From burke at thealmquists.net Sun Jan 10 04:55:17 2010 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 9 Jan 2010 22:55:17 -0600 Subject: [K12OSN] Clients with 64MB of RAM Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm curious if any of you running LTSP on Fedora have had any luck booting clients with 64MB of RAM. It doesn't seem to be enough for my HP t5000, but my Dell GX1s with 128MB boot and run without a hitch. I've been using Fedora 12. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAktJXbUACgkQxWV7OPa/g5E0sACaA04ToJCnOnQaOOr2Uh1eFSKp JEwAnRld088q+TWsw9c/LA1WZTa9armI =Ypsu -----END PGP SIGNATURE----- From microman at cmosnetworks.com Mon Jan 11 00:08:15 2010 From: microman at cmosnetworks.com (Terrell Prude' Jr.) Date: Sun, 10 Jan 2010 19:08:15 -0500 Subject: [K12OSN] Clients with 64MB of RAM In-Reply-To: References: Message-ID: <4B4A6BEF.8000700@cmosnetworks.com> Almquist Burke wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I'm curious if any of you running LTSP on Fedora have had any luck > booting clients with 64MB of RAM. It doesn't seem to be enough for my > HP t5000, but my Dell GX1s with 128MB boot and run without a hitch. > I've been using Fedora 12. Depends more specifically on what your server's running. If you're running K12LTSP 4.2EL or 5.0EL, then 64MB will work. I still have one Pentium-133 thin client that does exactly that. However, others have reported that X11 may crash on such thin clients when running Firefox 3.0 for any extended period of time, due to the Firefox/X11 caching relationship, IIRC. The vast majority of my time is spent in OpenOffice.org and Konqueror (I run the KDE 3.5 desktop on 5.0EL), so maybe that's why I don't see this problem. So yes, it'll work, but DRAM's cheap, even the old single-data-rate PC100/133 SDRAM. If you're talking about DDR1, that's still very easy to find. --TP From burke at thealmquists.net Mon Jan 11 00:36:15 2010 From: burke at thealmquists.net (Almquist Burke) Date: Sun, 10 Jan 2010 18:36:15 -0600 Subject: [K12OSN] Clients with 64MB of RAM In-Reply-To: <4B4A6BEF.8000700@cmosnetworks.com> References: <4B4A6BEF.8000700@cmosnetworks.com> Message-ID: <638BB273-2285-43D5-9ACD-B201262B7DFC@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Unfortunately the RAM on the client can't be upgraded, it's right on the board. Guess I'll just have to stick with the old LTSP for that one. On Jan 10, 2010, at 6:08 PM, Terrell Prude' Jr. wrote: > Almquist Burke wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> I'm curious if any of you running LTSP on Fedora have had any luck >> booting clients with 64MB of RAM. It doesn't seem to be enough for >> my HP t5000, but my Dell GX1s with 128MB boot and run without a >> hitch. I've been using Fedora 12. > > Depends more specifically on what your server's running. If you're > running K12LTSP 4.2EL or 5.0EL, then 64MB will work. I still have > one Pentium-133 thin client that does exactly that. However, > others have reported that X11 may crash on such thin clients when > running Firefox 3.0 for any extended period of time, due to the > Firefox/X11 caching relationship, IIRC. The vast majority of my > time is spent in OpenOffice.org and Konqueror (I run the KDE 3.5 > desktop on 5.0EL), so maybe that's why I don't see this problem. > > So yes, it'll work, but DRAM's cheap, even the old single-data-rate > PC100/133 SDRAM. If you're talking about DDR1, that's still very > easy to find. > > --TP > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAktKcn8ACgkQxWV7OPa/g5HCPwCfR5BlOFmgfawaU1JKzwRWRt02 aLQAoIt2SdaUNuYK8BpXmAvICJ8T1iI+ =yMIU -----END PGP SIGNATURE----- From microman at cmosnetworks.com Mon Jan 11 01:10:53 2010 From: microman at cmosnetworks.com (Terrell Prude' Jr.) Date: Sun, 10 Jan 2010 20:10:53 -0500 Subject: [K12OSN] Clients with 64MB of RAM In-Reply-To: <638BB273-2285-43D5-9ACD-B201262B7DFC@thealmquists.net> References: <4B4A6BEF.8000700@cmosnetworks.com> <638BB273-2285-43D5-9ACD-B201262B7DFC@thealmquists.net> Message-ID: <4B4A7A9D.3060508@cmosnetworks.com> Well, if that's the case, I'd seriously look into replacing that thin client with another Optiplex GX1 or even an HP Vectra VL400, just on principle. And yes, those are *DIRT* cheap as well. --TP Almquist Burke wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Unfortunately the RAM on the client can't be upgraded, it's right on > the board. Guess I'll just have to stick with the old LTSP for that one. > > On Jan 10, 2010, at 6:08 PM, Terrell Prude' Jr. wrote: > >> Almquist Burke wrote: >>> -----BEGIN PGP SIGNED MESSAGE----- >>> Hash: SHA1 >>> >>> I'm curious if any of you running LTSP on Fedora have had any luck >>> booting clients with 64MB of RAM. It doesn't seem to be enough for >>> my HP t5000, but my Dell GX1s with 128MB boot and run without a >>> hitch. I've been using Fedora 12. >> >> Depends more specifically on what your server's running. If you're >> running K12LTSP 4.2EL or 5.0EL, then 64MB will work. I still have >> one Pentium-133 thin client that does exactly that. However, others >> have reported that X11 may crash on such thin clients when running >> Firefox 3.0 for any extended period of time, due to the Firefox/X11 >> caching relationship, IIRC. The vast majority of my time is spent in >> OpenOffice.org and Konqueror (I run the KDE 3.5 desktop on 5.0EL), so >> maybe that's why I don't see this problem. >> >> So yes, it'll work, but DRAM's cheap, even the old single-data-rate >> PC100/133 SDRAM. If you're talking about DDR1, that's still very >> easy to find. >> >> --TP >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (Darwin) > > iEYEARECAAYFAktKcn8ACgkQxWV7OPa/g5HCPwCfR5BlOFmgfawaU1JKzwRWRt02 > aLQAoIt2SdaUNuYK8BpXmAvICJ8T1iI+ > =yMIU > -----END PGP SIGNATURE----- > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From rowens at ptd.net Mon Jan 11 01:14:29 2010 From: rowens at ptd.net (Rob Owens) Date: Sun, 10 Jan 2010 20:14:29 -0500 Subject: [K12OSN] Clients with 64MB of RAM In-Reply-To: <638BB273-2285-43D5-9ACD-B201262B7DFC@thealmquists.net> References: <4B4A6BEF.8000700@cmosnetworks.com> <638BB273-2285-43D5-9ACD-B201262B7DFC@thealmquists.net> Message-ID: <20100111011429.GB12285@aurora.owens.net> I've run thin clients with 64MB of RAM on a Debian server with LTSP 5. It was kind of sluggish, but it worked. I didn't really use it for long, so there might be problems that I didn't notice. I haven't tried this on a K12Linux system, though. -Rob On Sun, Jan 10, 2010 at 06:36:15PM -0600, Almquist Burke wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Unfortunately the RAM on the client can't be upgraded, it's right on the > board. Guess I'll just have to stick with the old LTSP for that one. > > On Jan 10, 2010, at 6:08 PM, Terrell Prude' Jr. wrote: > >> Almquist Burke wrote: >>> -----BEGIN PGP SIGNED MESSAGE----- >>> Hash: SHA1 >>> >>> I'm curious if any of you running LTSP on Fedora have had any luck >>> booting clients with 64MB of RAM. It doesn't seem to be enough for >>> my HP t5000, but my Dell GX1s with 128MB boot and run without a >>> hitch. I've been using Fedora 12. >> >> Depends more specifically on what your server's running. If you're >> running K12LTSP 4.2EL or 5.0EL, then 64MB will work. I still have one >> Pentium-133 thin client that does exactly that. However, others have >> reported that X11 may crash on such thin clients when running Firefox >> 3.0 for any extended period of time, due to the Firefox/X11 caching >> relationship, IIRC. The vast majority of my time is spent in >> OpenOffice.org and Konqueror (I run the KDE 3.5 desktop on 5.0EL), so >> maybe that's why I don't see this problem. >> >> So yes, it'll work, but DRAM's cheap, even the old single-data-rate >> PC100/133 SDRAM. If you're talking about DDR1, that's still very easy >> to find. >> >> --TP >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (Darwin) > > iEYEARECAAYFAktKcn8ACgkQxWV7OPa/g5HCPwCfR5BlOFmgfawaU1JKzwRWRt02 > aLQAoIt2SdaUNuYK8BpXmAvICJ8T1iI+ > =yMIU > -----END PGP SIGNATURE----- > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From brcisna at eazylivin.net Tue Jan 12 03:39:40 2010 From: brcisna at eazylivin.net (Barry Cisna) Date: Mon, 11 Jan 2010 21:39:40 -0600 Subject: [K12OSN] Clients with 64MB of RAM Message-ID: <1263267580.5556.122.camel@localhost.localdomain> Almquist, When you say Fedora12 ,I assume you mean K12linux/LTSP5? 64MB will never come close to being enough ram for ltsp5. 128mb on the TC is not enough realistically. Just an FYI: I found the following,and this does not mean anything other than a baseline. I merely done to this to try and determine why ltsp5 is so much more memory intensive than ltsp4.x. >From boot to a login screen on ltsp 4.x about 65MB of data is transfered to the TC from ETH0 of the server. >From boot to a login screen on ltsp 5.x about 375MB of data is transfered to the TC from ETH0 of the server. I know this doesn't really tell anything other than there is over 5times the data difference,to get a login screen on the TC from the server between ltsp 4.x & ltsp 5.x,,,? My Ebox2300/128MB will work( but not really usable) on LTSP5 but takes almost 4-5 minutes to boot up. On ltsp4.x it takes about 30 seconds to boot to login screen,and works great on anything you throw at it(Even Youtube and any video format). Wish I was smart enough to understand what the underpinnings of LTSP5 makes it so much more a memory pig as compared to ltsp 4.x? The bootup,PXE is the same protocol for both so my guess is 'just the kernel' has more voodoo thrown into it that causes the overhead for the TC's? Take Care, Barry Cisna From burke at thealmquists.net Tue Jan 12 04:16:24 2010 From: burke at thealmquists.net (Almquist Burke) Date: Mon, 11 Jan 2010 22:16:24 -0600 Subject: [K12OSN] Clients with 64MB of RAM In-Reply-To: <1263267580.5556.122.camel@localhost.localdomain> References: <1263267580.5556.122.camel@localhost.localdomain> Message-ID: <8C383992-3DDE-45B2-B0EC-55C0022AAE2C@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Jan 11, 2010, at 9:39 PM, Barry Cisna wrote: > Almquist, > > When you say Fedora12 ,I assume you mean K12linux/LTSP5? Yes I am. > 64MB will never come close to being enough ram for ltsp5. 128mb on the > TC is not enough realistically. I'd say 128 MB is the new minimum for LTSP 5. Or, more accurately, LTSP 5 as implemented on newer versions of fedora. I guess I don't expect them to get it working on 64MB clients, I'm just hoping my GX1s aren't about to become unusable in Fedora 13 or 14. > Just an FYI: I found the following,and this does not mean anything > other > than a baseline. I merely done to this to try and determine why > ltsp5 is > so much more memory intensive than ltsp4.x. > > > My Ebox2300/128MB will work( but not really usable) on LTSP5 but takes > almost 4-5 minutes to boot up. My GX1s only have 128MB but they seem to boot much more quickly > On ltsp4.x it takes about 30 seconds to boot to login screen,and works > great on anything you throw at it(Even Youtube and any video format). > Wish I was smart enough to understand what the underpinnings of LTSP5 > makes it so much more a memory pig as compared to ltsp 4.x? > The bootup,PXE is the same protocol for both so my guess is 'just the > kernel' has more voodoo thrown into it that causes the overhead for > the > TC's? > From what I can tell, the additional burden of LTSP 5 comes mostly from using the distro's own build tools, and a tiny bit from things like increasing the size of existing code or new LTSP features. LTSP before version 5 was essentially it's own mini-distro and it contained the minimum amount of software needed to boot up a thin client and get and working X session and stuff like sound and local devices. You could add it into any distro by putting it in a folder, exporting it as a read-only share (NFS usually) and setting up the needed DHCP/TFTP/NFS services plus some other optional ones. K12LTSP did this work for us and also bundled useful education apps. But maintaining a distro was a lot of work and the LTSP developers didn't feel they were adding a lot of value. They had to spend a lot of time on things that every distribution works on, and had much less time to work on the LTSP specific code. Thus with LTSP 5 they moved to a system where they only maintain certain LTSP specific code and each distribution has to integrate LTSP. Basically, the tradeoff that LTSP made was to make running local applications, sound, and supporting more hardware easier. But at the cost of requiring beefier clients. Whether you like this tradeoff depends on your pool of old clients that don't work well with LTSP 5 and your desire for the features that LTSP 5 added. > Take Care, > Barry Cisna > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAktL95gACgkQxWV7OPa/g5EvLACeK6AVU1ZyqNk1s8f+Z1PBnFqT EdQAn0oz0/D5h8jg3vIeGt4ZCtopF3Ep =qvjF -----END PGP SIGNATURE----- From microman at cmosnetworks.com Tue Jan 12 04:21:56 2010 From: microman at cmosnetworks.com (Terrell Prude' Jr.) Date: Mon, 11 Jan 2010 23:21:56 -0500 Subject: [K12OSN] Clients with 64MB of RAM In-Reply-To: <8C383992-3DDE-45B2-B0EC-55C0022AAE2C@thealmquists.net> References: <1263267580.5556.122.camel@localhost.localdomain> <8C383992-3DDE-45B2-B0EC-55C0022AAE2C@thealmquists.net> Message-ID: <4B4BF8E4.1060107@cmosnetworks.com> Almquist Burke wrote: > Basically, the tradeoff that LTSP made was to make running local > applications, sound, and supporting more hardware easier. But at the > cost of requiring beefier clients. Whether you like this tradeoff > depends on your pool of old clients that don't work well with LTSP 5 > and your desire for the features that LTSP 5 added. Sounds like a great reason to stick with the LTSP 4.x based K12LTSP 5EL for a good long time. I know I will be. :-) --TP From einfeldt at gmail.com Wed Jan 13 03:53:48 2010 From: einfeldt at gmail.com (Christian Einfeldt) Date: Tue, 12 Jan 2010 19:53:48 -0800 Subject: [K12OSN] big FOSS coding push for education Message-ID: <4b5781041001121953r87a35e1i3b984ec8ddf806cc@mail.gmail.com> hi, Anyone here know how to get in touch with Paul Nelson or Steve Hargadon? Brian Behlendorf just pinged me saying that he and a couple other people are organizing a coding push for FOSS for education. I am mostly involved with FOSS in four San Francisco Bay Area schools, and so I don't really know the bigger picture. I think that Steve and Paul probably have the bigger picture on whom to contact. Brian just happened to contact me because he is not normally in the education field, but he knows that I am doing that kind of stuff locally here. If there is someone who could help Brian with info as to whom he should contact for a national and international coding push, please do ping me either in this thread or directly at einfeldt @ gmail dot com. Unfortunately, time is an issue here. Thx for considering my request, and sorry that I am so short on details. -- Christian Einfeldt, Producer, The Digital Tipping Point -------------- next part -------------- An HTML attachment was scrubbed... URL: From tom.hoffman at gmail.com Wed Jan 13 13:50:35 2010 From: tom.hoffman at gmail.com (Tom Hoffman) Date: Wed, 13 Jan 2010 08:50:35 -0500 Subject: [K12OSN] big FOSS coding push for education In-Reply-To: <4b5781041001121953r87a35e1i3b984ec8ddf806cc@mail.gmail.com> References: <4b5781041001121953r87a35e1i3b984ec8ddf806cc@mail.gmail.com> Message-ID: <92de6c881001130550t47adea83hcbe2774abd9e5610@mail.gmail.com> What is a coding push? --Tom On Tue, Jan 12, 2010 at 10:53 PM, Christian Einfeldt wrote: > hi, > > Anyone here know how to get in touch with Paul Nelson or Steve Hargadon? > Brian Behlendorf just pinged me saying that he and a couple other people are > organizing a coding push for FOSS for education.? I am mostly involved with > FOSS in four San Francisco Bay Area schools, and so I don't really know the > bigger picture.? I think that Steve and Paul probably have the bigger > picture on whom to contact.? Brian just happened to contact me because he is > not normally in the education field, but he knows that I am doing that kind > of stuff locally here. > > If there is someone who could help Brian with info as to whom he should > contact for a national and international coding push, please do ping me > either in this thread or directly at einfeldt @ gmail dot com. > Unfortunately, time is an issue here.? Thx for considering my request, and > sorry that I am so short on details. > > -- > Christian Einfeldt, > Producer, The Digital Tipping Point > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From einfeldt at gmail.com Wed Jan 13 21:43:34 2010 From: einfeldt at gmail.com (Christian Einfeldt) Date: Wed, 13 Jan 2010 13:43:34 -0800 Subject: [K12OSN] big FOSS coding push for education In-Reply-To: <92de6c881001130550t47adea83hcbe2774abd9e5610@mail.gmail.com> References: <4b5781041001121953r87a35e1i3b984ec8ddf806cc@mail.gmail.com> <92de6c881001130550t47adea83hcbe2774abd9e5610@mail.gmail.com> Message-ID: <4b5781041001131343t618019bfk23a27be04051afd6@mail.gmail.com> Hi, On Wed, Jan 13, 2010 at 5:50 AM, Tom Hoffman wrote: > What is a coding push? > Not sure. I am a relatively simple end user who is active in bringing Linux computers to San Francisco Bay Area schools, where Brian lives. Brian knows that I am active in education with FOSS, and so he contacted me in a huge hurry. Brian talks really fast when he is excited about something, and he was using language that I didn't entirely understand. So I am sorry, I don't have any details. Bottom line is if you are interested in FOSS and education, please ping me. Again, my apologies for lack of detail. -------------- next part -------------- An HTML attachment was scrubbed... URL: From news at siddall.name Thu Jan 14 20:57:38 2010 From: news at siddall.name (Jeff Siddall) Date: Thu, 14 Jan 2010 15:57:38 -0500 Subject: [K12OSN] Fresh install of F12 w/ updates and problem running ltsp-build-client In-Reply-To: <1259775765.22126.121.camel@buran.hhp.hhprep.org> References: <1259775765.22126.121.camel@buran.hhp.hhprep.org> Message-ID: <4B4F8542.2090209@siddall.name> Burroughs, Henry wrote: > Dan, > > I followed norbert's steps, they didn't make a difference. I however > did get the rebuild to work on a fresh system boot without turning > "setenforce 0". Dumb me and I deleted the i386 directory to try again > with "setenforce 0" now it doesn't work. This is what is left over in > mount: > > /opt/ltsp/i386 on /var/tmp/imgcreate-cSSQL8/install_root type none (rw,bind) > > I am going to reboot the server again and see if it works again. Anyone have success with F12 and ltsp-build-client yet? I have a freshly upgraded F12 server which is fully updated, but ltsp-build-client _never_ finishes without this umount error. I have tried setenforce 0, clean reboot, no setenforce 0 -- nothing makes any difference. Every time it dies /opt/ltsp/i386 stays mounted as /var/tmp/imgcreate-something/install_root but it never complains when I try to umount it. I did some digging in the scripts by only found one script that did a umount. I added an echo statement above it to make sure I was in the right spot but either I was not in the correct spot or stdout is going to the bit bucket because I don't see anything printed to the screen. Anyone have any tips on how to troubleshoot this? Thanks, Jeff From burke at thealmquists.net Fri Jan 15 03:59:56 2010 From: burke at thealmquists.net (Almquist Burke) Date: Thu, 14 Jan 2010 21:59:56 -0600 Subject: [K12OSN] Fresh install of F12 w/ updates and problem running ltsp-build-client In-Reply-To: <4B4F8542.2090209@siddall.name> References: <1259775765.22126.121.camel@buran.hhp.hhprep.org> <4B4F8542.2090209@siddall.name> Message-ID: <025BE6C0-9CDD-4FB0-BAE4-D91C055ECFEC@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Jan 14, 2010, at 2:57 PM, Jeff Siddall wrote: > > Anyone have success with F12 and ltsp-build-client yet? > I didn't have any problems with it. Are you working from a fresh install? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAktP6DwACgkQxWV7OPa/g5H3nACgi8QaGnRSNw3QDBuUh1PKUGX6 4/sAoIPUeQdayDYywNdTvijbnOXZIAmF =eiS9 -----END PGP SIGNATURE----- From news at siddall.name Fri Jan 15 15:01:16 2010 From: news at siddall.name (Jeff Siddall) Date: Fri, 15 Jan 2010 10:01:16 -0500 Subject: [K12OSN] Fresh install of F12 w/ updates and problem running ltsp-build-client In-Reply-To: <025BE6C0-9CDD-4FB0-BAE4-D91C055ECFEC@thealmquists.net> References: <1259775765.22126.121.camel@buran.hhp.hhprep.org> <4B4F8542.2090209@siddall.name> <025BE6C0-9CDD-4FB0-BAE4-D91C055ECFEC@thealmquists.net> Message-ID: <4B50833C.2070801@siddall.name> Almquist Burke wrote: > > On Jan 14, 2010, at 2:57 PM, Jeff Siddall wrote: > >> Anyone have success with F12 and ltsp-build-client yet? > > > I didn't have any problems with it. Are you working from a fresh install? No, this is an upgrade from an F10 K12Linux install. The script was flaky there too, though occasionally it did complete successfully. Maybe a timing issue or something? I dunno. I looked into the scripts some more and it appears the umount is not being called directly in one of the LTSP scripts. It appears it is happening inside one of the python modules. Great. FWIW I did manage to reproduce the problem manually. If I kill the ltsp-build-client script after it has created the chroot but before it barfs at the umount I can see that inside /var/tmp/imgcreate-blah/install_root a bunch of sub directories are also mounted. If I try to umount /opt/ltsp/i386 it fails with the same "busy" message as the script. I think calling a lazy umount, or calling it twice in a row would fix the issue. I just don't know where to make the change. Any tips about how to find this bug would be great. Alternatively, does anyone know what the ltsp-build-client script does once all the packages are installed? Maybe I can just manually do the remaining steps. Jeff From dyoung at mesd.k12.or.us Fri Jan 15 16:11:50 2010 From: dyoung at mesd.k12.or.us (Dan Young) Date: Fri, 15 Jan 2010 08:11:50 -0800 Subject: [K12OSN] Fresh install of F12 w/ updates and problem running ltsp-build-client In-Reply-To: <4B50833C.2070801@siddall.name> References: <1259775765.22126.121.camel@buran.hhp.hhprep.org> <4B4F8542.2090209@siddall.name> <025BE6C0-9CDD-4FB0-BAE4-D91C055ECFEC@thealmquists.net> <4B50833C.2070801@siddall.name> Message-ID: <994441ae1001150811y5b2494dajc91220b3163accc9@mail.gmail.com> On Fri, Jan 15, 2010 at 7:01 AM, Jeff Siddall wrote: > Almquist Burke wrote: >> >> On Jan 14, 2010, at 2:57 PM, Jeff Siddall wrote: >> >>> Anyone have success with F12 and ltsp-build-client yet? >> >> >> I didn't have any problems with it. Are you working from a fresh install? > > No, this is an upgrade from an F10 K12Linux install. ?The script was > flaky there too, though occasionally it did complete successfully. > Maybe a timing issue or something? ?I dunno. Can you "rpm -V ltsp-server"? Maybe your kickstart file that the ltsp client builds from is goobered. -- Dan Young Multnomah ESD - Technology Services 503-257-1562 From news at siddall.name Fri Jan 15 21:10:11 2010 From: news at siddall.name (Jeff Siddall) Date: Fri, 15 Jan 2010 16:10:11 -0500 Subject: [K12OSN] Fresh install of F12 w/ updates and problem running ltsp-build-client In-Reply-To: <994441ae1001150811y5b2494dajc91220b3163accc9@mail.gmail.com> References: <1259775765.22126.121.camel@buran.hhp.hhprep.org> <4B4F8542.2090209@siddall.name> <025BE6C0-9CDD-4FB0-BAE4-D91C055ECFEC@thealmquists.net> <4B50833C.2070801@siddall.name> <994441ae1001150811y5b2494dajc91220b3163accc9@mail.gmail.com> Message-ID: <4B50D9B3.8010100@siddall.name> Dan Young wrote: > On Fri, Jan 15, 2010 at 7:01 AM, Jeff Siddall wrote: >> Almquist Burke wrote: >>> On Jan 14, 2010, at 2:57 PM, Jeff Siddall wrote: >>> >>>> Anyone have success with F12 and ltsp-build-client yet? >>> >>> I didn't have any problems with it. Are you working from a fresh install? >> No, this is an upgrade from an F10 K12Linux install. The script was >> flaky there too, though occasionally it did complete successfully. >> Maybe a timing issue or something? I dunno. > > Can you "rpm -V ltsp-server"? Maybe your kickstart file that the ltsp > client builds from is goobered. rpm -V ltsp-server shows: S.5....T. c /etc/ltsp/ltsp-build-client.conf S.5....T. c /etc/xinetd.d/nbdrootd S.5....T. c /etc/xinetd.d/nbdswapd .......T. /usr/sbin/ltsp-build-client S.5....T. c /var/lib/tftpboot/ltsp/i386/lts.conf I changed the cache setting in /etc/ltsp/ltsp-build-client.conf, I have a customized /var/lib/tftpboot/ltsp/i386/lts.conf and I tried some debug commands in /usr/sbin/ltsp-build-client so those are expected. I am not sure why the other files are different. rpm -qf /usr/sbin/ltsp-build-client revealed: ltsp-server-5.1.95-1.fc12.i686 So that all looks OK. Any other ideas? Jeff From einfeldt at gmail.com Sat Jan 16 01:26:18 2010 From: einfeldt at gmail.com (Christian Einfeldt) Date: Fri, 15 Jan 2010 17:26:18 -0800 Subject: [K12OSN] big FOSS coding push for education In-Reply-To: <4b5781041001121953r87a35e1i3b984ec8ddf806cc@mail.gmail.com> References: <4b5781041001121953r87a35e1i3b984ec8ddf806cc@mail.gmail.com> Message-ID: <4b5781041001151726s2253e98i73674b2a3995b18d@mail.gmail.com> Hi, On 1/12/10, Christian Einfeldt wrote: > > hi, > > Anyone here know how to get in touch with Paul Nelson or Steve Hargadon? > Brian Behlendorf just pinged me saying that he and a couple other people are > organizing a coding push for FOSS for education. I had completely misunderstood Brian when he and I talked on the phone. This is what he was referring to. http://blog.ostp.gov/2010/01/15/demonstrate-a-technologist%E2%80%99s-spirit-of-generosity-on-martin-luther-king-day/ Essentially, the US CTO Asheesh Chopra is trying to get people to volunteer for tech work in their communities. He is kicking off his push on MLK day. Unfortunately, the earthquake in Haiti totally scrambled his schedule, and so Chopra just got this notice posted today. -------------- next part -------------- An HTML attachment was scrubbed... URL: From HBurroughs at hhprep.org Sun Jan 17 17:05:16 2010 From: HBurroughs at hhprep.org (Burroughs, Henry) Date: Sun, 17 Jan 2010 12:05:16 -0500 Subject: [K12OSN] Fixing F12 ltsp-build-client Message-ID: <3437EBC2F7B463439E7CD8796349DCF25FF6C1@enterprise.hhp.hhprep.org> Jeff and Friends, disable the nscd service (ie: service nscd stop) There is an issue with that and some timeout problems... and then ltsp-build-client dumps out. That hit me on my production but not test build of F12 ltsp. Henry Burroughs Technology Director Hilton Head Preparatory School www.hhprep.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From news at siddall.name Wed Jan 20 14:54:35 2010 From: news at siddall.name (Jeff Siddall) Date: Wed, 20 Jan 2010 09:54:35 -0500 Subject: [K12OSN] SOLVED: Fresh install of F12 w/ updates and problem running ltsp-build-client In-Reply-To: <4B50D9B3.8010100@siddall.name> References: <1259775765.22126.121.camel@buran.hhp.hhprep.org> <4B4F8542.2090209@siddall.name> <025BE6C0-9CDD-4FB0-BAE4-D91C055ECFEC@thealmquists.net> <4B50833C.2070801@siddall.name> <994441ae1001150811y5b2494dajc91220b3163accc9@mail.gmail.com> <4B50D9B3.8010100@siddall.name> Message-ID: <4B57192B.2050107@siddall.name> On 01/15/2010 04:10 PM, Jeff Siddall wrote: > Dan Young wrote: >> On Fri, Jan 15, 2010 at 7:01 AM, Jeff Siddall wrote: >>> Almquist Burke wrote: >>>> On Jan 14, 2010, at 2:57 PM, Jeff Siddall wrote: >>>> >>>>> Anyone have success with F12 and ltsp-build-client yet? >>>> >>>> I didn't have any problems with it. Are you working from a fresh install? >>> No, this is an upgrade from an F10 K12Linux install. The script was >>> flaky there too, though occasionally it did complete successfully. >>> Maybe a timing issue or something? I dunno. >> >> Can you "rpm -V ltsp-server"? Maybe your kickstart file that the ltsp >> client builds from is goobered. > > rpm -V ltsp-server > > shows: > > S.5....T. c /etc/ltsp/ltsp-build-client.conf > S.5....T. c /etc/xinetd.d/nbdrootd > S.5....T. c /etc/xinetd.d/nbdswapd > .......T. /usr/sbin/ltsp-build-client > S.5....T. c /var/lib/tftpboot/ltsp/i386/lts.conf > > I changed the cache setting in /etc/ltsp/ltsp-build-client.conf, I have > a customized /var/lib/tftpboot/ltsp/i386/lts.conf and I tried some debug > commands in /usr/sbin/ltsp-build-client so those are expected. I am not > sure why the other files are different. > > rpm -qf /usr/sbin/ltsp-build-client > > revealed: > > ltsp-server-5.1.95-1.fc12.i686 > > So that all looks OK. > > Any other ideas? UPDATE: Henry suggested I disable nscd while running ltsp-build-client and it worked! There were some other errors like: sh: /usr/sbin/setenforce: No such file or directory and sed: warning: failed to set default file creation context to unconfined_u:object_r:usr_t:s0: No such file or directory... but the bottom line is: info: LTSP client installation completed successfully I'll try booting it on a client sometime this week and make sure everything is OK. Jeff From news at siddall.name Thu Jan 21 01:20:35 2010 From: news at siddall.name (Jeff Siddall) Date: Wed, 20 Jan 2010 20:20:35 -0500 Subject: [K12OSN] SOLVED: Fresh install of F12 w/ updates and problem running ltsp-build-client In-Reply-To: <4B57192B.2050107@siddall.name> References: <1259775765.22126.121.camel@buran.hhp.hhprep.org> <4B4F8542.2090209@siddall.name> <025BE6C0-9CDD-4FB0-BAE4-D91C055ECFEC@thealmquists.net> <4B50833C.2070801@siddall.name> <994441ae1001150811y5b2494dajc91220b3163accc9@mail.gmail.com> <4B50D9B3.8010100@siddall.name> <4B57192B.2050107@siddall.name> Message-ID: <4B57ABE3.3040205@siddall.name> On 01/20/2010 09:54 AM, Jeff Siddall wrote: > UPDATE: Henry suggested I disable nscd while running ltsp-build-client > and it worked! There were some other errors like: > > sh: /usr/sbin/setenforce: No such file or directory > > and > > sed: warning: failed to set default file creation context to > unconfined_u:object_r:usr_t:s0: No such file or directory... > > but the bottom line is: > > info: LTSP client installation completed successfully > > I'll try booting it on a client sometime this week and make sure > everything is OK. I tested the new image with a thin client and everything seems to work properly. Jeff From jthomas at bittware.com Mon Jan 25 13:59:50 2010 From: jthomas at bittware.com (j.w. thomas) Date: Mon, 25 Jan 2010 08:59:50 -0500 Subject: [K12OSN] Article of Interest on Slashdot Message-ID: <4B5DA3D6.7000400@bittware.com> Here's an article that I suspect this group will find somewhat interesting. http://linux.slashdot.org/story/10/01/25/0230231/NZ-School-Goes-Open-Source-Amid-Microsoft-Mandate?art_pos=5 -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 I thought I was wrong once, but I was mistaken. From pauls at tclcommunications.co.nz Mon Jan 25 18:41:53 2010 From: pauls at tclcommunications.co.nz (Paul Satherley) Date: Tue, 26 Jan 2010 07:41:53 +1300 Subject: [K12OSN] Article of Interest on Slashdot In-Reply-To: <4B5DA3D6.7000400@bittware.com> References: <4B5DA3D6.7000400@bittware.com> Message-ID: <4B5DE5F1.9030807@tclcommunications.co.nz> On 26/01/10 02:59, j.w. thomas wrote: > Here's an article that I suspect this group will find somewhat > interesting. > > http://linux.slashdot.org/story/10/01/25/0230231/NZ-School-Goes-Open-Source-Amid-Microsoft-Mandate?art_pos=5 > > As another example, Because the Internet speed of one of my sites that I look after The Ministry of Edu Told a school it was the Linux servers that were responsible so the ONE server doing Mail, File, Web Apps, Firewall etc Was blown away in came A Exchange, SQl, Proxy, AD and of course a hardware firewall to make it all "safe"..all for $40,000 Plus 600 licensees and Guess what the Internet speed is slower, The real reason is the Mind Control China style filtering. Oh of course Global Warming is a mandatory subject Cheers