wolfe Woodbury `- -' RHCT U The Line Eater is a boojum! -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From davej at redhat.com Wed May 25 22:24:20 2005 From: davej at redhat.com (Dave Jones) Date: Wed, 25 May 2005 18:24:20 -0400 Subject: What happened to the it821x IDE/PCI Raid driver? In-Reply-To: <20050525222249.GA23229@wolves.durham.nc.us> References: <20050525222249.GA23229@wolves.durham.nc.us> Message-ID: <20050525222419.GA28147@redhat.com> On Wed, May 25, 2005 at 06:22:49PM -0400, Gregory Woodbury wrote: > In the 2.6.10 series of kernels, Alan Cox' IT821x PCI/IDE-RAID driver > was included in the kernel as a module. In 2.6.11 kernels, the module > disappeared and the device is not recognized by the kernel. it came back in yesterdays errata kernel. Dave From alan at redhat.com Wed May 25 22:26:16 2005 From: alan at redhat.com (Alan Cox) Date: Wed, 25 May 2005 18:26:16 -0400 Subject: What happened to the it821x IDE/PCI Raid driver? In-Reply-To: <20050525222249.GA23229@wolves.durham.nc.us> References: <20050525222249.GA23229@wolves.durham.nc.us> Message-ID: <20050525222616.GA7050@devserv.devel.redhat.com> On Wed, May 25, 2005 at 06:22:49PM -0400, Gregory Woodbury wrote: > In the 2.6.10 series of kernels, Alan Cox' IT821x PCI/IDE-RAID driver > was included in the kernel as a module. In 2.6.11 kernels, the module > disappeared and the device is not recognized by the kernel. Its only in the -ac kernels. Fedora includes the -ac IDE code on the whole but it was missing for most of RHEL4 test because it needed porting to 2.6.12rc and Dave had a lot more things on his plate that were far more urgent. From kewley at gps.caltech.edu Wed May 25 22:38:22 2005 From: kewley at gps.caltech.edu (David Kewley) Date: Wed, 25 May 2005 15:38:22 -0700 Subject: ufs write safety Message-ID: <200505251538.22819.kewley@gps.caltech.edu> Hi all, Can anyone here provide experience reports or pointers to reports, regarding the safety of Linux's write support for Sun UFS? It is marked as experimental and dangerous, but it appears to have been around since 1998. I'm wondering whether this might be one of those things that works fine but still carries old labels. I've googled and found only the same warnings that I would write myself, if I only read fs/Kconfig & Documentation/filesystems/ufs.txt. I tried contacting the author, Daniel Pirkl, but his email address listed in the code the docs is now invalid. I have several external RAID arrays that up to now have been attached to Sun boxes and formatted with UFS filesystems. I will be attaching these arrays now to a RHEL 4 host. If I can safely get away with it, I'd prefer to keep writing to the UFS volumes, since we're talking about several TB of data. But if writing is still not believed to be safe, then I will need to transfer the data to a Linux filesystem with spare TB of space, reformat the external filesystems, and move the data back (or variations on that plan). These are filesystems that several people use for daily work, so I'd rather avoid the copy time if possible. Any pointers would be very welcome. Thanks, David From davej at redhat.com Wed May 25 22:45:39 2005 From: davej at redhat.com (Dave Jones) Date: Wed, 25 May 2005 18:45:39 -0400 Subject: ufs write safety In-Reply-To: <200505251538.22819.kewley@gps.caltech.edu> References: <200505251538.22819.kewley@gps.caltech.edu> Message-ID: <20050525224539.GA13733@redhat.com> On Wed, May 25, 2005 at 03:38:22PM -0700, David Kewley wrote: > Can anyone here provide experience reports or pointers to reports, > regarding the safety of Linux's write support for Sun UFS? It is > marked as experimental and dangerous, but it appears to have been > around since 1998. I'm wondering whether this might be one of those > things that works fine but still carries old labels. It doesn't get any real attention upstream, so I've no particular belief that its gotten any more stable than it ever was (wrt writes). For this reason, the write support is disabled in Fedora. > I have several external RAID arrays that up to now have been attached to > Sun boxes and formatted with UFS filesystems. I will be attaching > these arrays now to a RHEL 4 host. UFS is also unavailable in RHEL4. > If I can safely get away with it, I'd prefer to keep writing to the UFS volumes > since we're talking about several TB of data. I certainly wouldn't trust it with data I wanted to keep. > then I will need to transfer the data to a Linux filesystem with > spare TB of space, reformat the external filesystems, and move the data > back (or variations on that plan). These are filesystems that several > people use for daily work, so I'd rather avoid the copy time if > possible. This is unfortunatly, probably the best way forward. Dave From kewley at gps.caltech.edu Wed May 25 22:59:16 2005 From: kewley at gps.caltech.edu (David Kewley) Date: Wed, 25 May 2005 15:59:16 -0700 Subject: ufs write safety In-Reply-To: <20050525224539.GA13733@redhat.com> References: <200505251538.22819.kewley@gps.caltech.edu> <20050525224539.GA13733@redhat.com> Message-ID: <200505251559.16826.kewley@gps.caltech.edu> On Wednesday 25 May 2005 15:45, Dave Jones wrote: > On Wed, May 25, 2005 at 03:38:22PM -0700, David Kewley wrote: > > Can anyone here provide experience reports or pointers to reports, > > regarding the safety of Linux's write support for Sun UFS? It is > > marked as experimental and dangerous, but it appears to have been > > around since 1998. I'm wondering whether this might be one of > > those things that works fine but still carries old labels. > > It doesn't get any real attention upstream, so I've no particular > belief that its gotten any more stable than it ever was (wrt writes). > For this reason, the write support is disabled in Fedora. > > > I have several external RAID arrays that up to now have been > > attached to Sun boxes and formatted with UFS filesystems. I will > > be attaching these arrays now to a RHEL 4 host. > > UFS is also unavailable in RHEL4. > > > If I can safely get away with it, I'd prefer to keep writing to > > the UFS volumes since we're talking about several TB of data. > > I certainly wouldn't trust it with data I wanted to keep. > > > then I will need to transfer the data to a Linux filesystem with > > spare TB of space, reformat the external filesystems, and move the > > data back (or variations on that plan). These are filesystems > > that several people use for daily work, so I'd rather avoid the > > copy time if possible. > > This is unfortunatly, probably the best way forward. Thanks *very* much, Dave -- it's good to hear this from someone highly involved in the kernel. I'll take your advice. I am using UFS and XFS in RHEL4 by rebuilding the kernel with those filesystems enabled. The filesystems appear to work fine; I know others are also using XFS in RHEL4. I've just now for the first time mounted that Sun UFS volume with this kernel. I was able to do the one thing I tried -- 'ls' the top-level directory. :) I had enabled write support in the kernel and mounted it rw; I'm now going back & re-building that kernel without write support, per your advice. I could mount ro, but that's not as safe as disabling writes. I know that using these filesystems on RHEL4 is neither supported nor recommended by Red Hat. As an academic site-license customer, however, I don't get RH support anyway, so I get to make my own choices appropriate to my risk tolerance. ;) David From davej at redhat.com Wed May 25 23:07:47 2005 From: davej at redhat.com (Dave Jones) Date: Wed, 25 May 2005 19:07:47 -0400 Subject: ufs write safety In-Reply-To: <200505251559.16826.kewley@gps.caltech.edu> References: <200505251538.22819.kewley@gps.caltech.edu> <20050525224539.GA13733@redhat.com> <200505251559.16826.kewley@gps.caltech.edu> Message-ID: <20050525230747.GB13733@redhat.com> On Wed, May 25, 2005 at 03:59:16PM -0700, David Kewley wrote: > Thanks *very* much, Dave -- it's good to hear this from someone highly > involved in the kernel. I'll take your advice. > > I am using UFS and XFS in RHEL4 by rebuilding the kernel with those > filesystems enabled. The filesystems appear to work fine; I know > others are also using XFS in RHEL4. beware: XFS can use *lots* of stack space in certain conditions, which really doesn't play too nicely with the 4KB stack size. Dave From kewley at gps.caltech.edu Wed May 25 23:13:50 2005 From: kewley at gps.caltech.edu (David Kewley) Date: Wed, 25 May 2005 16:13:50 -0700 Subject: ufs write safety In-Reply-To: <20050525230747.GB13733@redhat.com> References: <200505251538.22819.kewley@gps.caltech.edu> <200505251559.16826.kewley@gps.caltech.edu> <20050525230747.GB13733@redhat.com> Message-ID: <200505251613.50430.kewley@gps.caltech.edu> On Wednesday 25 May 2005 16:07, Dave Jones wrote: > On Wed, May 25, 2005 at 03:59:16PM -0700, David Kewley wrote: > > Thanks *very* much, Dave -- it's good to hear this from someone > > highly involved in the kernel. I'll take your advice. > > > > I am using UFS and XFS in RHEL4 by rebuilding the kernel with > > those filesystems enabled. The filesystems appear to work fine; I > > know others are also using XFS in RHEL4. > > beware: XFS can use *lots* of stack space in certain conditions, > which really doesn't play too nicely with the 4KB stack size. Thanks, and acknowledged. I looked into 4k vs 8k stacks before I started using XFS on RHEL4. My conclusion was that I'm safe because I have x86_64, which has 8k stacks. Is that reasonable, or are stack items twice as big on 64- as on 32-bit, so that you still run a risk with 8k stacks on 64-bit? David From wbeebe at gmail.com Wed May 25 23:27:42 2005 From: wbeebe at gmail.com (William Beebe) Date: Wed, 25 May 2005 19:27:42 -0400 Subject: Physical location of Posix message queues Message-ID: I'm porting an application from Solaris to Fedora Core. Everything is fine except for one issue regarding the use of Posix message queues. In order to solve this problem I'd like to know where the message queues are physically located. Where on Fedora Core (or Linux for that matter) are the message queues located? I've tried just looking on the file system and I've tried looking in the code for a Clue (both the kernel and glibc). If someone could just tell me, or if there are pointers to other documentation explaining some of the finer details of Posix message queue implementation on Linux I'd appreciate it. From davej at redhat.com Wed May 25 23:54:59 2005 From: davej at redhat.com (Dave Jones) Date: Wed, 25 May 2005 19:54:59 -0400 Subject: ufs write safety In-Reply-To: <200505251613.50430.kewley@gps.caltech.edu> References: <200505251538.22819.kewley@gps.caltech.edu> <200505251559.16826.kewley@gps.caltech.edu> <20050525230747.GB13733@redhat.com> <200505251613.50430.kewley@gps.caltech.edu> Message-ID: <20050525235459.GA548@redhat.com> On Wed, May 25, 2005 at 04:13:50PM -0700, David Kewley wrote: > On Wednesday 25 May 2005 16:07, Dave Jones wrote: > > On Wed, May 25, 2005 at 03:59:16PM -0700, David Kewley wrote: > > > Thanks *very* much, Dave -- it's good to hear this from someone > > > highly involved in the kernel. I'll take your advice. > > > > > > I am using UFS and XFS in RHEL4 by rebuilding the kernel with > > > those filesystems enabled. The filesystems appear to work fine; I > > > know others are also using XFS in RHEL4. > > > > beware: XFS can use *lots* of stack space in certain conditions, > > which really doesn't play too nicely with the 4KB stack size. > > Thanks, and acknowledged. I looked into 4k vs 8k stacks before I > started using XFS on RHEL4. My conclusion was that I'm safe because I > have x86_64, which has 8k stacks. Is that reasonable, or are stack > items twice as big on 64- as on 32-bit, so that you still run a risk > with 8k stacks on 64-bit? Correct, sizeof(long) and sizeof(pointer) are doubled, so you're still at risk of an overflow. Dave From kewley at gps.caltech.edu Thu May 26 05:42:54 2005 From: kewley at gps.caltech.edu (David Kewley) Date: Wed, 25 May 2005 22:42:54 -0700 Subject: ufs write safety In-Reply-To: <20050525235459.GA548@redhat.com> References: <200505251538.22819.kewley@gps.caltech.edu> <200505251613.50430.kewley@gps.caltech.edu> <20050525235459.GA548@redhat.com> Message-ID: <200505252242.54835.kewley@gps.caltech.edu> Dave Jones wrote on Wednesday 25 May 2005 16:54: > On Wed, May 25, 2005 at 04:13:50PM -0700, David Kewley wrote: > > On Wednesday 25 May 2005 16:07, Dave Jones wrote: > > > On Wed, May 25, 2005 at 03:59:16PM -0700, David Kewley wrote: > > > > Thanks *very* much, Dave -- it's good to hear this from someone > > > > highly involved in the kernel. I'll take your advice. > > > > > > > > I am using UFS and XFS in RHEL4 by rebuilding the kernel with > > > > those filesystems enabled. The filesystems appear to work fine; I > > > > know others are also using XFS in RHEL4. > > > > > > beware: XFS can use *lots* of stack space in certain conditions, > > > which really doesn't play too nicely with the 4KB stack size. > > > > Thanks, and acknowledged. I looked into 4k vs 8k stacks before I > > started using XFS on RHEL4. My conclusion was that I'm safe because I > > have x86_64, which has 8k stacks. Is that reasonable, or are stack > > items twice as big on 64- as on 32-bit, so that you still run a risk > > with 8k stacks on 64-bit? > > Correct, sizeof(long) and sizeof(pointer) are doubled, so you're still at > risk of an overflow. Thanks again, Dave. I've started a conversation on the linux XFS list, because I know this isn't RH's problem. David From mihamina.rakotomandimby at etu.univ-orleans.fr Thu May 26 08:20:58 2005 From: mihamina.rakotomandimby at etu.univ-orleans.fr (Rakotomandimby (R12y) Mihamina) Date: Thu, 26 May 2005 10:20:58 +0200 Subject: specfile, request for comments Message-ID: <1117095658.3813.40.camel@vavahady> Hi, Attached you will find a specfile about the ocaml-cryptokit package. I'm not very clever on specfile :-), so may be you would have some comments... It was initially from PLD http://ftp.nest.pld-linux.org/test/SRPMS/ocaml-cryptokit-1.2-1.src.rpm My doubts are about: - the calling of "make" ( the CFLAGS...) - the %{_docdir} was %{_examplesdir} but I had to change it because it seems not to begin with a "/", so it fails because of that. -- Get a fully managed dedicated server for ?200/month ($257/month) No time limit for taking care of your server. You keep the "root" acces if you want. Billing periods are 3 months. See the conditions at http://aspo.rktmb.org/activities/managed_servers -------------- next part -------------- # $Revision: 1.6 $, $Date: 2003/07/25 09:02:05 $ Summary: Cryptographic toolkit for OCaml Summary(pl): Biblioteka kryptograficzna dla OCamla Name: ocaml-cryptokit Version: 1.3 Release: fc3.1 License: LGPL w/ linking exceptions Group: Libraries Vendor: Xavier Leroy URL: http://pauillac.inria.fr/~xleroy/software.html Source0: cryptokit-%{version}.tar.gz # Source0-md5: 0249135953f10c1515e88985b45ee4c9 BuildRequires: zlib-devel BuildRequires: ocaml >= 3.08 BuildRoot: %{_tmppath}/%{name}-%{version}-root %description The Cryptokit library for Objective Caml provides a variety of cryptographic primitives that can be used to implement cryptographic protocols in security-sensitive applications. The primitives provided include: symmetric-key cryptography: AES, DES, Triple-DES, ARCfour, in ECB, CBC, CFB and OFB modes; public-key cryptography: RSA; hash functions and MACs: SHA-1, MD5, and MACs based on AES and DES; random number generation; encodings and compression: base 64, hexadecimal, Zlib compression. This package contains files needed to run bytecode executables using this library. %package devel Summary: Cryptographic toolkit for OCaml - development part Group: Development/Libraries Requires: %{name} = %{version}-%{release} %requires_eq ocaml %description devel The Cryptokit library for Objective Caml provides a variety of cryptographic primitives that can be used to implement cryptographic protocols in security-sensitive applications. The primitives provided include: symmetric-key cryptography: AES, DES, Triple-DES, ARCfour, in ECB, CBC, CFB and OFB modes; public-key cryptography: RSA; hash functions and MACs: SHA-1, MD5, and MACs based on AES and DES; random number generation; encodings and compression: base 64, hexadecimal, Zlib compression. This package contains files needed to develop OCaml programs using this library. %prep %setup -q -n cryptokit-%{version} %build make CFLAGS=" -fPIC" all allopt %install rm -rf $RPM_BUILD_ROOT install -d $RPM_BUILD_ROOT%{_libdir}/ocaml/cryptokit make install INSTALLDIR=$RPM_BUILD_ROOT%{_libdir}/ocaml/cryptokit install -d $RPM_BUILD_ROOT%{_libdir}/ocaml/cryptokit install *.cm[ixa]* *.a dll*.so $RPM_BUILD_ROOT%{_libdir}/ocaml/cryptokit (cd $RPM_BUILD_ROOT%{_libdir}/ocaml && ln -s cryptokit/dll*.so .) install -d $RPM_BUILD_ROOT%{_docdir}/%{name}-%{version} cp -r *test.ml $RPM_BUILD_ROOT%{_docdir}/%{name}-%{version} install -d $RPM_BUILD_ROOT%{_libdir}/ocaml/site-lib/cryptokit cat > $RPM_BUILD_ROOT%{_libdir}/ocaml/site-lib/cryptokit/META < 1.3.fc3 - First build for FC3 - Taken from the PLF specfile (wich was an 1.2) From fedora at wir-sind-cool.org Thu May 26 09:57:53 2005 From: fedora at wir-sind-cool.org (Michael Schwendt) Date: Thu, 26 May 2005 11:57:53 +0200 Subject: What happened to pup? In-Reply-To: <41893.64.30.50.185.1116872542.squirrel@mail.vtlink.net> References: <41893.64.30.50.185.1116872542.squirrel@mail.vtlink.net> Message-ID: <20050526115753.564e5747.fedora@wir-sind-cool.org> On Mon, 23 May 2005 14:22:22 -0400 (EDT), James Sylvester wrote: > > Yesterday, at about this time, I flawlessly installed 5 CDs > containining a week old distro of Fedora Core 3.9, aka Fedora Core 4 > test 3. Having installed Red Hat from 7.1 to 9 and Fedora 1, 2, 3, I > thought the install was flawless until I tried to use Redhat Package > Manager, up2date and then yum to update the many changes that occur > in development in a week and a half's time. > RPM up2date hung on the second panel, yum could get to extras but > not to the development rpm repo at redhat or its mirrors. This is quite vague. What problems did you face? Overloaded download server mirrors? Broken package dependencies because Fedora Core Development is known to break the repository temporarily? From buildsys at redhat.com Thu May 26 11:49:22 2005 From: buildsys at redhat.com (Build System) Date: Thu, 26 May 2005 07:49:22 -0400 Subject: rawhide report: 20050526 changes Message-ID: <200505261149.j4QBnMsN029953@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.7-20050519.153509.FC4.7 ----------------------------------------- anaconda-10.2.1.3-1 ------------------- * Wed May 25 2005 Paul Nasrat - 10.2.1.3-1 - Fix to not write boot flag to first part on pmac with 2 disks * Wed May 25 2005 Jeremy Katz - 10.2.1.2-1 - Fix display size of PVs when PVs previously existed (clumens, #158696) * Wed May 25 2005 Jeremy Katz - 10.2.1.1-1 - Blacklist rpmdb-fedora on upgrade (pnasrat, #158666) - Langsupport fixes (clumens, #154572, #158389) binutils-2.15.94.0.2.2-2 ------------------------ * Wed May 25 2005 Jakub Jelinek 2.15.94.0.2.2-2 - bfd and readelf robustification (CAN-2005-1704, #158680) cman-kernel-2.6.11.4-20050517.141233.FC4.9 ------------------------------------------ coreutils-5.2.1-48 ------------------ * Wed May 25 2005 Tim Waugh 5.2.1-48 - Prevent buffer overflow in who(1) (bug #158405). dlm-kernel-2.6.11.4-20050517.141233.FC4.9 ----------------------------------------- eclipse-1:3.1.0_fc-0.M7.7 ------------------------- * Wed May 25 2005 Andrew Overholt 3.1.0_fc-0.M7.7 - Fix ecj symlink in /usr/share/java (rh#158734). * Sun May 22 2005 Andrew Overholt 3.1.0_fc-0.M7.4 - Remove compilation of jdt.ui jar.so on ppc. * Sat May 21 2005 Ben Konrath 3.1.0_fc-0.M7.3 - Add ecj-options patch to bootstrap source. - Make embedded browser widget work (Robin Green). - Bump required version of java-gcj-compat to the latest (-40jpp_24rh). - Use -lgcjawt when building with gcj. ethereal-0.10.11-2 ------------------ * Wed May 25 2005 Radek Vokal 0.10.11-2 - rebuilt evince-0.3.1-1 -------------- * Sun May 22 2005 Marco Pesenti Gritti - 0.3.1-1 - Update to 0.3.1 * Sat May 07 2005 Marco Pesenti Gritti - 0.3.0-1 - Update to 0.3.0 firefox-0:1.0.4-4 ----------------- * Tue May 24 2005 Christopher Aillon 0:1.0.4-4 - Only install searchplugins for en-US, since there isn't any way to dynamically select searchplugins per locale yet. * Mon May 23 2005 Christopher Aillon 0:1.0.4-3 - Add support for locales: af-ZA, ast-ES, ca-AD, cs-CZ, cy-GB, da-DK, de-DE, el-GR, en-GB es-AR, es-ES, eu-ES, fi-FI, fr-FR, ga-IE, he-IL, hu-HU, it-IT, ko-KR, ja-JP, ja-JPM, mk-MK, nb-NO, nl-NL, pa-IN, pl-PL, pt-BR, pt-PT, ro-RO, ru-RU, sk-SK, sl-SI, sq-AL, sv-SE, tr-TR, zh-CN, zh-TW firstboot-1.3.42-1 ------------------ * Wed May 25 2005 Jeremy Katz - 1.3.42-1 - Stop using deprecated gtk.{TRUE,FALSE} (#153033) glibc-2.3.5-9 ------------- * Tue May 24 2005 Jakub Jelinek 2.3.5-9 - update from CVS - increase bindresvport's LOWPORT to 512, apparently some broken daemons don't think 0 .. 511 ports are reserved * Mon May 23 2005 Jakub Jelinek 2.3.5-8 - update from CVS - fix kernel version check in ld.so - fix sendfile{,64} prototypes (BZ#961) - try more ports in bindresvport if all 600..1023 are used, don't use priviledged ports when talking to portmap (#141773) gnbd-kernel-2.6.11.2-20050420.133124.FC4.31 ------------------------------------------- initscripts-8.11.1-1 -------------------- * Wed May 25 2005 Bill Nottingham 8.11.1-1 - fix inability to rename devices brought about by the fix for java-1.4.2-gcj-compat-0:1.4.2.0-40jpp_26rh ------------------------------------------ * Wed May 25 2005 Thomas Fitzsimmons - 0:1.4.2.0-40jpp_26rh - Import java-gcj-compat 1.0.30. * Wed May 25 2005 Gary Benson - 0:1.4.2.0-40jpp_25rh - Update tools.jar with the ecj's new jarfile name (#158734). kernel-2.6.11-1.1363_FC4 ------------------------ * Wed May 25 2005 Dave Jones - Disable TPM driver, it breaks 8139 driver. - Revert to previous version of ipw2x00 drivers. The newer ones sadly brought too many problems this close to the release. I'll look at updating them again for an update. - Update to 2.6.12rc5 Fix potential local DoS. 1-2 other small fixes. - Tweak to fix up some vdso arithmetic. - Disable sysenter again for now. * Wed May 25 2005 David Woodhouse - Turn off CONFIG_ISA on PPC again. It makes some Macs unhappy (#149200) - Make Speedtouch DSL modem resync automatically libgal2-2:2.4.2-4 ----------------- * Wed May 25 2005 David Malcolm - 2:2.4.2-4 - added Akira Tagoh's patch to ensure input strings are committed in the correct order in Evolution's Tasks widget (#157398) net-snmp-5.2.1-12 ----------------- * Wed May 18 2005 Radek Vokal - 5.2.1-12 - session free fixed, agentx modules build fine (#157851) - fixed dependency for net-snmp libs (#156932) * Wed May 04 2005 Radek Vokal - 5.2.1-11 - report gigabit Ethernet speeds using Ethtool (#152480) poppler-0.3.2-1 --------------- * Sun May 22 2005 Marco Pesenti gritti - 0.3.2-1 - Update to 0.3.2 * Sat May 07 2005 Marco Pesenti Gritti - 0.3.1 - Update to 0.3.1 system-config-httpd-5:1.3.2-2 ----------------------------- * Mon May 23 2005 Phil Knirsch 1.3.2-2 - Fixed the KeepAlive bug that was already fixed as errata for RHEL3 (#137815) - Fixed use of True and False instead of gtk.TRUE and gtk.FALSE (#153039) - Fixed traceback when no X-server is running - Fixed problem with ServerSignature setting (#156639) - Removed static /manual alias from httpd.conf template (#144239) - Updated gtk.mainloop() and gtk.mainquit() to new API names - Added pa translation file (#157826) * Tue Sep 28 2004 Phil Knirsch 1.3.1-1 - Added a bunch of new translations * Fri Sep 10 2004 Phil Knirsch 1.3.0-1 - Huge HIG 2 compliance changes by Vladimir Djokic system-config-lvm-0.9.30-1.0 ---------------------------- * Mon May 23 2005 Jim Parsons 0.9.30-1.0 - Fixed bz157744 * Tue May 17 2005 Jim Parsons 0.9.29-1.0 - Version Bump * Tue May 17 2005 Jim Parsons 0.9.28-1.0 - Version Bump yum-2.3.2-7 ----------- * Wed May 25 2005 Paul Nasrat - 2.3.2-7 - Drop erase reversal patch From ndbecker2 at gmail.com Thu May 26 13:16:29 2005 From: ndbecker2 at gmail.com (Neal Becker) Date: Thu, 26 May 2005 09:16:29 -0400 Subject: Still failing deps for /lib/modules/2.6.11-1.1355_FC4 References: <200505261149.j4QBnMsN029953@porkchop.devel.redhat.com> Message-ID: yum upgrade: --> Running transaction check --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: GFS-kernel Importing Additional filelist information for dependency resolution filelists.xml.gz 100% |=========================| 2.8 MB 00:15 developmen: ################################################## 3767/3767 Added 101 new packages, deleted 101 old in 7.36 seconds --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: dlm-kernel Importing Additional filelist information for dependency resolution --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: gnbd-kernel Importing Additional filelist information for dependency resolution --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: cman-kernel Importing Additional filelist information for dependency resolution --> Finished Dependency Resolution Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by package GFS-kernel Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by package dlm-kernel Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by package gnbd-kernel Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by package cman-kernel From thias at spam.spam.spam.spam.spam.spam.spam.egg.and.spam.freshrpms.net Thu May 26 13:58:55 2005 From: thias at spam.spam.spam.spam.spam.spam.spam.egg.and.spam.freshrpms.net (Matthias Saou) Date: Thu, 26 May 2005 15:58:55 +0200 Subject: PPC boot.iso broken today? Message-ID: <20050526155855.4a3f575b@python2> Hi, I just started a network install of FC Devel on a PowerMac, and it all went amazingly good, but I got missing packages because the mirror I was using had been updating while I went through the installer... so I had to grab the newer boot.iso, re-burn it to the same CD-RW, but now it stops after 3 "agpgart:" and 1 "Serial:" lines very early at boot time :-( I'll try downloading the latest FC test disc1 and install from there... Matthias -- Clean custom Red Hat Linux rpm packages : http://freshrpms.net/ Fedora Core release 3 (Heidelberg) - Linux kernel 2.6.11-1.27_FC3 Load : 0.31 0.40 0.45 From notting at redhat.com Thu May 26 14:45:00 2005 From: notting at redhat.com (Bill Nottingham) Date: Thu, 26 May 2005 10:45:00 -0400 Subject: Still failing deps for /lib/modules/2.6.11-1.1355_FC4 In-Reply-To: References: <200505261149.j4QBnMsN029953@porkchop.devel.redhat.com> Message-ID: <20050526144500.GA20384@nostromo.devel.redhat.com> The kernel beat GFS into the repository. It will get sorted out by tomorrow. Unless Dave builds another late kernel. Bill From paul at city-fan.org Thu May 26 14:48:10 2005 From: paul at city-fan.org (Paul Howarth) Date: Thu, 26 May 2005 15:48:10 +0100 Subject: specfile, request for comments In-Reply-To: <1117095658.3813.40.camel@vavahady> References: <1117095658.3813.40.camel@vavahady> Message-ID: <4295E1AA.2010506@city-fan.org> Rakotomandimby (R12y) Mihamina wrote: > Attached you will find a specfile about the ocaml-cryptokit package. > I'm not very clever on specfile :-), so may be you would have some > comments... It was initially from PLD > > http://ftp.nest.pld-linux.org/test/SRPMS/ocaml-cryptokit-1.2-1.src.rpm This should probably have been posted to fedora-extras-list (since ocaml is in Extras, not Core). In fact I'd suggest posting again there for more feedback. > My doubts are about: > > - the calling of "make" ( the CFLAGS...) I'd use: make CFLAGS="%{optflags} -fPIC" %{?_smp_mflags} all allopt > - the %{_docdir} was %{_examplesdir} but I had to change it because it > seems not to begin with a "/", so it fails because of that. I'd just stick the examples in an "examples" subdirectory and then include that in %doc Suggested changes: - Use Extras Dist Tag in Release: (http://fedoraproject.org/wiki/DistTag) - Dispense with Source0-md5 comment that refers to a previous release - Use a full URL for Source0 - remove .cvsignore file(s) presumably included upstream by mistake - specifying modes with %attr shouldn't be necessary if the upstream Makefiles get it right - split documentation between base and -devel packages - be more specific in %files selections, with less globbing I've attached an updated spec file taking these suggestions into account. Paul. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ocaml-cryptokit.spec URL: From pedro.lamarao at mndfck.org Thu May 26 15:08:14 2005 From: pedro.lamarao at mndfck.org (=?ISO-8859-1?Q?Pedro_Lamar=E3o?=) Date: Thu, 26 May 2005 12:08:14 -0300 Subject: Physical location of Posix message queues In-Reply-To: References: Message-ID: <4295E65E.1080204@mndfck.org> William Beebe wrote: > I'm porting an application from Solaris to Fedora Core. Everything is > fine except for one issue regarding the use of Posix message queues. > In order to solve this problem I'd like to know where the message > queues are physically located. Where on Fedora Core (or Linux for that > matter) are the message queues located? I've tried just looking on the > file system and I've tried looking in the code for a Clue (both the > kernel and glibc). If someone could just tell me, or if there are > pointers to other documentation explaining some of the finer details > of Posix message queue implementation on Linux I'd appreciate it. > Check /var/spool/postfix . -- Pedro Lamar?o "Merda faz as flores crescerem e isso ? bonito." From mharris at www.linux.org.uk Thu May 26 15:13:04 2005 From: mharris at www.linux.org.uk (Mike A. Harris) Date: Thu, 26 May 2005 11:13:04 -0400 Subject: specfile, request for comments In-Reply-To: <1117095658.3813.40.camel@vavahady> References: <1117095658.3813.40.camel@vavahady> Message-ID: <4295E780.2030900@www.linux.org.uk> Rakotomandimby (R12y) Mihamina wrote: > Hi, > > Attached you will find a specfile about the ocaml-cryptokit package. > I'm not very clever on specfile :-), so may be you would have some > comments... It was initially from PLD > > http://ftp.nest.pld-linux.org/test/SRPMS/ocaml-cryptokit-1.2-1.src.rpm > > My doubts are about: > > - the calling of "make" ( the CFLAGS...) > - the %{_docdir} was %{_examplesdir} but I had to change it because it > seems not to begin with a "/", so it fails because of that. > > > > ------------------------------------------------------------------------ > > # $Revision: 1.6 $, $Date: 2003/07/25 09:02:05 $ > Summary: Cryptographic toolkit for OCaml > Summary(pl): Biblioteka kryptograficzna dla OCamla > Name: ocaml-cryptokit > Version: 1.3 > Release: fc3.1 I would avoid using a release name with a letter like that. Use something safer like: 1.fc3.1 > License: LGPL w/ linking exceptions > Group: Libraries > Vendor: Xavier Leroy "Vendor" should *NEVER* be put in a spec file. It should be put in per user .rpmmacros rpm config file, or system wide on the build system that is being used for package generation. Ditto for "Distribution:" and "Packager" > URL: http://pauillac.inria.fr/~xleroy/software.html > Source0: cryptokit-%{version}.tar.gz > # Source0-md5: 0249135953f10c1515e88985b45ee4c9 > BuildRequires: zlib-devel > BuildRequires: ocaml >= 3.08 > BuildRoot: %{_tmppath}/%{name}-%{version}-root > > %description > The Cryptokit library for Objective Caml provides a variety of > cryptographic primitives that can be used to implement cryptographic > protocols in security-sensitive applications. The primitives provided > include: symmetric-key cryptography: AES, DES, Triple-DES, ARCfour, in > ECB, CBC, CFB and OFB modes; public-key cryptography: RSA; hash > functions and MACs: SHA-1, MD5, and MACs based on AES and DES; random > number generation; encodings and compression: base 64, hexadecimal, > Zlib compression. > This package contains files needed to run bytecode executables using > this library. > > %package devel > Summary: Cryptographic toolkit for OCaml - development part > Group: Development/Libraries > Requires: %{name} = %{version}-%{release} > %requires_eq ocaml > > %description devel > The Cryptokit library for Objective Caml provides a variety of > cryptographic primitives that can be used to implement cryptographic > protocols in security-sensitive applications. The primitives provided > include: symmetric-key cryptography: AES, DES, Triple-DES, ARCfour, in > ECB, CBC, CFB and OFB modes; public-key cryptography: RSA; hash > functions and MACs: SHA-1, MD5, and MACs based on AES and DES; random > number generation; encodings and compression: base 64, hexadecimal, > Zlib compression. > This package contains files needed to develop OCaml programs using > this library. > > %prep > %setup -q -n cryptokit-%{version} > > %build > make CFLAGS=" -fPIC" all allopt Should be: make CFLAGS="$RPM_OPT_FLAGS -fPIC" all allopt Should -fPIC be there for all architectures? Sounds wrong to me. > install -d $RPM_BUILD_ROOT%{_docdir}/%{name}-%{version} > cp -r *test.ml $RPM_BUILD_ROOT%{_docdir}/%{name}-%{version} This should probably be done instead using %doc in %files section. > %files > %defattr(644,root,root,755) > %dir %{_libdir}/ocaml/cryptokit > %attr(755,root,root) %{_libdir}/ocaml/cryptokit/* In general, it's cleaner to use "install -m755" to install files mode 755 than to override it in %files. > %{_libdir}/ocaml/* > > %files devel > %defattr(644,root,root,755) > %doc LICENSE README doc > %{_libdir}/ocaml/cryptokit/*.cm[ixa]* > %{_libdir}/ocaml/cryptokit/*.a > %{_docdir}/%{name}-%{version} This _docdir line is unnecessary. From pnasrat at redhat.com Thu May 26 15:34:50 2005 From: pnasrat at redhat.com (Paul Nasrat) Date: Thu, 26 May 2005 11:34:50 -0400 Subject: PPC boot.iso broken today? In-Reply-To: <20050526155855.4a3f575b@python2> References: <20050526155855.4a3f575b@python2> Message-ID: <1117121690.3225.6.camel@enki.eridu> On Thu, 2005-05-26 at 15:58 +0200, Matthias Saou wrote: > Hi, > > I just started a network install of FC Devel on a PowerMac, and it all > went amazingly good, but I got missing packages because the mirror I was > using had been updating while I went through the installer... so I had to > grab the newer boot.iso, re-burn it to the same CD-RW, but now it stops > after 3 "agpgart:" and 1 "Serial:" lines very early at boot time :-( > > I'll try downloading the latest FC test disc1 and install from there... > More details on your system would be useful. Can you try booting with linux maxcpus=1 noisapnp Paul From cfeist at redhat.com Thu May 26 15:50:43 2005 From: cfeist at redhat.com (Chris Feist) Date: Thu, 26 May 2005 10:50:43 -0500 Subject: Still failing deps for /lib/modules/2.6.11-1.1355_FC4 In-Reply-To: References: <200505261149.j4QBnMsN029953@porkchop.devel.redhat.com> Message-ID: <4295F053.70804@redhat.com> I have a script which checks for changes in the latest kernel in dist-fc4. Anytime it's updated I get an email and rebuild the {GFS,cman,dlm,gnbd}-kernel packages. I just finished a build this morning for 2.6.11-1.1366. Thanks, Chris Neal Becker wrote: > yum upgrade: > --> Running transaction check > --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: > GFS-kernel > Importing Additional filelist information for dependency resolution > filelists.xml.gz 100% |=========================| 2.8 MB 00:15 > developmen: ################################################## 3767/3767 > Added 101 new packages, deleted 101 old in 7.36 seconds > --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: > dlm-kernel > Importing Additional filelist information for dependency resolution > --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: > gnbd-kernel > Importing Additional filelist information for dependency resolution > --> Processing Dependency: /lib/modules/2.6.11-1.1355_FC4 for package: > cman-kernel > Importing Additional filelist information for dependency resolution > --> Finished Dependency Resolution > Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by > package GFS-kernel > Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by > package dlm-kernel > Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by > package gnbd-kernel > Error: Missing Dependency: /lib/modules/2.6.11-1.1355_FC4 is needed by > package cman-kernel > > From jspaleta at gmail.com Thu May 26 16:01:29 2005 From: jspaleta at gmail.com (Jeff Spaleta) Date: Thu, 26 May 2005 12:01:29 -0400 Subject: Still failing deps for /lib/modules/2.6.11-1.1355_FC4 In-Reply-To: <4295F053.70804@redhat.com> References: <200505261149.j4QBnMsN029953@porkchop.devel.redhat.com> <4295F053.70804@redhat.com> Message-ID: <604aa79105052609013fcda9a6@mail.gmail.com> On 5/26/05, Chris Feist wrote: > I have a script which checks for changes in the latest kernel in dist-fc4. > Anytime it's updated I get an email and rebuild the {GFS,cman,dlm,gnbd}-kernel > packages. I just finished a build this morning for 2.6.11-1.1366. Might I suggest implementing a script feature to look 12 hours into the future, so you can be notified of new kernels before they actually exist. Some sophisticated tracking and statistical analysis of dave's daily activities might provide a strongly corrolated precursor event to watch for. -jef"it's time to look into the future.. all the way to the year 2000"spaleta From arjanv at redhat.com Thu May 26 16:19:51 2005 From: arjanv at redhat.com (Arjan van de Ven) Date: Thu, 26 May 2005 18:19:51 +0200 Subject: Physical location of Posix message queues In-Reply-To: References: Message-ID: <1117124392.5998.18.camel@laptopd505.fenrus.org> On Wed, 2005-05-25 at 19:27 -0400, William Beebe wrote: > I'm porting an application from Solaris to Fedora Core. Everything is > fine except for one issue regarding the use of Posix message queues. > In order to solve this problem I'd like to know where the message > queues are physically located. what do you mean by "physically located" ??? -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From nmiell at comcast.net Thu May 26 17:29:35 2005 From: nmiell at comcast.net (Nicholas Miell) Date: Thu, 26 May 2005 10:29:35 -0700 Subject: Physical location of Posix message queues In-Reply-To: <1117124392.5998.18.camel@laptopd505.fenrus.org> References: <1117124392.5998.18.camel@laptopd505.fenrus.org> Message-ID: <1117128575.5002.5.camel@localhost.localdomain> On Thu, 2005-05-26 at 18:19 +0200, Arjan van de Ven wrote: > On Wed, 2005-05-25 at 19:27 -0400, William Beebe wrote: > > I'm porting an application from Solaris to Fedora Core. Everything is > > fine except for one issue regarding the use of Posix message queues. > > In order to solve this problem I'd like to know where the message > > queues are physically located. > > what do you mean by "physically located" ??? > On Solaris, POSIX message queues have a physical representation on the filesystem. Linux just uses a small virtual filesystem which typically isn't even mounted. Note that the use of the mq_* syscalls doesn't require the mqueue fs to be mounted. -- Nicholas Miell From mbneto at gmail.com Thu May 26 19:38:54 2005 From: mbneto at gmail.com (mbneto) Date: Thu, 26 May 2005 15:38:54 -0400 Subject: Oom killer in kernel rpms ? Message-ID: <5cf776b8050526123828bd6e7a@mail.gmail.com> Hi, I was wondering how can find out if a given feature is enabled in the kernels provided with the distro. For instance, the oom killer. regards, mb From davej at redhat.com Thu May 26 20:15:39 2005 From: davej at redhat.com (Dave Jones) Date: Thu, 26 May 2005 16:15:39 -0400 Subject: Oom killer in kernel rpms ? In-Reply-To: <5cf776b8050526123828bd6e7a@mail.gmail.com> References: <5cf776b8050526123828bd6e7a@mail.gmail.com> Message-ID: <20050526201539.GB5894@redhat.com> On Thu, May 26, 2005 at 03:38:54PM -0400, mbneto wrote: > Hi, > > I was wondering how can find out if a given feature is enabled in the > kernels provided with the distro. > > For instance, the oom killer. It isnt an option. Dave From fedora at camperquake.de Thu May 26 22:13:07 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Fri, 27 May 2005 00:13:07 +0200 Subject: Still failing deps for /lib/modules/2.6.11-1.1355_FC4 In-Reply-To: <604aa79105052609013fcda9a6@mail.gmail.com> References: <200505261149.j4QBnMsN029953@porkchop.devel.redhat.com> <4295F053.70804@redhat.com> <604aa79105052609013fcda9a6@mail.gmail.com> Message-ID: <20050527001307.186f1088@nausicaa.camperquake.de> Hi. Jeff Spaleta wrote: > Might I suggest implementing a script feature to look 12 hours into > the future, so you can be notified of new kernels before they actually > exist. Now this would be cool. Could we get a bug-aware version of that, so we can fix bugs before they exist? -- Angie chalks another Blue, mother smiles, she did it too... -- Marillion, "Garden Party" From mbneto at gmail.com Thu May 26 22:31:08 2005 From: mbneto at gmail.com (mbneto) Date: Thu, 26 May 2005 18:31:08 -0400 Subject: Oom killer in kernel rpms ? In-Reply-To: <20050526201539.GB5894@redhat.com> References: <5cf776b8050526123828bd6e7a@mail.gmail.com> <20050526201539.GB5894@redhat.com> Message-ID: <5cf776b80505261531406a07ea@mail.gmail.com> Hi, It is not an option and it is not available our it is not an option but is enabled/configured/patched by default ? On 5/26/05, Dave Jones wrote: > On Thu, May 26, 2005 at 03:38:54PM -0400, mbneto wrote: > > Hi, > > > > I was wondering how can find out if a given feature is enabled in the > > kernels provided with the distro. > > > > For instance, the oom killer. > > It isnt an option. > > Dave > From fedora at camperquake.de Thu May 26 22:34:25 2005 From: fedora at camperquake.de (Ralf Ertzinger) Date: Fri, 27 May 2005 00:34:25 +0200 Subject: Oom killer in kernel rpms ? In-Reply-To: <5cf776b80505261531406a07ea@mail.gmail.com> References: <5cf776b8050526123828bd6e7a@mail.gmail.com> <20050526201539.GB5894@redhat.com> <5cf776b80505261531406a07ea@mail.gmail.com> Message-ID: <20050527003425.03a635da@nausicaa.camperquake.de> Hi. mbneto wrote: > It is not an option and it is not available our it is not an option > but is enabled/configured/patched by default ? The latter. -- Cat's urine glows under a blacklight. From wbeebe at gmail.com Thu May 26 23:07:29 2005 From: wbeebe at gmail.com (William Beebe) Date: Thu, 26 May 2005 19:07:29 -0400 Subject: Physical location of Posix message queues In-Reply-To: <1117128575.5002.5.camel@localhost.localdomain> References: <1117124392.5998.18.camel@laptopd505.fenrus.org> <1117128575.5002.5.camel@localhost.localdomain> Message-ID: That's what I thought, but I wasn't sure. Thanks. On 5/26/05, Nicholas Miell wrote: > On Thu, 2005-05-26 at 18:19 +0200, Arjan van de Ven wrote: > > On Wed, 2005-05-25 at 19:27 -0400, William Beebe wrote: > > > I'm porting an application from Solaris to Fedora Core. Everything is > > > fine except for one issue regarding the use of Posix message queues. > > > In order to solve this problem I'd like to know where the message > > > queues are physically located [on the file system]. > > > > what do you mean by "physically located" ??? > > > > On Solaris, POSIX message queues have a physical representation on the > filesystem. > > Linux just uses a small virtual filesystem which typically isn't even > mounted. Note that the use of the mq_* syscalls doesn't require the > mqueue fs to be mounted. > > -- > Nicholas Miell > > From nscheibl at gmail.com Fri May 27 00:14:27 2005 From: nscheibl at gmail.com (Nicolas Scheibling) Date: Fri, 27 May 2005 02:14:27 +0200 Subject: Fedora 3 kernel 2.6.10-1.770_FC3 and nfsroot problem for PXE/diskless (CONFIG_ROOT_NFS=y) Message-ID: <42966663.1080805@gmail.com> hi all, im trying to setup a diskless environnement with Fedora. I use etherboot (as a bootable CDROM) to load linux kernel via PXE. I have a dhcp, nfs and a tftp server working properly. I have rebuild a 2.6.10-1.770_FC3 fedora kernel because latest 2.6.11 kernel series can't be build on my FC3. I have rebuild a kernel (as an rpm, or via make bzImage), with the NFSROOT option, to allow / mounting via NFS pxelinux.0 boot normaly, load the kernel and the initrd, but when the kernel try to moun the / partition, i obtain a kernel panic message: VFS: Cannont open root device "nfs" or unknow-block(0,255) Please append a correct "root=" boot option Kernel Panic - not syncing : VFS: Unable to mount root fs on unknow-block(0,255) After that, i have tried many other options in the linux kernel (using devfs instead udev, yes i know it's obsolete :)), and some other tips i found but no one worked. Note that on the NFS Server i don't see any packet using the NFS protocol. Is there some special configuration with Fedora kernels and nfsroot option ? Thanks for your suggest ? Here is my pxelinux.cfg/default file : kernel FC3_PXE/vmlinuz APPEND ramdisk_size=100000 root=/dev/ram0 nfsroot=192.168.1.14:/DOC/PXE/root or APPEND initrd=FC3_PXE/initrd.img ramdisk_size=10000 root=/dev/nfs nfsroot=192.168.1.14:/DOC/PXE/root ? Here is the main kernel option needed (if i don't miss one!) and that i have in my kernel: grep -i "IP_PNP\|NFS\|RAMFS" /DOC/PXE/root/boot/config-2.6.10-1.770_FC3nscheibl.20050525 CONFIG_INITRAMFS_SOURCE="" CONFIG_IP_PNP=y CONFIG_IP_PNP_DHCP=y CONFIG_IP_PNP_BOOTP=y CONFIG_IP_PNP_RARP=y CONFIG_RAMFS=y CONFIG_CRAMFS=m CONFIG_NFS_FS=y CONFIG_NFS_V3=y CONFIG_NFS_V4=y CONFIG_NFS_DIRECTIO=y CONFIG_NFSD=m CONFIG_NFSD_V3=y CONFIG_NFSD_V4=y CONFIG_NFSD_TCP=y CONFIG_ROOT_NFS=y CONFIG_NCPFS_NFS_NS=y ? Here is the config of the diskless client in my dhcp server (/etc/dhcpd.conf): group { next-server 192.168.1.14; filename "linux-install/pxelinux.0"; option root-path "192.168.1.14:/DOC/PXE/root,v3,tcp,hard"; host PXED { hardware ethernet 00:48:54:XX:YY:ZZ ; fixed-address 192.168.1.133; } } ? And here is my nfsd config (/etc/exports): /DOC/PXE/root *(ro,no_root_squash,sync) /DOC/PXE/snapshot *(rw,no_root_squash,sync) From alan at balclutha.org Fri May 27 02:39:07 2005 From: alan at balclutha.org (Alan Milligan) Date: Fri, 27 May 2005 12:39:07 +1000 Subject: libgtkembedmoz.so Message-ID: <4296884B.7040902@balclutha.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, My locally built mozilla-*1.7.7-1.3.1.i386.rpm suite seems to not contain libgtkembedmoz.so. Is that to be expected? Has it been deprecated? I'm experimenting with gnophone and although the desktop client doesn't need it, I'm assuming with this plugin, I can use Mozilla to run up gnophone as a plugin. gnophone itself looks a little old as well - am I way off track on using this as a telephony desktop client?? Alan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFClohKCfroLk4EZpkRApllAJ0VRJO50In/gOIGYHhS+5tJMp+hKQCfSy0z zUHb8TBH5R4H/H5VL0MGdlw= =VYFg -----END PGP SIGNATURE----- From buildsys at redhat.com Fri May 27 11:42:40 2005 From: buildsys at redhat.com (Build System) Date: Fri, 27 May 2005 07:42:40 -0400 Subject: rawhide report: 20050527 changes Message-ID: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> Removed package perl-Text-Kakasi Removed package kakasi Removed package libtabe Removed package pmake Removed package autoconvert Removed package recode Removed package SDL_net Removed package SDL_mixer Removed package SDL_image Removed package tmake Removed package skkdic Updated Packages: GFS-kernel-2.6.11.7-20050519.153509.FC4.8 ----------------------------------------- ImageMagick-6.2.2.0-2 --------------------- * Thu May 26 2005 - 6.2.2.0-2 - fix a denial of service in the xwd coder (#158791, CAN-2005-1739) SDL-1.2.8-3.2 ------------- * Thu May 26 2005 Bill Nottingham 1.2.8-3.2 - fix configure script for libdir so library deps are identical on all arches (#158346) cman-kernel-2.6.11.4-20050517.141233.FC4.10 ------------------------------------------- dlm-kernel-2.6.11.4-20050517.141233.FC4.10 ------------------------------------------ eclipse-1:3.1.0_fc-0.M6.22 -------------------------- * Sun May 22 2005 Ben Konrath 3.1.0_fc-0.M6.22 - Bump required version of java-gcj-compat to the latest (-40jpp_24rh). - Add patch to make swt use libgcjawt instead of libjawt for gcj. * Fri May 20 2005 Ben Konrath 3.1.0_fc-0.M6.21 - Add ecj-options patch to bootstrap source. - Make embedded browser widget work (Robin Green). * Thu May 19 2005 Ben Konrath 3.1.0_fc-0.M6.20 - Add ecj-options patch. gnbd-kernel-2.6.11.2-20050420.133124.FC4.32 ------------------------------------------- java-1.4.2-gcj-compat-0:1.4.2.0-40jpp_24rh ------------------------------------------ * Fri May 20 2005 Thomas Fitzsimmons - 0:1.4.2.0-40jpp_24rh - Update libjawt.so symlink to reflect libgcjawt.so's new name. * Thu May 19 2005 Thomas Fitzsimmons - 0:1.4.2.0-40jpp_23rh - Import java-gcj-compat 1.0.29. * Wed May 18 2005 Thomas Fitzsimmons - 0:1.4.2.0-40jpp_22rh - Move gcc-java requirement from base to -devel. system-config-lvm-0.9.32-1.0 ---------------------------- * Thu May 26 2005 Jim Parsons 0.9.32-1.0 - Fix for 158872; x86_64 pam file path * Mon May 23 2005 Jim Parsons 0.9.31-1.0 - Version bump for RHEL4 From enrico.scholz at informatik.tu-chemnitz.de Fri May 27 13:22:02 2005 From: enrico.scholz at informatik.tu-chemnitz.de (Enrico Scholz) Date: Fri, 27 May 2005 15:22:02 +0200 Subject: rawhide report: 20050527 changes In-Reply-To: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> (Build System's message of "Fri, 27 May 2005 07:42:40 -0400") References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> Message-ID: <87u0kon6rp.fsf@kosh.bigo.ensc.de> buildsys at redhat.com (Build System) writes: > Removed package recode What are the alternatives? base64<->data or dos2unix conversions are common tasks and should be solved by Core packages. Enrico From skvidal at phy.duke.edu Fri May 27 13:28:45 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Fri, 27 May 2005 09:28:45 -0400 Subject: rawhide report: 20050527 changes In-Reply-To: <87u0kon6rp.fsf@kosh.bigo.ensc.de> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> Message-ID: <1117200525.9298.59.camel@cutter> On Fri, 2005-05-27 at 15:22 +0200, Enrico Scholz wrote: > buildsys at redhat.com (Build System) writes: > > > Removed package recode > > What are the alternatives? base64<->data or dos2unix conversions are > common tasks and should be solved by Core packages. > > solved in extras by uudeview. -sv From nicolas.mailhot at laposte.net Fri May 27 13:32:26 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Fri, 27 May 2005 15:32:26 +0200 (CEST) Subject: rawhide report: 20050527 changes In-Reply-To: <87u0kon6rp.fsf@kosh.bigo.ensc.de> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> Message-ID: <49998.192.54.193.37.1117200746.squirrel@rousalka.dyndns.org> On Ven 27 mai 2005 15:22, Enrico Scholz a ?crit : > buildsys at redhat.com (Build System) writes: > >> Removed package recode > > What are the alternatives? base64<->data or dos2unix conversions are > common tasks and should be solved by Core packages. dos2unix covers he main recode use case but recode does a lot more than dos2unix. It's invaluable when you need encoding conversions in a shell script (rescuing a borked directory structure for example) Regards, -- Nicolas Mailhot From nicolas.mailhot at laposte.net Fri May 27 13:38:03 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Fri, 27 May 2005 15:38:03 +0200 (CEST) Subject: rawhide report: 20050527 changes In-Reply-To: <1117200525.9298.59.camel@cutter> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> <1117200525.9298.59.camel@cutter> Message-ID: <34378.192.54.193.37.1117201083.squirrel@rousalka.dyndns.org> On Ven 27 mai 2005 15:28, seth vidal a ?crit : > On Fri, 2005-05-27 at 15:22 +0200, Enrico Scholz wrote: >> buildsys at redhat.com (Build System) writes: >> >> > Removed package recode >> >> What are the alternatives? base64<->data or dos2unix conversions are >> common tasks and should be solved by Core packages. >> >> > > solved in extras by uudeview. can you do something like recoded=$(echo $string | recode encoding..utf8) with uudeview ? Thats the 75% most crucial part of my recode usage Regards -- Nicolas Mailhot From skvidal at phy.duke.edu Fri May 27 13:40:34 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Fri, 27 May 2005 09:40:34 -0400 Subject: rawhide report: 20050527 changes In-Reply-To: <34378.192.54.193.37.1117201083.squirrel@rousalka.dyndns.org> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> <1117200525.9298.59.camel@cutter> <34378.192.54.193.37.1117201083.squirrel@rousalka.dyndns.org> Message-ID: <1117201234.9298.65.camel@cutter> On Fri, 2005-05-27 at 15:38 +0200, Nicolas Mailhot wrote: > On Ven 27 mai 2005 15:28, seth vidal a ?crit : > > On Fri, 2005-05-27 at 15:22 +0200, Enrico Scholz wrote: > >> buildsys at redhat.com (Build System) writes: > >> > >> > Removed package recode > >> > >> What are the alternatives? base64<->data or dos2unix conversions are > >> common tasks and should be solved by Core packages. > >> > >> > > > > solved in extras by uudeview. > > can you do something like > recoded=$(echo $string | recode encoding..utf8) > with uudeview ? > > Thats the 75% most crucial part of my recode usage > I dunno, install it and find out :) if not - then grab recode and bring it into extras. -sv From kaboom at oobleck.net Fri May 27 13:37:50 2005 From: kaboom at oobleck.net (Chris Ricker) Date: Fri, 27 May 2005 09:37:50 -0400 (EDT) Subject: rawhide report: 20050527 changes In-Reply-To: <87u0kon6rp.fsf@kosh.bigo.ensc.de> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> Message-ID: On Fri, 27 May 2005, Enrico Scholz wrote: > buildsys at redhat.com (Build System) writes: > > > Removed package recode > > What are the alternatives? base64<->data or dos2unix conversions are > common tasks and should be solved by Core packages. recode's in extras later, chris From jakub at redhat.com Fri May 27 13:42:31 2005 From: jakub at redhat.com (Jakub Jelinek) Date: Fri, 27 May 2005 09:42:31 -0400 Subject: rawhide report: 20050527 changes In-Reply-To: <49998.192.54.193.37.1117200746.squirrel@rousalka.dyndns.org> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> <49998.192.54.193.37.1117200746.squirrel@rousalka.dyndns.org> Message-ID: <20050527134231.GQ9038@devserv.devel.redhat.com> On Fri, May 27, 2005 at 03:32:26PM +0200, Nicolas Mailhot wrote: > > On Ven 27 mai 2005 15:22, Enrico Scholz a ?crit : > > buildsys at redhat.com (Build System) writes: > > > >> Removed package recode > > > > What are the alternatives? base64<->data or dos2unix conversions are > > common tasks and should be solved by Core packages. > > dos2unix covers he main recode use case but recode does a lot more than > dos2unix. It's invaluable when you need encoding conversions in a shell > script Encoding conversions in a shell script can be done with iconv(1). Jakub From nicolas.mailhot at laposte.net Fri May 27 13:48:49 2005 From: nicolas.mailhot at laposte.net (Nicolas Mailhot) Date: Fri, 27 May 2005 15:48:49 +0200 (CEST) Subject: rawhide report: 20050527 changes In-Reply-To: <20050527134231.GQ9038@devserv.devel.redhat.com> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> <49998.192.54.193.37.1117200746.squirrel@rousalka.dyndns.org> <20050527134231.GQ9038@devserv.devel.redhat.com> Message-ID: <37955.192.54.193.37.1117201729.squirrel@rousalka.dyndns.org> On Ven 27 mai 2005 15:42, Jakub Jelinek a ?crit : > On Fri, May 27, 2005 at 03:32:26PM +0200, Nicolas Mailhot wrote: >> >> On Ven 27 mai 2005 15:22, Enrico Scholz a ?crit : >> > buildsys at redhat.com (Build System) writes: >> > >> >> Removed package recode >> > >> > What are the alternatives? base64<->data or dos2unix conversions are >> > common tasks and should be solved by Core packages. >> >> dos2unix covers he main recode use case but recode does a lot more than >> dos2unix. It's invaluable when you need encoding conversions in a shell >> script > > Encoding conversions in a shell script can be done with iconv(1). Seems nice. Thanks for the tip. I won't miss recode afterall. Regards, -- Nicolas Mailhot From ville.skytta at iki.fi Fri May 27 13:52:35 2005 From: ville.skytta at iki.fi (Ville =?ISO-8859-1?Q?Skytt=E4?=) Date: Fri, 27 May 2005 16:52:35 +0300 Subject: rawhide report: 20050527 changes In-Reply-To: <34378.192.54.193.37.1117201083.squirrel@rousalka.dyndns.org> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <87u0kon6rp.fsf@kosh.bigo.ensc.de> <1117200525.9298.59.camel@cutter> <34378.192.54.193.37.1117201083.squirrel@rousalka.dyndns.org> Message-ID: <1117201955.25731.36.camel@bobcat.mine.nu> On Fri, 2005-05-27 at 15:38 +0200, Nicolas Mailhot wrote: > can you do something like > recoded=$(echo $string | recode encoding..utf8) > with uudeview ? Probably not, but how about iconv(1)? From byte at aeon.com.my Fri May 27 13:30:10 2005 From: byte at aeon.com.my (Colin Charles) Date: Fri, 27 May 2005 23:30:10 +1000 Subject: PPC boot.iso broken today? In-Reply-To: <1117121690.3225.6.camel@enki.eridu> References: <20050526155855.4a3f575b@python2> <1117121690.3225.6.camel@enki.eridu> Message-ID: <1117200610.31902.175.camel@arena.soho.bytebot.net> On Thu, 2005-05-26 at 11:34 -0400, Paul Nasrat wrote: > > linux maxcpus=1 noisapnp isapnp went away with the boot.iso required for the 20050526 tree, fwiw -- Colin Charles, http://www.bytebot.net/ From pedro.lamarao at mndfck.org Fri May 27 16:32:58 2005 From: pedro.lamarao at mndfck.org (=?ISO-8859-1?Q?Pedro_Lamar=E3o?=) Date: Fri, 27 May 2005 13:32:58 -0300 Subject: Physical location of Posix message queues In-Reply-To: <4295E65E.1080204@mndfck.org> References: <4295E65E.1080204@mndfck.org> Message-ID: <42974BBA.8020109@mndfck.org> Pedro Lamar?o wrote: > William Beebe wrote: > >>I'm porting an application from Solaris to Fedora Core. Everything is >>fine except for one issue regarding the use of Posix message queues. >>In order to solve this problem I'd like to know where the message >>queues are physically located. Where on Fedora Core (or Linux for that >>matter) are the message queues located? I've tried just looking on the >>file system and I've tried looking in the code for a Clue (both the >>kernel and glibc). If someone could just tell me, or if there are >>pointers to other documentation explaining some of the finer details >>of Posix message queue implementation on Linux I'd appreciate it. >> > > > Check /var/spool/postfix . > I'm feeling ridiculous. I swear I read "Postfix". :| -- Pedro Lamar?o "Merda faz as flores crescerem e isso ? bonito." From ronny-vlug at vlugnet.org Fri May 27 16:53:13 2005 From: ronny-vlug at vlugnet.org (Ronny Buchmann) Date: Fri, 27 May 2005 18:53:13 +0200 Subject: rawhide report: 20050527 changes In-Reply-To: <34378.192.54.193.37.1117201083.squirrel@rousalka.dyndns.org> References: <200505271142.j4RBgeCG010521@porkchop.devel.redhat.com> <1117200525.9298.59.camel@cutter> <34378.192.54.193.37.1117201083.squirrel@rousalka.dyndns.org> Message-ID: <200505271853.14195.ronny-vlug@vlugnet.org> On Friday 27 May 2005 15:38, Nicolas Mailhot wrote: > On Ven 27 mai 2005 15:28, seth vidal a ?crit : > > On Fri, 2005-05-27 at 15:22 +0200, Enrico Scholz wrote: > >> buildsys at redhat.com (Build System) writes: > >> > Removed package recode > >> > >> What are the alternatives? base64<->data or dos2unix conversions are > >> common tasks and should be solved by Core packages. > > > > solved in extras by uudeview. > > can you do something like > recoded=$(echo $string | recode encoding..utf8) > with uudeview ? I assume convmv (in extras) and/or iconv will help you -- http://LinuxWiki.org/RonnyBuchmann From Axel.Thimm at ATrpms.net Fri May 27 17:30:07 2005 From: Axel.Thimm at ATrpms.net (Axel Thimm) Date: Fri, 27 May 2005 19:30:07 +0200 Subject: alsa 1.0.9 out: epoch issue with FC4 test and rawhide Message-ID: <20050527173007.GG5107@neu.nirvana> Hi, alsa 1.0.9 is out. Any chance this will make it into FC4? Probably not, but could be one of the earliest updates. alsa 1.0.9rcX has been packaged in such a way that it will be newer than 1.0.9 unless an epoch bump is issued. It has been discussed in the past that rawhide/test release do not give any upgrade path guarantees for various reasons (among other to be able to downgrade a bunch of packages like kde w/o having an epoch inflation). Should alsa be epochized (poor alsa), or the upgrade paths from rawhide FC4tX broken? Personally I think epochs are really evil, and setups with test releases need to be reinstalled anyway. bugzilla is here: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=158859 -- Axel.Thimm at ATrpms.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ggw at wolves.homeip.net Fri May 27 17:28:39 2005 From: ggw at wolves.homeip.net (G.Wolfe Woodbury) Date: Fri, 27 May 2005 17:28:39 +0000 (UTC) Subject: What happened to the it821x IDE/PCI Raid driver? References: <20050525222249.GA23229@wolves.durham.nc.us> <20050525222616.GA7050@devserv.devel.redhat.com> Message-ID: On Wed, 25 May 2005 18:26:16 -0400, Alan shaped electrons to write: > On Wed, May 25, 2005 at 06:22:49PM -0400, Gregory Woodbury wrote: >> In the 2.6.10 series of kernels, Alan Cox' IT821x PCI/IDE-RAID driver >> was included in the kernel as a module. In 2.6.11 kernels, the module >> disappeared and the device is not recognized by the kernel. > > Its only in the -ac kernels. Fedora includes the -ac IDE code on the whole > but it was missing for most of RHEL4 test because it needed porting to > 2.6.12rc and Dave had a lot more things on his plate that were far more > urgent. > Thank you Alan, Dave said it re-appeared in a recent update kernel, and it did! It works much better than the previous incarnation and is running strong with full DMA in Raid mode. A big Thank You to you and Dave for this driver! -- G.Wolfe Woodbury `- -' RHCT U The Line Eater is a boojum! From rdieter at math.unl.edu Fri May 27 17:45:06 2005 From: rdieter at math.unl.edu (Rex Dieter) Date: Fri, 27 May 2005 12:45:06 -0500 Subject: alsa 1.0.9 out: epoch issue with FC4 test and rawhide In-Reply-To: <20050527173007.GG5107@neu.nirvana> References: <20050527173007.GG5107@neu.nirvana> Message-ID: Axel Thimm wrote: > alsa 1.0.9 is out. Any chance this will make it into FC4? Probably > not, but could be one of the earliest updates. > > alsa 1.0.9rcX has been packaged in such a way that it will be newer > than 1.0.9 unless an epoch bump is issued. One good reason why one should always try to avoid non-numeric bits in the "version" tag. -- rex From pjones at redhat.com Fri May 27 20:15:45 2005 From: pjones at redhat.com (Peter Jones) Date: Fri, 27 May 2005 16:15:45 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <20050524222445.GA23131@jadzia.bu.edu> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> <1116462767.3483.19.camel@jellyfish.redfishdemo.com> <1116973123.7655.2.camel@localhost.localdomain> <20050524222445.GA23131@jadzia.bu.edu> Message-ID: <1117224945.6021.5.camel@localhost.localdomain> On Tue, 2005-05-24 at 18:24 -0400, Matthew Miller wrote: > On Tue, May 24, 2005 at 06:18:43PM -0400, Peter Jones wrote: > > > 1. Each user should have there own ~/tmp space which only they can > > > access. This could be used for the users agent sockets, but also just > [...] > > Of course, ~/tmp/ really sucks for this -- lots of places use nfs > > homedirs, and you'd rather not put the sort of stuff you use /tmp for on > > the network. > > Which goes back to my original post here, which cleverly (?) checks if ~/tmp > is on a local filesystem and uses mktemp in /tmp otherwise. Yeah, that's better than just blindly using ~/tmp/. But why have the extra complexity? Why not always do mktemp and the bind+namespace magic? This does have some advantage -- all users' tmp dirs are created the way the admin intended when he set the system up, and they're easy to find if he needs to look for them, for whatever reason. -- Peter From alan at redhat.com Fri May 27 20:54:47 2005 From: alan at redhat.com (Alan Cox) Date: Fri, 27 May 2005 16:54:47 -0400 Subject: enhance security via private TMP/TMPDIR by default In-Reply-To: <1117224945.6021.5.camel@localhost.localdomain> References: <20050512155504.GA25624@jadzia.bu.edu> <1115913808.3424.5.camel@nexus.verbum.private> <87is1gjt7v.fsf@kosh.bigo.ensc.de> <1116455960.3552.42.camel@nexus.verbum.private> <20050518224834.GB25468@nostromo.devel.redhat.com> <1116462767.3483.19.camel@jellyfish.redfishdemo.com> <1116973123.7655.2.camel@localhost.localdomain> <20050524222445.GA23131@jadzia.bu.edu> <1117224945.6021.5.camel@localhost.localdomain> Message-ID: <20050527205447.GD9772@devserv.devel.redhat.com> On Fri, May 27, 2005 at 04:15:45PM -0400, Peter Jones wrote: > Yeah, that's better than just blindly using ~/tmp/. But why have the > extra complexity? Why not always do mktemp and the bind+namespace > magic? This does have some advantage -- all users' tmp dirs are created > the way the admin intended when he set the system up, and they're easy > to find if he needs to look for them, for whatever reason. There is another ~/tmp problem to worry about. Quite a few programs behave badly on start up if they can't write to their tmp files. ~/tmp combined with quota will make this worse. Alan (who used /tmp/$USER as /tmp when playing with this stuff) From Matt_Domsch at dell.com Fri May 27 20:57:41 2005 From: Matt_Domsch at dell.com (Matt Domsch) Date: Fri, 27 May 2005 15:57:41 -0500 Subject: [PATCH 1.4.11 1/2] OpenIPMI initscript and config file In-Reply-To: <20050520095457.GA32316@lists.us.dell.com> References: <20050520040924.GA25048@lists.us.dell.com> <20050520041031.GB25048@lists.us.dell.com> <20050520041924.GA24328@nostromo.devel.redhat.com> <20050520095457.GA32316@lists.us.dell.com> Message-ID: <20050527205741.GA6192@lists.us.dell.com> On Fri, May 20, 2005 at 04:54:57AM -0500, Matt Domsch wrote: > On Fri, May 20, 2005 at 12:19:24AM -0400, Bill Nottingham wrote: > > Matt Domsch (Matt_Domsch at dell.com) said: > > > + if [ "${DEV_IPMI}" = "1" ]; then > > > + modprobe ipmi_devintf || RETVAL=2 > > > + if [ "${RETVAL}" != "2" ]; then > > > + # Note, this really should be done by udev on 2.6 > > > + DEVMAJOR=`cat /proc/devices | awk '/ipmidev/{print $1}'` > > > + mknod -m 0600 /dev/ipmi${INTF_NUM} c ${DEVMAJOR} 0 || RETVAL=2 > > > + ln -sf /dev/ipmi${INTF_NUM} /dev/ipmi || RETVAL=2 > > > + fi > > > + fi > > > > Why isn't the driver fixed to populate sysfs properly? > > Just-in-time programming. Corey submitted class_simple support this > evening. > http://marc.theaimsgroup.com/?l=linux-kernel&m=111654581209165& With class_simple support, the initscripts now look as follows. Confirmed this works on a RHEL4-U1 beta kernel build, sysfs makes /dev/ipmi0 for us. Per Corey, there's no real tool that uses the /dev/ipmi symlink, they should be using /dev/ipmi0. So I removed the creation of that symlink in the patch below. The initscripts are necessary until there's a mechanism that can autodetect the presence of IPMI-capable hardware and load the modules automatically. That's a future task, and not one I want to hold this up for. This will work for now. Thanks, Matt -- Matt Domsch Software Architect Dell Linux Solutions linux.dell.com & www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com --- /dev/null Thu Apr 11 09:25:15 2002 +++ OpenIPMI-1.4.11/ipmi.init Fri May 27 15:51:29 2005 @@ -0,0 +1,192 @@ +#!/bin/sh +############################################################################# +# +# ipmi: OpenIPMI Driver init script +# +# Authors: Matt Domsch +# Chris Poblete +# +# chkconfig: 2345 04 96 +# description: OpenIPMI Driver init script +# +### BEGIN INIT INFO +# Provides: ipmidrv +# Required-Start: $localfs $remotefs $syslog +# Required-Stop: $localfs $remotefs $syslog +# Default-Start: 2 3 4 5 +# Default-Stop: +# Short-Description: OpenIPMI Driver init script +# Description: OpenIPMI Driver init script +### END INIT INFO +# +############################################################################# +# for log_success_msg and friends +[ -r /lib/lsb/init-functions ] && . /lib/lsb/init-functions +# source config info +[ -r /etc/sysconfig/ipmi ] && . /etc/sysconfig/ipmi + +############################################################################# +# GLOBALS +############################################################################# +MODULE_NAME="ipmi" +INTF_NUM=0 + +IPMI_SMB_MODULE_NAME="ipmi_smb" +IPMI_SI_MODULE_NAME="ipmi_si" +kernel=`uname -r | cut -d. -f1-2` +if [ "${kernel}" == "2.4" ]; then + IPMI_SMB_MODULE_NAME="ipmi_smb_intf" + IPMI_SI_MODULE_NAME="ipmi_si_drv" +fi + +MODULES="ipmi_radisys ipmi_imb ipmi_poweroff ipmi_watchdog \ + ipmi_devintf ${IPMI_SMB_MODULE_NAME} ${IPMI_SI_MODULE_NAME} \ + ipmi_msghandler" + + +RETVAL=0 +LOCKFILE=/var/lock/subsys/ipmi + +############################################################################# +start_watchdog() +{ + if [ "${IPMI_WATCHDOG}" = "1" ]; then + modprobe ipmi_watchdog || RETVAL=2 + fi +} + +stop_watchdog() +{ + modprobe -r ipmi_watchdog +} + +start_powercontrol() +{ + local poweroff_opts="" + if [ "${IPMI_POWEROFF}" = "1" ]; then + [ "${IPMI_POWERCYCLE}" == "1" ] && poweroff_opts="chassis_ctrl_cmd_param=2" + modprobe ipmi_poweroff "${poweroff_opts}" || RETVAL=2 + fi +} + +stop_powercontrol() +{ + modprobe -r ipmi_poweroff +} + +############################################################################# +unload_ipmi_modules() +{ + [ ! -x /sbin/udev ] && rm -f "/dev/ipmi${INTF_NUM}" + for m in ${MODULES}; do + modprobe -q -r ${m} + done +} + +############################################################################# +load_ipmi_modules () +{ + modprobe ipmi_msghandler || RETVAL=1 + if [ "${IPMI_SI}" = "1" ]; then + modprobe ${IPMI_SI_MODULE_NAME} || RETVAL=1 + fi + if [ "${IPMI_SMB}" = "1" ]; then + modprobe ${IPMI_SMB_MODULE_NAME} || RETVAL=1 + fi + [ "${RETVAL}" = "1" ] && unload_ipmi_modules && return + + if [ "${DEV_IPMI}" = "1" ]; then + modprobe ipmi_devintf || RETVAL=2 + if [ "${RETVAL}" != "2" ]; then + if [ ! -x /sbin/udev ]; then + DEVMAJOR=`cat /proc/devices | awk '/ipmidev/{print $1}'` + mknod -m 0600 /dev/ipmi${INTF_NUM} c ${DEVMAJOR} 0 || RETVAL=2 + fi + fi + fi + + start_watchdog + start_powercontrol + if [ "${IPMI_IMB}" = "1" ]; then + modprobe ipmi_imb || RETVAL=2 + # FIXME create canonical /dev/foo entry + fi + if [ "${IPMI_RADISYS}" = "1" ]; then + modprobe ipmi_radisys || RETVAL=2 + # FIXME create canonical /dev/foo entry + fi + return +} + +############################################################################# +start() +{ + echo -n $"Starting ${MODULE_NAME} drivers: " + load_ipmi_modules + [ "${RETVAL}" = "1" ] && log_failure_msg && return + [ "${RETVAL}" = "2" ] && touch ${LOCKFILE} && log_warning_msg + [ "${RETVAL}" = "0" ] && touch ${LOCKFILE} && log_success_msg +} + +############################################################################# +stop() +{ + echo -n $"Stopping ${MODULE_NAME} drivers: " + unload_ipmi_modules + rm -f ${LOCKFILE} + log_success_msg +} + +############################################################################# +restart() +{ + stop + start +} + +############################################################################# +status () +{ + for m in ${MODULES}; do + if /sbin/lsmod | grep $m >/dev/null 2>&1 ; then + echo "$m module loaded" + else + echo "$m module not loaded" + fi + done +} + +usage () +{ + echo $"Usage: $0 {start|stop|status|restart|condrestart|" + echo $" start-watchdog|stop-watchdog|start-powercontrol|stop-powercontrol}" 1>&2 + RETVAL=1 +} + +condrestart () +{ + [ -e ${LOCKFILE} ] && restart +} + +############################################################################# +# MAIN +############################################################################# +case "$1" in + start) start ;; + stop) stop ;; + restart) restart ;; + status) status ;; + condrestart) condrestart ;; + start-watchdog) start_watchdog ;; + stop-watchdog) stop_watchdog ;; + start-powercontrol) start_powercontrol ;; + stop-powercontrol) stop_powercontrol ;; + *) usage ;; +esac + +exit ${RETVAL} + +############################################################################# +# end of file +############################################################################# + --- /dev/null Thu Apr 11 09:25:15 2002 +++ OpenIPMI-1.4.11/ipmi.sysconf Fri May 27 15:51:37 2005 @@ -0,0 +1,30 @@ +# Enable standard hardware interfaces (KCS, BT, SMIC) +# You probably want this enabled. +IPMI_SI=1 + +# Enable nonstandard interfaces (SMB via i2c) +# IPMI_SMB=1 + +# Enable /dev/ipmi0 interface, used by ipmitool, ipmicmd, +# and other userspace IPMI-using applications. +# You probably want this enabled. +DEV_IPMI=1 + +# Enable IPMI_WATCHDOG if you want the IPMI watchdog +# to reboot the system if it hangs +# IPMI_WATCHDOG=1 + +# Enable IPMI_POWEROFF if you want the IPMI +# poweroff module to be loaded. +# IPMI_POWEROFF=1 + +# Enable IPMI_POWERCYCLE if you want the system to be power-cycled (power +# down, delay briefly, power on) rather than power off, on systems +# that support such. IPMI_POWEROFF=1 is also required. +# IPMI_POWERCYCLE=1 + +# Enable "legacy" interfaces for applications +# Intel IMB driver interface +# IPMI_IMB=1 +# Radisys driver interface +# IPMI_RADISYS=1 From tibbs at math.uh.edu Fri May 27 21:00:37 2005 From: tibbs at math.uh.edu (Jason L Tibbitts III) Date: Fri, 27 May 2005 16:00:37 -0500 Subject: alsa 1.0.9 out: epoch issue with FC4 test and rawhide In-Reply-To: <20050527173007.GG5107@neu.nirvana> (Axel Thimm's message of "Fri, 27 May 2005 19:30:07 +0200") References: <20050527173007.GG5107@neu.nirvana> Message-ID: >>>>> "AT" == Axel Thimm writes: AT> alsa 1.0.9rcX has been packaged in such a way that it will be AT> newer than 1.0.9 unless an epoch bump is issued. Or the new package is given version "1.0.9release". - J< From Matt_Domsch at dell.com Fri May 27 22:01:21 2005 From: Matt_Domsch at dell.com (Matt Domsch) Date: Fri, 27 May 2005 17:01:21 -0500 Subject: [PATCH 1.4.11 1/2] OpenIPMI initscript and config file In-Reply-To: <20050527205741.GA6192@lists.us.dell.com> References: <20050520040924.GA25048@lists.us.dell.com> <20050520041031.GB25048@lists.us.dell.com> <20050520041924.GA24328@nostromo.devel.redhat.com> <20050520095457.GA32316@lists.us.dell.com> <20050527205741.GA6192@lists.us.dell.com> Message-ID: <20050527220121.GA10606@lists.us.dell.com> On Fri, May 27, 2005 at 03:57:41PM -0500, Matt Domsch wrote: > On Fri, May 20, 2005 at 04:54:57AM -0500, Matt Domsch wrote: > > On Fri, May 20, 2005 at 12:19:24AM -0400, Bill Nottingham wrote: > > > Matt Domsch (Matt_Domsch at dell.com) said: > > > > + if [ "${DEV_IPMI}" = "1" ]; then > > > > + modprobe ipmi_devintf || RETVAL=2 > > > > + if [ "${RETVAL}" != "2" ]; then > > > > + # Note, this really should be done by udev on 2.6 > > > > + DEVMAJOR=`cat /proc/devices | awk '/ipmidev/{print $1}'` > > > > + mknod -m 0600 /dev/ipmi${INTF_NUM} c ${DEVMAJOR} 0 || RETVAL=2 > > > > + ln -sf /dev/ipmi${INTF_NUM} /dev/ipmi || RETVAL=2 > > > > + fi > > > > + fi > > > > > > Why isn't the driver fixed to populate sysfs properly? > > > > Just-in-time programming. Corey submitted class_simple support this > > evening. > > http://marc.theaimsgroup.com/?l=linux-kernel&m=111654581209165& > > With class_simple support, the initscripts now look as follows. > > Confirmed this works on a RHEL4-U1 beta kernel build, sysfs makes > /dev/ipmi0 for us. Per Corey, there's no real tool that uses the > /dev/ipmi symlink, they should be using /dev/ipmi0. So I removed the > creation of that symlink in the patch below. > > The initscripts are necessary until there's a mechanism that can > autodetect the presence of IPMI-capable hardware and load the modules > automatically. That's a future task, and not one I want to hold this > up for. This will work for now. Oh, the ipmi_poweroff module option for powercycle changed names in Corey's last set of patches. As nothing was using the old name really yet, here's the initscripts using the new poweroff_control option name. Thanks, Matt -- Matt Domsch Software Architect Dell Linux Solutions linux.dell.com & www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com --- /dev/null Thu Apr 11 09:25:15 2002 +++ OpenIPMI-1.4.11/ipmi.init Fri May 27 16:59:23 2005 @@ -0,0 +1,192 @@ +#!/bin/sh +############################################################################# +# +# ipmi: OpenIPMI Driver init script +# +# Authors: Matt Domsch +# Chris Poblete +# +# chkconfig: 2345 04 96 +# description: OpenIPMI Driver init script +# +### BEGIN INIT INFO +# Provides: ipmidrv +# Required-Start: $localfs $remotefs $syslog +# Required-Stop: $localfs $remotefs $syslog +# Default-Start: 2 3 4 5 +# Default-Stop: +# Short-Description: OpenIPMI Driver init script +# Description: OpenIPMI Driver init script +### END INIT INFO +# +############################################################################# +# for log_success_msg and friends +[ -r /lib/lsb/init-functions ] && . /lib/lsb/init-functions +# source config info +[ -r /etc/sysconfig/ipmi ] && . /etc/sysconfig/ipmi + +############################################################################# +# GLOBALS +############################################################################# +MODULE_NAME="ipmi" +INTF_NUM=0 + +IPMI_SMB_MODULE_NAME="ipmi_smb" +IPMI_SI_MODULE_NAME="ipmi_si" +kernel=`uname -r | cut -d. -f1-2` +if [ "${kernel}" == "2.4" ]; then + IPMI_SMB_MODULE_NAME="ipmi_smb_intf" + IPMI_SI_MODULE_NAME="ipmi_si_drv" +fi + +MODULES="ipmi_radisys ipmi_imb ipmi_poweroff ipmi_watchdog \ + ipmi_devintf ${IPMI_SMB_MODULE_NAME} ${IPMI_SI_MODULE_NAME} \ + ipmi_msghandler" + + +RETVAL=0 +LOCKFILE=/var/lock/subsys/ipmi + +############################################################################# +start_watchdog() +{ + if [ "${IPMI_WATCHDOG}" = "1" ]; then + modprobe ipmi_watchdog || RETVAL=2 + fi +} + +stop_watchdog() +{ + modprobe -r ipmi_watchdog +} + +start_powercontrol() +{ + local poweroff_opts="" + if [ "${IPMI_POWEROFF}" = "1" ]; then + [ "${IPMI_POWERCYCLE}" == "1" ] && poweroff_opts="poweroff_control=2" + modprobe ipmi_poweroff "${poweroff_opts}" || RETVAL=2 + fi +} + +stop_powercontrol() +{ + modprobe -r ipmi_poweroff +} + +############################################################################# +unload_ipmi_modules() +{ + [ ! -x /sbin/udev ] && rm -f "/dev/ipmi${INTF_NUM}" + for m in ${MODULES}; do + modprobe -q -r ${m} + done +} + +############################################################################# +load_ipmi_modules () +{ + modprobe ipmi_msghandler || RETVAL=1 + if [ "${IPMI_SI}" = "1" ]; then + modprobe ${IPMI_SI_MODULE_NAME} || RETVAL=1 + fi + if [ "${IPMI_SMB}" = "1" ]; then + modprobe ${IPMI_SMB_MODULE_NAME} || RETVAL=1 + fi + [ "${RETVAL}" = "1" ] && unload_ipmi_modules && return + + if [ "${DEV_IPMI}" = "1" ]; then + modprobe ipmi_devintf || RETVAL=2 + if [ "${RETVAL}" != "2" ]; then + if [ ! -x /sbin/udev ]; then + DEVMAJOR=`cat /proc/devices | awk '/ipmidev/{print $1}'` + mknod -m 0600 /dev/ipmi${INTF_NUM} c ${DEVMAJOR} 0 || RETVAL=2 + fi + fi + fi + + start_watchdog + start_powercontrol + if [ "${IPMI_IMB}" = "1" ]; then + modprobe ipmi_imb || RETVAL=2 + # FIXME create canonical /dev/foo entry + fi + if [ "${IPMI_RADISYS}" = "1" ]; then + modprobe ipmi_radisys || RETVAL=2 + # FIXME create canonical /dev/foo entry + fi + return +} + +############################################################################# +start() +{ + echo -n $"Starting ${MODULE_NAME} drivers: " + load_ipmi_modules + [ "${RETVAL}" = "1" ] && log_failure_msg && return + [ "${RETVAL}" = "2" ] && touch ${LOCKFILE} && log_warning_msg + [ "${RETVAL}" = "0" ] && touch ${LOCKFILE} && log_success_msg +} + +############################################################################# +stop() +{ + echo -n $"Stopping ${MODULE_NAME} drivers: " + unload_ipmi_modules + rm -f ${LOCKFILE} + log_success_msg +} + +############################################################################# +restart() +{ + stop + start +} + +############################################################################# +status () +{ + for m in ${MODULES}; do + if /sbin/lsmod | grep $m >/dev/null 2>&1 ; then + echo "$m module loaded" + else + echo "$m module not loaded" + fi + done +} + +usage () +{ + echo $"Usage: $0 {start|stop|status|restart|condrestart|" + echo $" start-watchdog|stop-watchdog|start-powercontrol|stop-powercontrol}" 1>&2 + RETVAL=1 +} + +condrestart () +{ + [ -e ${LOCKFILE} ] && restart +} + +############################################################################# +# MAIN +############################################################################# +case "$1" in + start) start ;; + stop) stop ;; + restart) restart ;; + status) status ;; + condrestart) condrestart ;; + start-watchdog) start_watchdog ;; + stop-watchdog) stop_watchdog ;; + start-powercontrol) start_powercontrol ;; + stop-powercontrol) stop_powercontrol ;; + *) usage ;; +esac + +exit ${RETVAL} + +############################################################################# +# end of file +############################################################################# + --- /dev/null Thu Apr 11 09:25:15 2002 +++ OpenIPMI-1.4.11/ipmi.sysconf Fri May 27 15:51:37 2005 @@ -0,0 +1,30 @@ +# Enable standard hardware interfaces (KCS, BT, SMIC) +# You probably want this enabled. +IPMI_SI=1 + +# Enable nonstandard interfaces (SMB via i2c) +# IPMI_SMB=1 + +# Enable /dev/ipmi0 interface, used by ipmitool, ipmicmd, +# and other userspace IPMI-using applications. +# You probably want this enabled. +DEV_IPMI=1 + +# Enable IPMI_WATCHDOG if you want the IPMI watchdog +# to reboot the system if it hangs +# IPMI_WATCHDOG=1 + +# Enable IPMI_POWEROFF if you want the IPMI +# poweroff module to be loaded. +# IPMI_POWEROFF=1 + +# Enable IPMI_POWERCYCLE if you want the system to be power-cycled (power +# down, delay briefly, power on) rather than power off, on systems +# that support such. IPMI_POWEROFF=1 is also required. +# IPMI_POWERCYCLE=1 + +# Enable "legacy" interfaces for applications +# Intel IMB driver interface +# IPMI_IMB=1 +# Radisys driver interface +# IPMI_RADISYS=1 From seandarcy2 at gmail.com Fri May 27 22:06:28 2005 From: seandarcy2 at gmail.com (sean) Date: Fri, 27 May 2005 18:06:28 -0400 Subject: 2.6.11-1.1287_FC4 kernel panic on intel 915 In-Reply-To: References: Message-ID: sean wrote: > I've installed kernel-smp-2.6.11-1.1287_FC4 on a Gigabyte GA-8i915G Pro > ( with has ICH6 ) with a pentium 4 2.8g HT chip. It has a 300g maxtor > SATA drive ( only drive ). > > It boots, but then kernel panic mounting root. > > It looks alright until: > > SCSI subsystem initialized > Loading libata.ko module > Loading ata_piix.ko module > Loading dm-mod.ko module > ata_piix: Unknown symbol ata_scsi_ioctl > ata_piix: Unknown symbol ata_std_bios_param > ata_piix: Unknown symbol ata_tf_read > ata_piix: Unknown symbol ata_tf_write > ata_piix: Unknown symbol ata_bmdma_start > ata_piix: Unknown symbol ata_bmdma_setup > ata_piix: Unknown symbol ata_bmdma_stop > [...........] > ata_piix: Unknown symbol ata_port_start > insmod: error inserting '/lib/ata_piix.ko': -1 Unknown symbol in module > > > and then all sorts of bad stuff happens. > > I did go get 2.6.12-rc4 today. Built it with ata_piix compiled in ( not > as a module ). That worked. Booted. Mounted root. However, same kernel > panic if ahci.ko was also compiled in. I also got a kernel panic when > ata_piix was a module and I removed ahci.ko from modules. > > sean > Tried kernel-smp-2.6.11-1.1341_FC4. Same result. Has anybody been able to boot on a 915 mobo with only a SATA drive or / on a SATA drive? sean From buildsys at redhat.com Sat May 28 11:38:59 2005 From: buildsys at redhat.com (Build System) Date: Sat, 28 May 2005 07:38:59 -0400 Subject: rawhide report: 20050528 changes Message-ID: <200505281138.j4SBcxOa032276@porkchop.devel.redhat.com> Updated Packages: alsa-lib-1.0.9rc4-2 ------------------- * Fri May 27 2005 Martin Stransky 1.0.9rc4-2 - alsacard utility for s-c-s gnome-applets-1:2.10.1-9 ------------------------ * Fri May 27 2005 Bill Nottingham 1:2.10.1-9 - remove setuid bit from cpufreq-selector, usermode-ify it gnu-crypto-0:2.0.1-1jpp_5fc --------------------------- * Thu May 26 2005 Thomas Fitzsimmons - 0:2.0.1-1jpp_5fc - Separate post and postun lines. - Require libgcj, not gcc-java. isdn4k-utils-3.2-28 ------------------- * Fri May 27 2005 Bill Nottingham 3.2-28 - remove setuid bit from vboxbeep jessie-0:1.0.0-8 ---------------- * Thu May 26 2005 Thomas Fitzsimmons - 0:1.0.0-8 - Require libgcj. - Separate post and postun requirement lines. - Specify gnu-crypto jar directly. nfs-utils-1.0.7-8 ----------------- * Thu May 26 2005 Steve Dickson 1.0.7-8 - Fixed subscripting problem in idmapd (bz 158188) screen-4.0.2-9 -------------- * Fri May 27 2005 Bill Nottingham - 4.0.2-9 - don't use utmp group for socket dir; use a dedicated screen gid system-config-netboot-0.1.16-1_FC3 ---------------------------------- * Thu May 26 2005 Jason Vas Dias 0.1.16-1 - fix bugs 144240, 148022, 149000, 153047, 154982 From buildsys at redhat.com Sun May 29 11:35:39 2005 From: buildsys at redhat.com (Build System) Date: Sun, 29 May 2005 07:35:39 -0400 Subject: rawhide report: 20050529 changes Message-ID: <200505291135.j4TBZd6r010525@porkchop.devel.redhat.com> Updated Packages: (none) From frankxchen at yahoo.com Sun May 29 17:03:07 2005 From: frankxchen at yahoo.com (Frank Chen) Date: Sun, 29 May 2005 10:03:07 -0700 (PDT) Subject: How to debug a hung system? In-Reply-To: 6667 Message-ID: <20050529170307.70658.qmail@web54501.mail.yahoo.com> I installed FC3 on a ASUS Terminator C3 PC. It has a VIA C3 CPU. The system becomes completely unresponsive from time to time. The only thing I could do was to do a hard reboot. Any suggestions on how to debug such a system? Thanks, Frank __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From adrian at lisas.de Sun May 29 17:50:46 2005 From: adrian at lisas.de (Adrian Reber) Date: Sun, 29 May 2005 19:50:46 +0200 Subject: How to debug a hung system? In-Reply-To: <20050529170307.70658.qmail@web54501.mail.yahoo.com> References: <20050529170307.70658.qmail@web54501.mail.yahoo.com> Message-ID: <20050529175046.GA29411@lisas.de> On Sun, May 29, 2005 at 10:03:07AM -0700, Frank Chen wrote: > I installed FC3 on a ASUS Terminator C3 PC. It has a > VIA C3 CPU. The system becomes completely unresponsive > from time to time. The only thing I could do was to do > a hard reboot. Any suggestions on how to debug such a > system? No suggestions, but this sounds like https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=140873 Adrian From jerone at gmail.com Mon May 30 03:22:31 2005 From: jerone at gmail.com (Jerone Young) Date: Sun, 29 May 2005 22:22:31 -0500 Subject: FC4 Display GUI tool does not have 1920x1200 as selectable resolution Message-ID: <9f50a7a00505292022c087fa6@mail.gmail.com> I just got a Samsung 243T with Nvidia Geforce 6600GT (this connection is over DVI). When setting up FC4 , the highest resolution I can choose is 1680x1050. One thing FC4 is doing is that it is using the "vesa" driver by default for my card instead of the "nv" driver. But even once I do select the nv driver in the GUI I still don't see 1920x1200 as a selectable resolution. Going into xorg.conf and puting 1920x1200 as a resolution solves the problem & I X works in 1920x1200. "lscpi -nv" outputf for my video card: 01:00.0 Class 0300: 10de:00f1 (rev a2) Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 5 Memory at f8000000 (32-bit, non-prefetchable) [size=16M] Memory at e0000000 (32-bit, prefetchable) [size=256M] Memory at f9000000 (32-bit, non-prefetchable) [size=16M] Capabilities: [60] Power Management version 2 Capabilities: [44] AGP version 3.0 From jerone at gmail.com Mon May 30 03:41:57 2005 From: jerone at gmail.com (Jerone Young) Date: Sun, 29 May 2005 22:41:57 -0500 Subject: FC4 Gnome Large Mouse Curser only has effect on first login Message-ID: <9f50a7a005052920416e2984c6@mail.gmail.com> Choosing under Preferences->Mouse:Cursors you can choose to make your mouse cursor Small , Medium, or Large. Well it seems this only affects the first person who logs in. Once that person logs out, the next person to login or the same person who just logged in will nolonger have a large (or Medium) mouse cursor, it will stay the same small size untill you move the mouse over a application window (ex. firefox), but once you have the mouse back on the desktop it becomes very small. If you use high resolution monitors or have disabilities this can be a real pain. From kms at passback.co.uk Mon May 30 07:52:33 2005 From: kms at passback.co.uk (Keith Sharp) Date: Mon, 30 May 2005 08:52:33 +0100 Subject: How to debug a hung system? In-Reply-To: <20050529170307.70658.qmail@web54501.mail.yahoo.com> References: <20050529170307.70658.qmail@web54501.mail.yahoo.com> Message-ID: <1117439553.6474.13.camel@animal.passback.co.uk> On Sun, 2005-05-29 at 10:03 -0700, Frank Chen wrote: > I installed FC3 on a ASUS Terminator C3 PC. It has a > VIA C3 CPU. The system becomes completely unresponsive > from time to time. The only thing I could do was to do > a hard reboot. Any suggestions on how to debug such a > system? You will probably need to use a serial console to connect to the system and capture any output. See: http://www.tldp.org/HOWTO/Remote-Serial-Console-HOWTO/ and you may want to make grub use the serial console as well: http://www.rajeevnet.com/linux/grub_serial_console.html If you are using the system when this occurs then rebooting and looking is /var/log/messages for entries around the time of the problem may help. This sort of question is not appropriate for the Fedora Development List. You would be better asking on the Fedora List: http://www.redhat.com/mailman/listinfo/fedora-list Keith. From nscheibl at gmail.com Mon May 30 08:40:02 2005 From: nscheibl at gmail.com (Nicolas Scheibling) Date: Mon, 30 May 2005 10:40:02 +0200 Subject: Fedora 3 kernel 2.6.10-1.770_FC3 and nfsroot problem for PXE/diskless (CONFIG_ROOT_NFS=y) Message-ID: <429AD162.8010103@gmail.com> This work with a kernel from kernel.org (linux-2.6.11.10.tar.bz2) and exactly the same options set! I think Redhat/Fedora kernel patches corrupt the Network booting options. ------------------------------------------------------------------- hi all, im trying to setup a diskless environnement with Fedora. I use etherboot (as a bootable CDROM) to load linux kernel via PXE. I have a dhcp, nfs and a tftp server working properly. I have rebuild a 2.6.10-1.770_FC3 fedora kernel because latest 2.6.11 kernel series can't be build on my FC3. I have rebuild a kernel (as an rpm, or via make bzImage), with the NFSROOT option, to allow / mounting via NFS pxelinux.0 boot normaly, load the kernel and the initrd, but when the kernel try to moun the / partition, i obtain a kernel panic message: VFS: Cannont open root device "nfs" or unknow-block(0,255) Please append a correct "root=" boot option Kernel Panic - not syncing : VFS: Unable to mount root fs on unknow-block(0,255) After that, i have tried many other options in the linux kernel (using devfs instead udev, yes i know it's obsolete :)), and some other tips i found but no one worked. Note that on the NFS Server i don't see any packet using the NFS protocol. Is there some special configuration with Fedora kernels and nfsroot option ? Thanks for your suggest ? Here is my pxelinux.cfg/default file : kernel FC3_PXE/vmlinuz APPEND ramdisk_size=100000 root=/dev/ram0 nfsroot=192.168.1.14:/DOC/PXE/root or APPEND initrd=FC3_PXE/initrd.img ramdisk_size=10000 root=/dev/nfs nfsroot=192.168.1.14:/DOC/PXE/root ? Here is the main kernel option needed (if i don't miss one!) and that i have in my kernel: grep -i "IP_PNP\|NFS\|RAMFS" /DOC/PXE/root/boot/config-2.6.10-1.770_FC3nscheibl.20050525 CONFIG_INITRAMFS_SOURCE="" CONFIG_IP_PNP=y CONFIG_IP_PNP_DHCP=y CONFIG_IP_PNP_BOOTP=y CONFIG_IP_PNP_RARP=y CONFIG_RAMFS=y CONFIG_CRAMFS=m CONFIG_NFS_FS=y CONFIG_NFS_V3=y CONFIG_NFS_V4=y CONFIG_NFS_DIRECTIO=y CONFIG_NFSD=m CONFIG_NFSD_V3=y CONFIG_NFSD_V4=y CONFIG_NFSD_TCP=y CONFIG_ROOT_NFS=y CONFIG_NCPFS_NFS_NS=y ? Here is the config of the diskless client in my dhcp server (/etc/dhcpd.conf): group { next-server 192.168.1.14; filename "linux-install/pxelinux.0"; option root-path "192.168.1.14:/DOC/PXE/root,v3,tcp,hard"; host PXED { hardware ethernet 00:48:54:XX:YY:ZZ ; fixed-address 192.168.1.133; } } ? And here is my nfsd config (/etc/exports): /DOC/PXE/root *(ro,no_root_squash,sync) /DOC/PXE/snapshot *(rw,no_root_squash,sync) From hlprasu at yahoo.com Mon May 30 08:56:00 2005 From: hlprasu at yahoo.com (Prasad H.L.) Date: Mon, 30 May 2005 01:56:00 -0700 (PDT) Subject: Firstboot: Module to add other OS partitions' entry to fstab Message-ID: <20050530085600.53686.qmail@web30801.mail.mud.yahoo.com> Hi, I'm using Fedora Core 3. I've developed a firstboot python module which can detect other OS(currently only windows) partitions and add entries to the fstab. It also creates a user group which has access to those partitions. The version of firstboot is 1.3.33-2. I would like to know whether the firstboot developers are interested in seeing and using it in future releases of Fedora. If so, I'll send the module as an E-mail attachment. With regards, Prasad H.L. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From sundaram at redhat.com Mon May 30 09:00:02 2005 From: sundaram at redhat.com (Rahul Sundaram) Date: Mon, 30 May 2005 14:30:02 +0530 Subject: Firstboot: Module to add other OS partitions' entry to fstab In-Reply-To: <20050530085600.53686.qmail@web30801.mail.mud.yahoo.com> References: <20050530085600.53686.qmail@web30801.mail.mud.yahoo.com> Message-ID: <429AD612.5040507@redhat.com> Prasad H.L. wrote: >Hi, > >I'm using Fedora Core 3. > >I've developed a firstboot python module which can >detect other OS(currently only windows) partitions and >add entries to the fstab. It also creates a user group >which has access to those partitions. > >The version of firstboot is 1.3.33-2. > >I would like to know whether the firstboot developers >are interested in seeing and using it in future >releases of Fedora. If so, I'll send the module as an >E-mail attachment. > >With regards, >Prasad H.L. > > Please file it in bugzilla.redhat.com against firstboot with your module as an attachment regards Rahul From buildsys at redhat.com Mon May 30 11:32:27 2005 From: buildsys at redhat.com (Build System) Date: Mon, 30 May 2005 07:32:27 -0400 Subject: rawhide report: 20050530 changes Message-ID: <200505301132.j4UBWRv8006582@porkchop.devel.redhat.com> Updated Packages: (none) From seandarcy2 at gmail.com Mon May 30 15:04:06 2005 From: seandarcy2 at gmail.com (sean) Date: Mon, 30 May 2005 11:04:06 -0400 Subject: Firefox: You cannot print while in print preview Message-ID: I updated from development, now I keep getting this message in firefox whenever I try to print. How do I fix this? sean From byte at aeon.com.my Mon May 30 16:46:41 2005 From: byte at aeon.com.my (Colin Charles) Date: Tue, 31 May 2005 02:46:41 +1000 Subject: Firstboot: Module to add other OS partitions' entry to fstab In-Reply-To: <20050530085600.53686.qmail@web30801.mail.mud.yahoo.com> References: <20050530085600.53686.qmail@web30801.mail.mud.yahoo.com> Message-ID: <1117471601.3559.227.camel@arena.soho.bytebot.net> On Mon, 2005-05-30 at 01:56 -0700, Prasad H.L. wrote: > I've developed a firstboot python module which can > detect other OS(currently only windows) partitions and > add entries to the fstab. It also creates a user group > which has access to those partitions. I'm wondering if this is best done automatically in anaconda (the installer), rather than getting the user to do it manually during firstboot? Currently anaconda does this on PPC, and enters the OS X partition in yaboot automatically... > I would like to know whether the firstboot developers > are interested in seeing and using it in future > releases of Fedora. If so, I'll send the module as an > E-mail attachment. But yeah, as Rahul mentioned, Bugzilla it. -- Colin Charles, http://www.bytebot.net/ From byte at aeon.com.my Mon May 30 16:48:28 2005 From: byte at aeon.com.my (Colin Charles) Date: Tue, 31 May 2005 02:48:28 +1000 Subject: How to debug a hung system? In-Reply-To: <1117439553.6474.13.camel@animal.passback.co.uk> References: <20050529170307.70658.qmail@web54501.mail.yahoo.com> <1117439553.6474.13.camel@animal.passback.co.uk> Message-ID: <1117471708.3559.230.camel@arena.soho.bytebot.net> On Mon, 2005-05-30 at 08:52 +0100, Keith Sharp wrote: > > I installed FC3 on a ASUS Terminator C3 PC. It has a > > VIA C3 CPU. The system becomes completely unresponsive > > from time to time. The only thing I could do was to do > > a hard reboot. Any suggestions on how to debug such a > > system? > > You will probably need to use a serial console to connect to the > system > and capture any output. See: > > http://www.tldp.org/HOWTO/Remote-Serial-Console-HOWTO/ We build netconsole as a module, so having that loaded would be somewhat useful. Be warned of the UDP traffic. Read Documentation/networking/netconsole.txt from your kernel documentation for more information about it -- Colin Charles, http://www.bytebot.net/ From warren at togami.com Tue May 31 01:16:57 2005 From: warren at togami.com (Warren Togami) Date: Mon, 30 May 2005 15:16:57 -1000 Subject: Firefox: You cannot print while in print preview In-Reply-To: References: Message-ID: <429BBB09.7040306@togami.com> sean wrote: > I updated from development, now I keep getting this message in firefox > whenever I try to print. > > How do I fix this? > > sean > This list is only for discussion of development issues in the Fedora Project. THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION ONLY. Users should use one of these sites: FedoraForum.org Fedora @ LinuxQuestions.org Fedora Test Users http://www.redhat.com/mailman/listinfo/fedora-test-list or file a bug in Bugzilla From warren at togami.com Tue May 31 01:17:15 2005 From: warren at togami.com (Warren Togami) Date: Mon, 30 May 2005 15:17:15 -1000 Subject: FC4 Gnome Large Mouse Curser only has effect on first login In-Reply-To: <9f50a7a005052920416e2984c6@mail.gmail.com> References: <9f50a7a005052920416e2984c6@mail.gmail.com> Message-ID: <429BBB1B.7060502@togami.com> Jerone Young wrote: > Choosing under Preferences->Mouse:Cursors you can choose to make your > mouse cursor Small , Medium, or Large. Well it seems this only affects > the first person who logs in. Once that person logs out, the next > person to login or the same person who just logged in will nolonger > have a large (or Medium) mouse cursor, it will stay the same small > size untill you move the mouse over a application window (ex. > firefox), but once you have the mouse back on the desktop it becomes > very small. If you use high resolution monitors or have disabilities > this can be a real pain. > This list is only for discussion of development issues in the Fedora Project. THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION ONLY. Users should use one of these sites: FedoraForum.org Fedora @ LinuxQuestions.org Fedora Test Users http://www.redhat.com/mailman/listinfo/fedora-test-list or file a bug in Bugzilla From warren at togami.com Tue May 31 01:17:24 2005 From: warren at togami.com (Warren Togami) Date: Mon, 30 May 2005 15:17:24 -1000 Subject: FC4 Display GUI tool does not have 1920x1200 as selectable resolution In-Reply-To: <9f50a7a00505292022c087fa6@mail.gmail.com> References: <9f50a7a00505292022c087fa6@mail.gmail.com> Message-ID: <429BBB24.1070708@togami.com> Jerone Young wrote: > I just got a Samsung 243T with Nvidia Geforce 6600GT (this connection > is over DVI). When setting up FC4 , the highest resolution I can > choose is 1680x1050. One thing FC4 is doing is that it is using the > "vesa" driver by default for my card instead of the "nv" driver. But > even once I do select the nv driver in the GUI I still don't see > 1920x1200 as a selectable resolution. Going into xorg.conf and puting > 1920x1200 as a resolution solves the problem & I X works in 1920x1200. > > "lscpi -nv" outputf for my video card: > 01:00.0 Class 0300: 10de:00f1 (rev a2) > Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 5 > Memory at f8000000 (32-bit, non-prefetchable) [size=16M] > Memory at e0000000 (32-bit, prefetchable) [size=256M] > Memory at f9000000 (32-bit, non-prefetchable) [size=16M] > Capabilities: [60] Power Management version 2 > Capabilities: [44] AGP version 3.0 > This list is only for discussion of development issues in the Fedora Project. THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION ONLY. Users should use one of these sites: FedoraForum.org Fedora @ LinuxQuestions.org Fedora Test Users http://www.redhat.com/mailman/listinfo/fedora-test-list or file a bug in Bugzilla From wtogami at redhat.com Tue May 31 01:23:20 2005 From: wtogami at redhat.com (Warren Togami) Date: Mon, 30 May 2005 15:23:20 -1000 Subject: ANNOUNCE: Fedora Perl Devel List Message-ID: <429BBC88.9020504@redhat.com> http://www.redhat.com/mailman/listinfo/fedora-perl-devel-list Mission Statement: * Agreement and documentation of perl packaging standards * Cleanup and maintenance of Core and Extras perl packages * Discussion of perl package development issues * Progressive packaging of CPAN toward Extras THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR FEDORA PERL PACKAGE DEVELOPMENT DISCUSSION ONLY. IT IS INAPPROPRIATE TO ASK FOR PERL PROGRAMMING HELP HERE. Please join only if you intend on helping to fix perl packages in Fedora. Be prepared to filter the mail because all perl* bugmail is CC there. (And no I am not turning off bugmail. It is tolerable as long as it helps bugs to become fixed, which it does.) Warren Togami wtogami at redhat.com From green at redhat.com Tue May 31 02:20:38 2005 From: green at redhat.com (Anthony Green) Date: Mon, 30 May 2005 19:20:38 -0700 Subject: ANNOUNCE: Fedora Perl Devel List In-Reply-To: <429BBC88.9020504@redhat.com> References: <429BBC88.9020504@redhat.com> Message-ID: <1117506038.6092.14.camel@localhost.localdomain> On Mon, 2005-05-30 at 15:23 -1000, Warren Togami wrote: > http://www.redhat.com/mailman/listinfo/fedora-perl-devel-list My request for fedora-java-devel-list was transformed into fedora-devel- java-list by the time it actually got created. I don't know why, and I didn't think too much about it at the time. But having both fedora- perl-devel-list and fedora-devel-java-list is kind of odd. I don't suppose we can do anything about this now, however. AG From sopwith at redhat.com Mon May 30 20:14:43 2005 From: sopwith at redhat.com (Elliot Lee) Date: Mon, 30 May 2005 16:14:43 -0400 (EDT) Subject: Making FC-4 branches in CVS (Core & Extras) Message-ID: Just a heads-up that the branching of Fedora Core and Fedora Extras CVS repos is imminent. This means that the devel branch will start to be used for Fedora Core 5 & Fedora Extras for FC5, while the FC-4 branch will be the place to make updates targetted at FC4. This branching should happen sometime today, depending on the builds that are going on and how the FC4 release shapes up. -- Elliot From j.w.r.degoede at hhs.nl Tue May 31 07:05:44 2005 From: j.w.r.degoede at hhs.nl (Hans de Goede) Date: Tue, 31 May 2005 09:05:44 +0200 Subject: Firefox: You cannot print while in print preview In-Reply-To: <429BBB09.7040306@togami.com> References: <429BBB09.7040306@togami.com> Message-ID: <429C0CC8.9030006@hhs.nl> Aren't these more bug reports then support questions. He solved the X-problem himself, but the problem is that Fedora's X-tools don't know/list the resolution he has. I guess a polite note asking himto bugzilla this is more in place. Esp since he did provide additonal info like lspci output which would make this a usefull bugreport. So sean, please bugzilla this and other (reproducable) problems you have. Regards, Hans Warren Togami wrote: > sean wrote: > >> I updated from development, now I keep getting this message in firefox >> whenever I try to print. >> >> How do I fix this? >> >> sean >> > > This list is only for discussion of development issues in the Fedora > Project. > > THIS IS NOT A SUPPORT LIST. THIS LIST IS FOR CORE DEVELOPMENT DISCUSSION > ONLY. > > Users should use one of these sites: > FedoraForum.org > Fedora @ LinuxQuestions.org > Fedora Test Users http://www.redhat.com/mailman/listinfo/fedora-test-list > or file a bug in Bugzilla > From jss at ast.cam.ac.uk Tue May 31 11:23:12 2005 From: jss at ast.cam.ac.uk (Jeremy Sanders) Date: Tue, 31 May 2005 12:23:12 +0100 (BST) Subject: alsa-oss Message-ID: Is there any reason why the alsa-oss package isn't part of fedora? This is quite useful if you want to use the dmix alsa mixer and get oss apps to mix their output too. I've managed to hack together an updated version of the alsa-oss spec file (from livna) which appears to work. Jeremy -- Jeremy Sanders http://www-xray.ast.cam.ac.uk/~jss/ X-Ray Group, Institute of Astronomy, University of Cambridge, UK. Public Key Server PGP Key ID: E1AAE053 From d.lesca at solinos.it Tue May 31 11:27:55 2005 From: d.lesca at solinos.it (Dario Lesca) Date: Tue, 31 May 2005 13:27:55 +0200 Subject: How To debug why the server freezed Message-ID: <1117538875.5432.100.camel@localhost.localdomain> Hi, there is a method for know why the server during the normal work, without any register into /var/log/messages, freeze? I use a standard kernel 2.6.11-1.27_FC3smp on a i686 hp proliant 110 and I have rebuild and use a conexand driver for a USB ADSL Modem (http://accessrunner.sourceforge.net/driver.shtml) Also I use the linux-atm package ( linux-atm-2.5.0-0.20050118.2) The server work great for many hours, then freeze. Please, suggest me some kind of method for know what is the problem and what component caused the server freeze. Many thanks -- Dario Lesca From buildsys at redhat.com Tue May 31 11:43:08 2005 From: buildsys at redhat.com (Build System) Date: Tue, 31 May 2005 07:43:08 -0400 Subject: rawhide report: 20050531 changes Message-ID: <200505311143.j4VBh8SM017814@porkchop.devel.redhat.com> Updated Packages: GFS-kernel-2.6.11.7-20050519.153509.FC4.9 ----------------------------------------- cman-kernel-2.6.11.4-20050517.141233.FC4.11 ------------------------------------------- dlm-kernel-2.6.11.4-20050517.141233.FC4.11 ------------------------------------------ glibc-2.3.5-10 -------------- * Mon May 30 2005 Jakub Jelinek 2.3.5-10 - fix LD_ASSUME_KERNEL (since 2.3.5-8 GLRO(dl_osversion) has been always overwritten with the version of currently running kernel) - remove linuxthreads man pages other than those covered in 3p section, as 3p man pages are far better quality and describe POSIX behaviour that NPTL implements (#159084) gnbd-kernel-2.6.11.2-20050420.133124.FC4.33 ------------------------------------------- iptraf-2.7.0-15 --------------- * Tue May 17 2005 Karsten Hopp 2.7.0-15 - move config file to %{_sysconfdir}/iptraf.cfg to prevent deletion at bootup (#157794) * Tue May 10 2005 Karsten Hopp 2.7.0-14 - enable debuginfo kernel-2.6.11-1.1366_FC4 ------------------------ * Sun May 29 2005 Dave Jones - Fix slab corruption in firewire (#158424) * Fri May 27 2005 Dave Jones - remove non-cleanroom pwc driver compression. - Fix unintialised value in single bit error detector. (#158825) From notting at redhat.com Tue May 31 13:42:34 2005 From: notting at redhat.com (Bill Nottingham) Date: Tue, 31 May 2005 09:42:34 -0400 Subject: alsa-oss In-Reply-To: References: Message-ID: <20050531134234.GD8744@nostromo.devel.redhat.com> Jeremy Sanders (jss at ast.cam.ac.uk) said: > Is there any reason why the alsa-oss package isn't part of fedora? This is > quite useful if you want to use the dmix alsa mixer and get oss apps to > mix their output too. Honestly, IMO it seemed like quite a hack for Core use, especially since it would require per-app setup anyway. For Core I think the best thing to do is to make sure all apps are using ALSA natively. > I've managed to hack together an updated version of the alsa-oss spec file > (from livna) which appears to work. If you've got a working version, it would be great for Extras. Bill From ndbecker2 at gmail.com Tue May 31 13:47:11 2005 From: ndbecker2 at gmail.com (Neal Becker) Date: Tue, 31 May 2005 09:47:11 -0400 Subject: Mailman 2.1.6 released (includes security related updates) Message-ID: FYI From mbneto at gmail.com Tue May 31 13:57:32 2005 From: mbneto at gmail.com (mbneto) Date: Tue, 31 May 2005 09:57:32 -0400 Subject: Using fc4 tomcat/mod_jk in fc3 Message-ID: <5cf776b80505310657421156df@mail.gmail.com> Hi, I was wondering if anyone has already tried to use fc4's tomcat packages with fc3. In order to avoid upgrading a lot I'd like to rebuild the src.rpms but not sure which ones (besides the obvious tomcat5-5.0.30-5jpp_6fc.src.rpm) tks. From ndbecker2 at gmail.com Tue May 31 14:04:11 2005 From: ndbecker2 at gmail.com (Neal Becker) Date: Tue, 31 May 2005 10:04:11 -0400 Subject: lzo 2.00 released (important 64-bit speedups) Message-ID: http://freshmeat.net/projects/lzo/?branch_id=6066&release_id=197588 From overholt at redhat.com Tue May 31 14:23:02 2005 From: overholt at redhat.com (Andrew Overholt) Date: Tue, 31 May 2005 10:23:02 -0400 Subject: Using fc4 tomcat/mod_jk in fc3 In-Reply-To: <5cf776b80505310657421156df@mail.gmail.com> References: <5cf776b80505310657421156df@mail.gmail.com> Message-ID: <20050531142302.GA9252@redhat.com> * mbneto [2005-05-31 09:58]: > > I was wondering if anyone has already tried to use fc4's tomcat > packages with fc3. > > In order to avoid upgrading a lot I'd like to rebuild the src.rpms but > not sure which ones (besides the obvious > tomcat5-5.0.30-5jpp_6fc.src.rpm) You'll have to upgrade your toolchain to gcc 4.0.x. Andrew From mike at navi.cx Tue May 31 14:47:13 2005 From: mike at navi.cx (Mike Hearn) Date: Tue, 31 May 2005 15:47:13 +0100 Subject: alsa-oss References: <20050531134234.GD8744@nostromo.devel.redhat.com> Message-ID: On Tue, 31 May 2005 09:42:34 -0400, Bill Nottingham wrote: > Honestly, IMO it seemed like quite a hack for Core use, especially > since it would require per-app setup anyway. For Core I think the > best thing to do is to make sure all apps are using ALSA natively. Hack it is, but large numbers of apps still don't support ALSA properly. We need backwards compatibility with OSS in a way that doesn't avoid dmix, and alsa-oss provides that. As to whether it should be in Core or Extras, well, it's not a large program and it's the sort of thing that could end up being quite frequently used. Especially as nearly every commercial game for Linux uses OSS and not ALSA. But the criteria for core vs extras is a bit vague, I'm not sure where it should be ... but alsa-oss feels like operating system level software to me, as opposed to some app. thanks -mike From mjc at redhat.com Tue May 31 15:19:23 2005 From: mjc at redhat.com (Mark J Cox) Date: Tue, 31 May 2005 16:19:23 +0100 (BST) Subject: Mailman 2.1.6 released (includes security related updates) In-Reply-To: References: Message-ID: <0505311615510.21666@dell1.moose.awe.com> Note that 2.1.6 fixes CAN-2005-0202, but FC4 already contains a backported fix for this issue, and FC3 and FC2 were previously issued with updates (FEDORA-2005-131 and FEDORA-2005-132) (The XSS mentioned is CAN-2004-1177, fixed by FEDORA-2005-241, FEDORA-2005-242, and included by backport in FC4 already) Thanks, Mark -- Mark J Cox / Red Hat Security Response Team From dwmw2 at infradead.org Tue May 31 16:01:16 2005 From: dwmw2 at infradead.org (David Woodhouse) Date: Tue, 31 May 2005 17:01:16 +0100 Subject: How To debug why the server freezed In-Reply-To: <1117538875.5432.100.camel@localhost.localdomain> References: <1117538875.5432.100.camel@localhost.localdomain> Message-ID: <1117555276.13273.1.camel@hades.cambridge.redhat.com> On Tue, 2005-05-31 at 13:27 +0200, Dario Lesca wrote: > I use a standard kernel 2.6.11-1.27_FC3smp on a i686 hp proliant 110 and > I have rebuild and use a conexand driver for a USB ADSL Modem > (http://accessrunner.sourceforge.net/driver.shtml) > Also I use the linux-atm package ( linux-atm-2.5.0-0.20050118.2) > > The server work great for many hours, then freeze. > > Please, suggest me some kind of method for know what is the problem and > what component caused the server freeze. Do you have another computer near it? If so, the easiest answer is to hook up a serial console. Otherwise, switch to a text console and try to reproduce the problem -- you should probably see an oops on the screen. If it looks like it's the DSL modem driver, mail the report to usbatm at lists.infradead.org -- dwmw2 From jss at ast.cam.ac.uk Tue May 31 16:10:10 2005 From: jss at ast.cam.ac.uk (Jeremy Sanders) Date: Tue, 31 May 2005 17:10:10 +0100 (BST) Subject: alsa-oss In-Reply-To: References: Message-ID: On Tue, 31 May 2005, Jeremy Sanders wrote: > I've managed to hack together an updated version of the alsa-oss spec file > (from livna) which appears to work. Sorry for the lack of threading here, but I stupidly must have chosen plain text instead of MIME for my digests. I've included the alsa-oss spec file (which is just a slightly modified copy of the livna one). It appears to work with alsa-oss-1.0.9. I think it's pretty useful to have, because so many apps want to use OSS, especially non-free stuff. These apps include the flash plugin, games, skype and realplayer (does helix do alsa now?). Jeremy -- Jeremy Sanders http://www-xray.ast.cam.ac.uk/~jss/ X-Ray Group, Institute of Astronomy, University of Cambridge, UK. Public Key Server PGP Key ID: E1AAE053 -------------- next part -------------- Summary: OSS Emulation for ALSA. Name: alsa-oss Version: 1.0.9 Release: 2.xray Epoch: 0 License: GPL Group: System Environment/Libraries URL: http://www.alsa-project.org/ Source0: %{name}-%{version}.tar.bz2 BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-buildroot BuildRequires: alsa-lib-devel >= 0:0.9.0 %package devel Summary: Development package for %{name} Group: Development/Libraries Requires: %{name} = %{epoch}:%{version}-%{release} Requires: alsa-lib-devel # ----------------------------------------------------------------------------- %description This package contains an emulation/compatibility library which can make oss-linked program use ALSA instead. %description devel This package contains development files for %{name} # ----------------------------------------------------------------------------- %prep %setup -q # ----------------------------------------------------------------------------- %build %configure make %{?_smp_mflags} # ----------------------------------------------------------------------------- %install rm -rf $RPM_BUILD_ROOT %makeinstall # ----------------------------------------------------------------------------- %clean rm -rf $RPM_BUILD_ROOT # ----------------------------------------------------------------------------- %post -p /sbin/ldconfig %postun -p /sbin/ldconfig %files %defattr(-,root,root,-) %doc COPYING %{_bindir}/aoss %{_libdir}/libaoss.so.* %{_libdir}/libalsatoss.so.* %{_libdir}/libaoss.so %{_mandir}/man1/aoss.1* %files devel %defattr(-,root,root,-) %exclude %{_libdir}/lib*.la %{_includedir}/oss-redir.h %{_libdir}/libaoss.a %{_libdir}/libalsatoss.a %{_libdir}/libossredir.a # ----------------------------------------------------------------------------- %changelog * Tue May 31 2005 Jeremy Sanders 1.0.9-0.fdr.2 - Updated to 1.0.9 * Fri Oct 24 2003 Dams 0:0.9.8-0.fdr.1 - Updated to 0.9.8 * Fri Oct 3 2003 Dams 0:0.9.6-0.fdr.2 - Removed #-- marker after scriptlets * Wed Aug 13 2003 Dams - Initial build. From ivazquez at ivazquez.net Tue May 31 17:14:18 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Tue, 31 May 2005 13:14:18 -0400 Subject: alsa-oss In-Reply-To: References: Message-ID: <1117559658.6931.3.camel@ignacio.ignacio.lan> On Tue, 2005-05-31 at 12:23 +0100, Jeremy Sanders wrote: > Is there any reason why the alsa-oss package isn't part of fedora? This is > quite useful if you want to use the dmix alsa mixer and get oss apps to > mix their output too. With the pcm.!dsp target alsa-oss is obsolete. -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From ivazquez at ivazquez.net Tue May 31 17:16:46 2005 From: ivazquez at ivazquez.net (Ignacio Vazquez-Abrams) Date: Tue, 31 May 2005 13:16:46 -0400 Subject: lzo 2.00 released (important 64-bit speedups) In-Reply-To: References: Message-ID: <1117559806.6931.5.camel@ignacio.ignacio.lan> On Tue, 2005-05-31 at 10:04 -0400, Neal Becker wrote: > http://freshmeat.net/projects/lzo/?branch_id=6066&release_id=197588 http://bugzilla.redhat.com/ -- Ignacio Vazquez-Abrams http://fedora.ivazquez.net/ gpg --keyserver hkp://subkeys.pgp.net --recv-key 38028b72 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From fedora at leemhuis.info Tue May 31 17:44:07 2005 From: fedora at leemhuis.info (Thorsten Leemhuis) Date: Tue, 31 May 2005 19:44:07 +0200 Subject: alsa-oss In-Reply-To: <20050531134234.GD8744@nostromo.devel.redhat.com> References: <20050531134234.GD8744@nostromo.devel.redhat.com> Message-ID: <1117561447.5440.16.camel@notebook.thl.home> Am Dienstag, den 31.05.2005, 09:42 -0400 schrieb Bill Nottingham: > Jeremy Sanders (jss at ast.cam.ac.uk) said: > > I've managed to hack together an updated version of the alsa-oss spec file > > (from livna) which appears to work. > > If you've got a working version, it would be great for Extras. There was one in extras, it was removed after a short discussion: https://www.redhat.com/archives/fedora-devel-list/2005-February/msg00655.html I was the maintainer. The last version is here afaik: http://ftp-stud.fht-esslingen.de/pub/Mirrors/fedora/fedora/fedora/1/i386/SRPMS.stable/alsa-oss-1.0.4-0.fdr.1.1.src.rpm -- Thorsten Leemhuis From sopwith at redhat.com Tue May 31 18:51:04 2005 From: sopwith at redhat.com (Elliot Lee) Date: Tue, 31 May 2005 14:51:04 -0400 (EDT) Subject: CVS branching is done Message-ID: I've created FC-4 branches for all the Fedora Extras packages, and for all the packages included in Fedora Core 4. For distro CVS, the devel branch is targetting FC5 now and the FC-4 branch needs to be used for all FC4 updates. For extras CVS, the devel branch is for packages on top of rawhide (FC5-to-be), and the FC-4 branch needs to be used for the FE4 release/updates. Best, -- Elliot From sdohn at sdohn.com Tue May 31 19:57:29 2005 From: sdohn at sdohn.com (Stefan Dohn) Date: Tue, 31 May 2005 21:57:29 +0200 Subject: Fedora Rescue CD and Source - Checked by AntiVir - Message-ID: <000001c5661a$fcddbbe0$2d11a8c0@YAROW> I'm looking for the sourcecode of the fedora rescue cd. Can anyone tell me where I can find it, or who is the maintainer. The code for the rescue cd'S from redhat 7.3 to 9 was available under: ftp://ftp.redhat.de/pub/rh-addons/rescue-cd/ But where is the code for Fedora. From pnasrat at redhat.com Tue May 31 20:22:18 2005 From: pnasrat at redhat.com (Paul Nasrat) Date: Tue, 31 May 2005 16:22:18 -0400 Subject: Fedora Rescue CD and Source - Checked by AntiVir - In-Reply-To: <000001c5661a$fcddbbe0$2d11a8c0@YAROW> References: <000001c5661a$fcddbbe0$2d11a8c0@YAROW> Message-ID: <1117570938.3649.47.camel@enki.eridu> On Tue, 2005-05-31 at 21:57 +0200, Stefan Dohn wrote: > I'm looking for the sourcecode of the fedora rescue cd. > Can anyone tell me where I can find it, or who is the maintainer. > > The code for the rescue cd'S from redhat 7.3 to 9 was available under: > ftp://ftp.redhat.de/pub/rh-addons/rescue-cd/ > But where is the code for Fedora. See mk-rescueimage.i386 which is part of anaconda http://fedora.redhat.com/projects/anaconda-installer/ This article describes rescue CD creation http://fedoranews.org/contributors/gene_czarcinski/update_distro/ Paul