From burke at thealmquists.net Wed Apr 1 04:14:42 2009 From: burke at thealmquists.net (Almquist Burke) Date: Tue, 31 Mar 2009 23:14:42 -0500 Subject: [K12OSN] Custom xorg.conf In-Reply-To: References: Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mar 31, 2009, at 3:32 PM, David Hopkins wrote: > With FC10 and ltsp5, how do I get X -configure to run at the client to > generate an xorg.conf file that I can then use to force certain > settings? In particular, I need to force 1024/x768 resolution at > 75Hz. The autodetect sets the screens to 1280x1024 at 60Hz, but the > preferred resolution of the monitor is the lower setting. I have > tried simply running X -configure (via using a shell at the client) > and I get that the file //xorg.conf.new can't be located. For that > matter, is an xorg.conf file created automatically at client boot? If > so, I can't find that file either. I do have the xorg.log files if > needed. > Can you specify a resolution in the lts.conf file? > Sincerely, > Dave Hopkins > > _______________________________________________ > 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) iEYEARECAAYFAknS6jIACgkQxWV7OPa/g5HKVwCfRQ+RRxPZmK/i58NQytAfTYZd MhcAnRd+Qlizuodi/lqJHTyogYNQncwd =O9n5 -----END PGP SIGNATURE----- From tonythaaloca at yahoo.com Wed Apr 1 09:40:18 2009 From: tonythaaloca at yahoo.com (mbiatem joben) Date: Wed, 1 Apr 2009 02:40:18 -0700 (PDT) Subject: [K12OSN] help for linux friends Message-ID: <893870.22425.qm@web39501.mail.mud.yahoo.com> Thanks ;-) --- On Tue, 3/31/09, M Rathburn wrote: From: M Rathburn Subject: RE: [K12OSN] help for linux friends To: "'Support list for open source software in schools.'" Date: Tuesday, March 31, 2009, 8:24 AM Start here:? www.linuxfoundation.org _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see -------------- next part -------------- An HTML attachment was scrubbed... URL: From dahopkins429 at gmail.com Wed Apr 1 12:35:26 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 1 Apr 2009 08:35:26 -0400 Subject: [K12OSN] Custom xorg.conf In-Reply-To: References: Message-ID: > Can you specify a resolution in the lts.conf file? Supposedly, but I tried XMODE and that didn't change anything ... unless I'm looking at the wrong lts.conf? I'm modifying the one in tftpboot/i386. There is also an lts.conf in /opt/ltsp/i386/etc. And, finally, with ltsp4.2, there was an /etc/hosts file with entries for all the thin clients. with ltsp5 that has also changed? Sort of a change in topic, but I'll check tonight to see if the client knows its name. From whatch at anwsu.org Wed Apr 1 14:41:35 2009 From: whatch at anwsu.org (Will Hatch) Date: Wed, 01 Apr 2009 10:41:35 -0400 Subject: [K12OSN] off topic - sound redirection and XDMCP Message-ID: <49D3358A.0948.00D4.0@anwsu.org> I haven't been able to find a good answer to this issue, so I though the wizards at the k12ltsp list might know the answer. I have my devonIT thin clients connecting to my Edubuntu server via XDMCP. All is well with the graphics, but sound does not work. As a result, I think some programs such as tux type will not work because they require sound. Either way, it would be nice to have sound if the kids want to play any multimedia. Otherwise, they will have to connect to the Windows network! Nooooooooo!!! Any advice appreciated. *********************************** PRIVILEGED AND CONFIDENTIAL: This communication, including attachments, is for the exclusive use of addressee and may contain proprietary, confidential or privileged information. If you are not the intended recipient, any use, copying, disclosure, dissemination or distribution is strictly prohibited. If you're not the intended recipient, please notify the sender immediately by return email and delete this communication and destroy all copies. It is the policy of ANWSU not to discriminate on the basis or race, color, religion, national origin, gender, disability, or gender orientation in its educational programs or activities, or in its employment policies as required by Title IX of the 1972 Educational Amendments, by Section 504 of the Rehabilitation Act of 1973, by Title VI of the Civil Rights Act of 1964, and by Vermont State Law. From acerothstein at hotmail.fr Wed Apr 1 15:53:19 2009 From: acerothstein at hotmail.fr (thomas Thomas) Date: Wed, 1 Apr 2009 15:53:19 +0000 Subject: [K12OSN] Problem with USB Stick (NTFS) Message-ID: Hi all, I'm Ace, and i'm a french user of K12Linux :) I have a server LTSP with Fedora 10. And,i have a problem with NTFS USB Stick. When an USB Stick is mounted on a client, a folder appears on the client's desktop, but is empty.... However, when the USB stick is mounted on the server, his folder appears on the desktop's server, but the folder is readable ! I'm new user of LTSP, and i did'nt know where to search. Thanks Linux, ya moins bien, mais c'est plus cher.... _________________________________________________________________ Vous voulez savoir ce que vous pouvez faire avec le nouveau Windows Live?? Lancez-vous ! http://www.microsoft.com/windows/windowslive/default.aspx -------------- next part -------------- An HTML attachment was scrubbed... URL: From rowens at ptd.net Wed Apr 1 20:37:30 2009 From: rowens at ptd.net (Rob Owens) Date: Wed, 1 Apr 2009 16:37:30 -0400 Subject: [K12OSN] off topic - sound redirection and XDMCP In-Reply-To: <49D3358A.0948.00D4.0@anwsu.org> References: <49D3358A.0948.00D4.0@anwsu.org> Message-ID: <20090401203730.GB8443@aurora.owens.net> I think you need to PXE boot (and therefore utilize LTSP) in order to get sound. I don't think straight XDMCP will get you sound. Although, having used Tux Type, I don't think sound is critical to its performance. have you tried? -Rob On Wed, Apr 01, 2009 at 10:41:35AM -0400, Will Hatch wrote: > I haven't been able to find a good answer to this issue, so I though the wizards at the k12ltsp list might know the answer. > > I have my devonIT thin clients connecting to my Edubuntu server via XDMCP. All is well with the graphics, but sound does not work. As a result, I think some programs such as tux type will not work because they require sound. Either way, it would be nice to have sound if the kids want to play any multimedia. Otherwise, they will have to connect to the Windows network! Nooooooooo!!! > > Any advice appreciated. > > > *********************************** > PRIVILEGED AND CONFIDENTIAL: This communication, including attachments, is for the exclusive use of addressee and may contain proprietary, confidential or privileged information. If you are not the intended recipient, any use, copying, disclosure, dissemination or distribution is strictly prohibited. If you're not the intended recipient, please notify the sender immediately by return email and delete this communication and destroy all copies. > It is the policy of ANWSU not to discriminate on the basis or race, color, religion, national origin, gender, disability, or gender orientation in its educational programs or activities, or in its employment policies as required by Title IX of the 1972 Educational Amendments, by Section 504 of the Rehabilitation Act of 1973, by Title VI of the Civil Rights Act of 1964, and by Vermont State Law. > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From jthomas at bittware.com Fri Apr 3 02:18:51 2009 From: jthomas at bittware.com (j.w. thomas) Date: Thu, 02 Apr 2009 22:18:51 -0400 Subject: [K12OSN] IPCop DNS Issues Message-ID: <49D5720B.2020201@bittware.com> Hi everyone, I recently set up a box and installed IPCop on it to serve as our firewall. It apparently runs a caching DNS server, but for the life of me, I cannot figure out where it gets ITS dns info. And as you might guess, I'm having some pretty severe DNS troubles. Lookups are taking a full minute. That means that any webpage with say... five different domains in it takes a minimum of five minutes to load. In the K12ltsp box I changed /etc/resolv.conf so that it uses a name server other than ipcop, but that didn't seem to help. If I ping something by ip address, it comes back immediately, but if I ping by name, it takes a minute, and THEN I get good ping times. I know this is not the IPCop mailing list, but I'm asking here because this is a friendly (and knowledgeable) crowd. Also, I would entertain the thought of using something other than IPCop, so if anyone has a suggestion, please! I'm all ears! -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Anything is possible if you don't know what you are talking about. From moon at smbis.com Fri Apr 3 02:34:40 2009 From: moon at smbis.com (Moon) Date: Thu, 02 Apr 2009 22:34:40 -0400 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <49D5720B.2020201@bittware.com> References: <49D5720B.2020201@bittware.com> Message-ID: <1238726080.4027.25.camel@t3002.localdomain> Try pfSense, it's way more robust and stable than IPCop. Also has a lot of package add-ons, like squid and squidguard for web caching and filtering. Gives more granular control. Runs on low power systems like PC Engines ALEX Geode based SBCs. On Thu, 2009-04-02 at 22:18 -0400, j.w. thomas wrote: > Hi everyone, > > I recently set up a box and installed IPCop on it to serve as our > firewall. It apparently runs a caching DNS server, but for the life of > me, I cannot figure out where it gets ITS dns info. And as you might > guess, I'm having some pretty severe DNS troubles. > > Lookups are taking a full minute. That means that any webpage with > say... five different domains in it takes a minimum of five minutes to load. > > In the K12ltsp box I changed /etc/resolv.conf so that it uses a name > server other than ipcop, but that didn't seem to help. If I ping > something by ip address, it comes back immediately, but if I ping by > name, it takes a minute, and THEN I get good ping times. > > I know this is not the IPCop mailing list, but I'm asking here because > this is a friendly (and knowledgeable) crowd. Also, I would entertain > the thought of using something other than IPCop, so if anyone has a > suggestion, please! I'm all ears! > -------------- next part -------------- An HTML attachment was scrubbed... URL: From moon at smbis.com Fri Apr 3 02:38:34 2009 From: moon at smbis.com (Moon) Date: Thu, 02 Apr 2009 22:38:34 -0400 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <49D5720B.2020201@bittware.com> References: <49D5720B.2020201@bittware.com> Message-ID: <1238726314.4027.28.camel@t3002.localdomain> Also, you may want to check into using OpenDNS as opposed to your service providers DNS servers. I have AT&T here in Georgia and they are notorious for their DNS problems... I changed my gateway to point to OpenDNS and haven't had any more DNS problems... On Thu, 2009-04-02 at 22:18 -0400, j.w. thomas wrote: > Hi everyone, > > I recently set up a box and installed IPCop on it to serve as our > firewall. It apparently runs a caching DNS server, but for the life of > me, I cannot figure out where it gets ITS dns info. And as you might > guess, I'm having some pretty severe DNS troubles. > > Lookups are taking a full minute. That means that any webpage with > say... five different domains in it takes a minimum of five minutes to load. > > In the K12ltsp box I changed /etc/resolv.conf so that it uses a name > server other than ipcop, but that didn't seem to help. If I ping > something by ip address, it comes back immediately, but if I ping by > name, it takes a minute, and THEN I get good ping times. > > I know this is not the IPCop mailing list, but I'm asking here because > this is a friendly (and knowledgeable) crowd. Also, I would entertain > the thought of using something other than IPCop, so if anyone has a > suggestion, please! I'm all ears! > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jthomas at bittware.com Fri Apr 3 02:45:41 2009 From: jthomas at bittware.com (j.w. thomas) Date: Thu, 02 Apr 2009 22:45:41 -0400 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <1238726314.4027.28.camel@t3002.localdomain> References: <49D5720B.2020201@bittware.com> <1238726314.4027.28.camel@t3002.localdomain> Message-ID: <49D57855.8070103@bittware.com> Moon wrote: > Also, you may want to check into using OpenDNS as opposed to your > service providers DNS servers. I have AT&T here in Georgia and they are > notorious for their DNS problems... I changed my gateway to point to > OpenDNS and haven't had any more DNS problems... > Thanks. I was planning to use 4.2.2.1, 2, & 3 as my dns servers but was having trouble figuring out how to make ipcop ignore the dns servers assigned by our isp via dhcp. I DID finally figure that out, but I'm not at the school so I can't make the adjustment. It's under Network->Dial-up of all places. I never would have looked there since we are most definitely NOT using dial-up (but these DNS problems sure make it feel that way). Anyhow, I'm going to try that tomorrow, and if I can't get it resolved that way, I will give pfsense a try (as per your suggestion). -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Anything is possible if you don't know what you are talking about. From jthomas at bittware.com Fri Apr 3 03:06:52 2009 From: jthomas at bittware.com (j.w. thomas) Date: Thu, 02 Apr 2009 23:06:52 -0400 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <49D57855.8070103@bittware.com> References: <49D5720B.2020201@bittware.com> <1238726314.4027.28.camel@t3002.localdomain> <49D57855.8070103@bittware.com> Message-ID: <49D57D4C.1010109@bittware.com> j.w. thomas wrote: > Moon wrote: >> Also, you may want to check into using OpenDNS as opposed to your >> service providers DNS servers. I have AT&T here in Georgia and they >> are notorious for their DNS problems... I changed my gateway to point >> to OpenDNS and haven't had any more DNS problems... >> > > Thanks. I was planning to use 4.2.2.1, 2, & 3 as my dns servers but was > having trouble figuring out how to make ipcop ignore the dns servers > assigned by our isp via dhcp. Oh man, oh man, oh man, oh man! Major thanks for pointing me towards OpenDNS. Wow. I think I can mark web filtering off my todo list now. I had no idea! -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Anything is possible if you don't know what you are talking about. From nils at breun.nl Fri Apr 3 06:44:39 2009 From: nils at breun.nl (Nils Breunese) Date: Fri, 3 Apr 2009 08:44:39 +0200 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <49D57D4C.1010109@bittware.com> References: <49D5720B.2020201@bittware.com> <1238726314.4027.28.camel@t3002.localdomain> <49D57855.8070103@bittware.com> <49D57D4C.1010109@bittware.com> Message-ID: j.w. thomas wrote: > j.w. thomas wrote: >> Moon wrote: >>> Also, you may want to check into using OpenDNS as opposed to your >>> service providers DNS servers. I have AT&T here in Georgia and >>> they are notorious for their DNS problems... I changed my gateway >>> to point to OpenDNS and haven't had any more DNS problems... >>> >> Thanks. I was planning to use 4.2.2.1, 2, & 3 as my dns servers >> but was having trouble figuring out how to make ipcop ignore the >> dns servers assigned by our isp via dhcp. > > Oh man, oh man, oh man, oh man! Major thanks for pointing me > towards OpenDNS. Wow. I think I can mark web filtering off my todo > list now. I had no idea! I have never used OpenDNS, but the paragraph "Privacy issues, conflicts and covert redirection" on Wikipedia doesn't really make me want to use it: http://en.wikipedia.org/wiki/OpenDNS#Privacy_issues.2C_conflicts_and_covert_redirection Nils Breunese. From robark at gmail.com Fri Apr 3 07:11:04 2009 From: robark at gmail.com (Robert Arkiletian) Date: Fri, 3 Apr 2009 00:11:04 -0700 Subject: [K12OSN] Fl_TeacherTool 0.70 released Message-ID: Thank goodness for student teachers. I finally found some time to code a new version. I have wanted to add Internet access control for a long time. http://www3.telus.net/public/robark/Fl_TeacherTool/download.html Here is a screencast video of the new feature. http://blip.tv/file/1951333/ BTW this version *should* also work if you are using squid proxy. But of course it will only work if you have a standard 2 nic setup. Enjoy! -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From arturo at susetic.com Fri Apr 3 08:05:56 2009 From: arturo at susetic.com (=?ISO-8859-1?Q?Arturo_Lim=F3n?=) Date: Fri, 3 Apr 2009 10:05:56 +0200 Subject: [K12OSN] Broken updates for K12LTSP ...!!!! Message-ID: I have just installed a K12LTSP server. I have other 8 similar installations running in schools. Trying to update my just installed system, I run into a suprise. Yum tells me it cannot find http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/.... what is true, because going into its web page, now it has changed to 5.3: http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.3/.... This is my yum output: [root at linux yum.repos.d]# yum update Loading "installonlyn" plugin Setting up Update Process Setting up repositories k12ltsp 100% |=========================| 1.1 kB 00:00 addons 100% |=========================| 951 B 00:00 webmin 100% |=========================| 951 B 00:00 macromedia 100% |=========================| 1.9 kB 00:00 updates 100% |=========================| 951 B 00:00 base 100% |=========================| 1.1 kB 00:00 extras 100% |=========================| 1.1 kB 00:00 Reading repository metadata in from local files http://mirror.silyus.net/centos/5.3/updates/i386/repodata/primary.xml.gz: [Errno 12] Timeout: Trying other mirror. primary.xml.gz 100% |=========================| 69 kB 00:00 updates : ################################################## 83/83 Added 42 new packages, deleted 0 old in 0.77 seconds Resolving Dependencies --> Populating transaction set with selected packages. Please wait. ---> Downloading header for chkconfig to pack into transaction set. http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/chkconfig-1.3.30.1-2.i386.rpm: [Errno 14] HTTP Error 404: Not Found Trying other mirror. Error: failure: 5.2/chkconfig-1.3.30.1-2.i386.rpm from k12ltsp: [Errno 256] No more mirrors to try. What do I have todo to make yum work with 5.3 and not with 5.2? Thanks in advance. Arturo Limon -------------- next part -------------- An HTML attachment was scrubbed... URL: From nils at breun.nl Fri Apr 3 08:17:07 2009 From: nils at breun.nl (Nils Breunese) Date: Fri, 3 Apr 2009 10:17:07 +0200 Subject: [K12OSN] Broken updates for K12LTSP ...!!!! In-Reply-To: References: Message-ID: <12CBF116-A33A-431D-8FC2-A5CBC4F49EB5@breun.nl> Arturo Lim?n wrote: > I have just installed a K12LTSP server. I have other 8 similar > installations running in schools. > > Trying to update my just installed system, I run into a suprise. Yum > tells me it cannot find http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/ > .... what is true, because going into its web page, now it has > changed to 5.3: http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.3/ > .... > > This is my yum output: > > [root at linux yum.repos.d]# yum update > Loading "installonlyn" plugin > Setting up Update Process > Setting up repositories > k12ltsp 100% |=========================| 1.1 kB > 00:00 > addons 100% |=========================| 951 B > 00:00 > webmin 100% |=========================| 951 B > 00:00 > macromedia 100% |=========================| 1.9 kB > 00:00 > updates 100% |=========================| 951 B > 00:00 > base 100% |=========================| 1.1 kB > 00:00 > extras 100% |=========================| 1.1 kB > 00:00 > Reading repository metadata in from local files > http://mirror.silyus.net/centos/5.3/updates/i386/repodata/primary.xml.gz > : [Errno 12] Timeout: > Trying other mirror. > primary.xml.gz 100% |=========================| 69 kB > 00:00 > updates : ################################################## 83/83 > Added 42 new packages, deleted 0 old in 0.77 seconds > Resolving Dependencies > --> Populating transaction set with selected packages. Please wait. > ---> Downloading header for chkconfig to pack into transaction set. > http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/chkconfig-1.3.30.1-2.i386.rpm > : [Errno 14] HTTP Error 404: Not Found > Trying other mirror. > Error: failure: 5.2/chkconfig-1.3.30.1-2.i386.rpm from k12ltsp: > [Errno 256] No more mirrors to try. > > What do I have todo to make yum work with 5.3 and not with 5.2? Might have something to do with this week's CentOS 5.3 release? (K12LTSP 5 is based on CentOS 5.) Nils Breunese. From dahopkins429 at gmail.com Fri Apr 3 13:59:26 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Fri, 3 Apr 2009 09:59:26 -0400 Subject: [K12OSN] OT: Testing FC10 results Message-ID: This is sort-of off-topic, but the server I have installed with FC10, LTSP5 continues to perform rather poorly with packages such as tuxpaint, gcompris and the like. Not sure where the slowdown is. I have the latest version of both packages installed. I am curious though if using the bridge might not be a source of the slowdown? I could be grasping at straws but I need to find a way to determine why this system which worked well with K12LTSP5 (CentOS) now is sluggish with FC10. When it is just a few users, the reason for the upgrade are obvious as sound works very well, videos play correctly, etc. But, past about 10 users and it just crawls. I am not running out of memory, nor is the load average sky high. It is just slow. So, could it be that the ltspbr0 bridge is a source of the issue? Or using nbdswap for the clients? Any ideas on how to tell? Sincerely, Dave Hopkins From microman at cmosnetworks.com Fri Apr 3 15:02:28 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Fri, 03 Apr 2009 11:02:28 -0400 Subject: [K12OSN] Broken updates for K12LTSP ...!!!! In-Reply-To: <12CBF116-A33A-431D-8FC2-A5CBC4F49EB5@breun.nl> References: <12CBF116-A33A-431D-8FC2-A5CBC4F49EB5@breun.nl> Message-ID: <49D62504.800@cmosnetworks.com> Nils Breunese wrote: > Arturo Lim?n wrote: > >> I have just installed a K12LTSP server. I have other 8 similar >> installations running in schools. >> >> Trying to update my just installed system, I run into a suprise. Yum >> tells me it cannot find >> http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/.... >> what is true, because going into its web page, now it has changed to >> 5.3: http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.3/.... >> >> This is my yum output: >> >> [root at linux yum.repos.d]# yum update >> Loading "installonlyn" plugin >> Setting up Update Process >> Setting up repositories >> k12ltsp 100% |=========================| 1.1 kB >> 00:00 >> addons 100% |=========================| 951 B >> 00:00 >> webmin 100% |=========================| 951 B >> 00:00 >> macromedia 100% |=========================| 1.9 kB >> 00:00 >> updates 100% |=========================| 951 B >> 00:00 >> base 100% |=========================| 1.1 kB >> 00:00 >> extras 100% |=========================| 1.1 kB >> 00:00 >> Reading repository metadata in from local files >> http://mirror.silyus.net/centos/5.3/updates/i386/repodata/primary.xml.gz: >> [Errno 12] Timeout: >> Trying other mirror. >> primary.xml.gz 100% |=========================| 69 kB >> 00:00 >> updates : ################################################## 83/83 >> Added 42 new packages, deleted 0 old in 0.77 seconds >> Resolving Dependencies >> --> Populating transaction set with selected packages. Please wait. >> ---> Downloading header for chkconfig to pack into transaction set. >> http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/chkconfig-1.3.30.1-2.i386.rpm: >> [Errno 14] HTTP Error 404: Not Found >> Trying other mirror. >> Error: failure: 5.2/chkconfig-1.3.30.1-2.i386.rpm from k12ltsp: >> [Errno 256] No more mirrors to try. >> >> What do I have todo to make yum work with 5.3 and not with 5.2? > > Might have something to do with this week's CentOS 5.3 release? > (K12LTSP 5 is based on CentOS 5.) > > Nils Breunese. > Yep, probably does. Give it a couple of days and try it again. --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From reb at taco.com Fri Apr 3 15:07:39 2009 From: reb at taco.com (Phydeaux) Date: Fri, 03 Apr 2009 11:07:39 -0400 Subject: [K12OSN] Problem with USB Stick (NTFS) In-Reply-To: References: Message-ID: <20090403150742.4290B61EF2C@althea.taco.com> At 11:53 AM 04/01/09, thomas Thomas wrote: >I'm Ace, and i'm a french user of K12Linux :) > >I have a server LTSP with Fedora 10. >And,i have a problem with NTFS USB Stick. > >When an USB Stick is mounted on a client, a folder appears on the client's desktop, but is empty.... >However, when the USB stick is mounted on the server, his folder appears on the desktop's server, but the folder is readable ! That sounds as if it might be related to this problem that I've been looking for a solution to: https://bugzilla.redhat.com/show_bug.cgi?id=484580 Do you see any related error messages? reb From acerothstein at hotmail.fr Fri Apr 3 21:16:28 2009 From: acerothstein at hotmail.fr (thomas Thomas) Date: Fri, 3 Apr 2009 21:16:28 +0000 Subject: [K12OSN] RE: K12OSN Digest, Vol 62, Issue 3 In-Reply-To: <20090403160044.5479A619E7B@hormel.redhat.com> References: <20090403160044.5479A619E7B@hormel.redhat.com> Message-ID: > Message: 12 > Date: Fri, 03 Apr 2009 11:07:39 -0400 > From: Phydeaux > Subject: Re: [K12OSN] Problem with USB Stick (NTFS) > To: "Support list for open source software in schools." > , > Message-ID: <20090403150742.4290B61EF2C at althea.taco.com> > Content-Type: text/plain; charset="us-ascii" > > At 11:53 AM 04/01/09, thomas Thomas wrote: > >I'm Ace, and i'm a french user of K12Linux :) > > > >I have a server LTSP with Fedora 10. > >And,i have a problem with NTFS USB Stick. > > > >When an USB Stick is mounted on a client, a folder appears on the client's desktop, but is empty.... > >However, when the USB stick is mounted on the server, his folder appears on the desktop's server, but the folder is readable ! > > That sounds as if it might be related to this problem that I've been looking for a > solution to: > > https://bugzilla.redhat.com/show_bug.cgi?id=484580 > > Do you see any related error messages? > > reb > > > Hi, Thank you for your answer :) But,i have not any error messages when the usb stick is mounted... So if the error message are present, Where find it on the system? I go to read your bugzilla. @+ _________________________________________________________________ T?l?phonez gratuitement ? tous vos proches avec Windows Live Messenger? !? T?l?chargez-le maintenant !? http://www.windowslive.fr/messenger/1.asp -------------- next part -------------- An HTML attachment was scrubbed... URL: From news at siddall.name Sat Apr 4 02:38:33 2009 From: news at siddall.name (Jeff Siddall) Date: Fri, 03 Apr 2009 22:38:33 -0400 Subject: [K12OSN] ltsp-build-client failing Message-ID: <49D6C829.2080105@siddall.name> ltsp-build-client is failing. It makes it through downloading and installing all the packages then reports an error. I can't see anything obviously wrong, and I can manually umount /var/tmp/imgcreate-Slotq5/install_root. Any ideas what is going wrong? Thanks, Jeff ---------- ... Installing: xorg-x11-drivers ##################### [478/481] Installing: hpijs ##################### [479/481] Installing: system-config-display ##################### [480/481] Installing: libsane-hpaio ##################### [481/481] umount: /var/tmp/imgcreate-Slotq5/install_root: device is busy. (In some cases useful info about processes that use the device is found by lsof(8) or fuser(1)) /usr/share/ltsp/plugins/ltsp-build-client/common/010-chroot-tagging: line 3: /opt/ltsp/i386/etc/ltsp_chroot: No such file or directory error: LTSP client installation ended abnormally From fcdanilo at gmail.com Sat Apr 4 02:47:36 2009 From: fcdanilo at gmail.com (Danilo =?ISO-8859-1?Q?C=E2mara?=) Date: Fri, 03 Apr 2009 23:47:36 -0300 Subject: [K12OSN] Re: "umount: device is busy" error in ltsp-build-client In-Reply-To: <1234541696.3548.79.camel@localhost.localdomain> References: <1234541696.3548.79.camel@localhost.localdomain> Message-ID: <1238813256.3709.33.camel@localhost.localdomain> I'm replying to myself to register a workaround about this post. What I did was to proceed the installation as described in https://fedorahosted.org/k12linux/wiki/InstallGuide but instead of running "ltsp-build-client", I copied "/opt/ltsp" and "/var/lib/tftpboot" from an system running "K12Linux Live Server" (https://fedorahosted.org/k12linux/wiki/LiveServer). I also had to fix "/opt/ltsp/i386/etc/ssh/ssh_known_hosts" to match the keys of the new server. More precisely: (opt_ltsp.k12linux.i386.tgz and var_lib_tftpboot.k12linux.i386.tgz were created from K12Linux Live Server) mv /opt/ltsp /opt/ltsp.bak tar -xzf opt_ltsp.k12linux.i386.tgz -C /opt mv /var/lib/tftpboot /var/lib/tftpboot.bak tar -xzf var_lib_tftpboot.k12linux.i386.tgz -C /var/lib chcon -u system_u /opt/ltsp -R restorecon /var/lib/tftpboot -R # replace these with your values IP_ADDR_ETH0=192.168.0.1 IP_ADDR_ETH1=10.0.0.254 FILE=/opt/ltsp/i386/etc/ssh/ssh_known_hosts echo -n "$(hostname) " > $FILE cat /etc/ssh/ssh_host_dsa_key.pub >> $FILE echo -n "$(hostname) " >> $FILE cat /etc/ssh/ssh_host_rsa_key.pub >> $FILE echo -n "$IP_ADDR_ETH0 " >> $FILE cat /etc/ssh/ssh_host_dsa_key.pub >> $FILE echo -n "$IP_ADDR_ETH0 " >> $FILE cat /etc/ssh/ssh_host_rsa_key.pub >> $FILE echo -n "$IP_ADDR_ETH1 " >> $FILE cat /etc/ssh/ssh_host_dsa_key.pub >> $FILE echo -n "$IP_ADDR_ETH1 " >> $FILE cat /etc/ssh/ssh_host_rsa_key.pub >> $FILE Em Sex, 2009-02-13 ?s 13:15 -0300, Danilo C?mara escreveu: > So far I have successfully installed LTSP in Fedora 10 x86_64 servers > (for i386 thin-clients) but when installing on Fedora 10 i386 servers I > always get the following error message in ltsp-build-client: > > ... > Installing: perl-Pod-Simple ##################### [387/388] > Installing: hpijs ##################### [388/388] > > umount: /var/tmp/imgcreate-mDg5Ft/install_root: device is busy. > (In some cases useful info about processes that use > the device is found by lsof(8) or fuser(1)) > /usr/share/ltsp/plugins/ltsp-build-client/common/010-chroot-tagging: line 3: /opt/ltsp/i386/etc/ltsp_chroot: No such file or directory > error: LTSP client installation ended abnormally > > > The best match I found on the Web about this is on > https://www.redhat.com/archives/fedora-livecd-list/2008-February/msg00120.html > but there was no conclusive solution there. > > > I am installing from the LTSP packages in Fedora 10 distribution and did > the following configurations: > > diff /etc/ltsp/ltsp-build-client.conf.original /etc/ltsp/ltsp-build-client.conf > 8c8 > < # option_cache_value=/var/cache/chroot > --- > > option_cache_value=/var/cache/chroot > 10c10 > < # option_arch_value=i386 > --- > > option_arch_value=i386 > > > and these configuration to use our local network mirror of the packages > (downloading from web I got the same results) > > diff /etc/ltsp/kickstart/Fedora/10/ltsp-i386.ks.original /etc/ltsp/kickstart/Fedora/10/ltsp-i386.ks > 6,8c6,8 > < repo --name=released-10-i386 --mirrorlist=http://mirrors.fedoraproject.org/mirrorlist?repo=fedora-10&arch=i386 > < repo --name=updates-10-i386 --mirrorlist=http://mirrors.fedoraproject.org/mirrorlist?repo=updates-released-f10&arch=i386 > < repo --name=temporary-10-i386 --baseurl=http://togami.com/~k12linux-temporary/fedora/10/i386/ > --- > > repo --name=released-10-i386 --baseurl=file:///home/Packages/Linux/Fedora/10/i386/os/ > > repo --name=updates-10-i386 --baseurl=file:///home/Packages/Linux/Fedora/10/i386/updates/ > > repo --name=temporary-10-i386 --baseurl=file:///home/Packages/Linux/Fedora/10/i386/k12linux/ > > > Danilo Camara From josh.fryman at gmail.com Sat Apr 4 05:53:21 2009 From: josh.fryman at gmail.com (Josh Fryman) Date: Fri, 3 Apr 2009 22:53:21 -0700 Subject: [K12OSN] Guide to enabling Local Apps support? Message-ID: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> Hi, I've set up from the "K12Linux F10 Stable 4 x86-64" a fully working server. I have several clients booting into it perfectly. My clients are (right now) all i386 units. The trick is that I want to enable local apps for Firefox + Flash to begin with, and a few other things later. The documentation link on the K12LTSP site is broken; the Ubuntu documentation link is also broken; there are no mailing list posts that I can find that explain how to actually do this; and the only things that seem to mention it are LTSP 4.x based. I find a lot of references to people saying they got parts working, but no actual guides. As a bit of background, I'm a long-time Linux user (since mid 90's) but have never tried to set this up before. My kids' school district is facing a massive budget cut (who isn't right now?) and are looking at any avenue to cut costs and save jobs. The district's Microsoft license is due next year, and I've suggested that LTSP + OpenOffice + etc are a known viable replacement for their needs. The only glitch I've found in my test setup, which I'm supposed to be demonstrating next week, is the really slow support for Flash videos and certain highly-interactive programs from the server-hosted Firefox. Some of the material the schools present to kids is in Flash format from external sites, and it just crawls on my test setup with beefy server and quite powerful thin clients. It sounds like the only solution is local app support for Firefox + flash, but . . . I cannot seem to find any explanation of how to actually configure it. If there's a useful (and applicable) URL out there that I've just not found, or you can provide me with information on how to do this, I would be grateful. I'm also willing to write the document that explains how to make this work, if someone can give me enough bits to get it going. Thanks for your time, Josh Fryman From robark at gmail.com Sat Apr 4 06:14:28 2009 From: robark at gmail.com (Robert Arkiletian) Date: Fri, 3 Apr 2009 23:14:28 -0700 Subject: [K12OSN] Guide to enabling Local Apps support? In-Reply-To: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> References: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> Message-ID: look at this thread https://www.redhat.com/archives/k12linux-devel-list/2008-October/msg00035.html On Fri, Apr 3, 2009 at 10:53 PM, Josh Fryman wrote: > Hi, > > I've set up from the "K12Linux F10 Stable 4 x86-64" a fully working > server. I have several clients booting into it perfectly. My clients > are (right now) all i386 units. > > The trick is that I want to enable local apps for Firefox + Flash to > begin with, and a few other things later. > > The documentation link on the K12LTSP site is broken; the Ubuntu > documentation link is also broken; there are no mailing list posts > that I can find that explain how to actually do this; and the only > things that seem to mention it are LTSP 4.x based. I find a lot of > references to people saying they got parts working, but no actual > guides. > > As a bit of background, I'm a long-time Linux user (since mid 90's) > but have never tried to set this up before. My kids' school district > is facing a massive budget cut (who isn't right now?) and are looking > at any avenue to cut costs and save jobs. The district's Microsoft > license is due next year, and I've suggested that LTSP + OpenOffice + > etc are a known viable replacement for their needs. > > The only glitch I've found in my test setup, which I'm supposed to be > demonstrating next week, is the really slow support for Flash videos > and certain highly-interactive programs from the server-hosted > Firefox. Some of the material the schools present to kids is in Flash > format from external sites, and it just crawls on my test setup with > beefy server and quite powerful thin clients. It sounds like the only > solution is local app support for Firefox + flash, but . . . I cannot > seem to find any explanation of how to actually configure it. > > If there's a useful (and applicable) URL out there that I've just not > found, or you can provide me with information on how to do this, I > would be grateful. I'm also willing to write the document that > explains how to make this work, if someone can give me enough bits to > get it going. > > Thanks for your time, > > Josh Fryman > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From burke at thealmquists.net Sat Apr 4 10:48:58 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 05:48:58 -0500 Subject: [K12OSN] Guide to enabling Local Apps support? In-Reply-To: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> References: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 4, 2009, at 12:53 AM, Josh Fryman wrote: > Hi, > > I've set up from the "K12Linux F10 Stable 4 x86-64" a fully working > server. I have several clients booting into it perfectly. My clients > are (right now) all i386 units. > > The trick is that I want to enable local apps for Firefox + Flash to > begin with, and a few other things later. > You should also note, from this page https://fedorahosted.org/ k12linux/wiki/Tips, the part about Yum in the chroot. Since you need that for local apps. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAknXOxoACgkQxWV7OPa/g5HwyQCfRv9xpR4kjhzkhay1NlUf73c5 uTkAoIgxYI1TAPJWBHwq+vSbaGhiGg0c =T6er -----END PGP SIGNATURE----- From burke at thealmquists.net Sat Apr 4 11:26:36 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 06:26:36 -0500 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <49D5720B.2020201@bittware.com> References: <49D5720B.2020201@bittware.com> Message-ID: <1443710F-1A29-406B-9630-99ACEF0635A8@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 2, 2009, at 9:18 PM, j.w. thomas wrote: > Hi everyone, > > I recently set up a box and installed IPCop on it to serve as our > firewall. It apparently runs a caching DNS server, but for the > life of me, I cannot figure out where it gets ITS dns info. And as > you might guess, I'm having some pretty severe DNS troubles. > > Lookups are taking a full minute. That means that any webpage with > say... five different domains in it takes a minimum of five minutes > to load. > Are you using LDAP or some other network authentication system by chance? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAknXQ+wACgkQxWV7OPa/g5Hl8wCfSeCb0gbMpmSduEIoG0L6bcki 2M8An3eAm6mk/ZahxxrLBIU/3fslGNrf =cMLh -----END PGP SIGNATURE----- From burke at thealmquists.net Sat Apr 4 12:13:33 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 07:13:33 -0500 Subject: [K12OSN] Single NIC Setup In-Reply-To: <49CC8E6D.4040000@arkki.info> References: <49CB9AC3.1020902@cnc-usa.com> <49CC8E6D.4040000@arkki.info> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mar 27, 2009, at 3:29 AM, Asmo Koskinen wrote: > Almquist Burke kirjoitti: > >> https://fedorahosted.org/k12linux/wiki/AdvancedNetworkSetup Having installed a single NIC setup of K12Linux (LTSP 5 on Fedora 10) is there anything you think is missing from the wiki page that would trip someone up? Something they need to know to get this working that is not obvious from what has been written on the site? > > I used that howto when I installed Fedora in KVM (Ubuntu). > > http://www.arkki.info/howto/Fedora10/Eee_FC10_00.png > http://www.arkki.info/howto/Fedora10/ > > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAknXTu0ACgkQxWV7OPa/g5EJmQCcCowaxMf+Kj874VQlah7KRxmi tt4Anj9HMyUpYw+YfqILem3ApbXO7PcE =2LSC -----END PGP SIGNATURE----- From peter at scheie.homedns.org Sat Apr 4 13:33:27 2009 From: peter at scheie.homedns.org (Peter Scheie) Date: Sat, 04 Apr 2009 08:33:27 -0500 Subject: [K12OSN] Guide to enabling Local Apps support? In-Reply-To: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> References: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> Message-ID: <49D761A7.6010903@scheie.homedns.org> One point you may want to make in your presentation is that having a 'one architecture to fill all wants' is really expensive, and frankly rather wasteful. Viewing flash video is nice, but how much time each day/week/month do the kids actually spend doing that? Word processing, presentations, non-flash web browsing (e.g., flash isn't needed for reading Wikipedia), etc., probably consume a greater amount of kids' computer time in school, and on LTSP can be done so cheaply, both in software & hardware, that it is practical to put computers everywhere (put a lab in each classroom if you want). IOW, Linux can provide what is needed 90% of the time at 10% of the cost; Windows eats 90% of the budget but only adds value 10% of the time. Perhaps a small Windows lab is necessary for those edge cases, but the primary architecture should be LTSP. Peter Josh Fryman wrote: > Hi, > > I've set up from the "K12Linux F10 Stable 4 x86-64" a fully working > server. I have several clients booting into it perfectly. My clients > are (right now) all i386 units. > > The trick is that I want to enable local apps for Firefox + Flash to > begin with, and a few other things later. > > The documentation link on the K12LTSP site is broken; the Ubuntu > documentation link is also broken; there are no mailing list posts > that I can find that explain how to actually do this; and the only > things that seem to mention it are LTSP 4.x based. I find a lot of > references to people saying they got parts working, but no actual > guides. > > As a bit of background, I'm a long-time Linux user (since mid 90's) > but have never tried to set this up before. My kids' school district > is facing a massive budget cut (who isn't right now?) and are looking > at any avenue to cut costs and save jobs. The district's Microsoft > license is due next year, and I've suggested that LTSP + OpenOffice + > etc are a known viable replacement for their needs. > > The only glitch I've found in my test setup, which I'm supposed to be > demonstrating next week, is the really slow support for Flash videos > and certain highly-interactive programs from the server-hosted > Firefox. Some of the material the schools present to kids is in Flash > format from external sites, and it just crawls on my test setup with > beefy server and quite powerful thin clients. It sounds like the only > solution is local app support for Firefox + flash, but . . . I cannot > seem to find any explanation of how to actually configure it. > > If there's a useful (and applicable) URL out there that I've just not > found, or you can provide me with information on how to do this, I > would be grateful. I'm also willing to write the document that > explains how to make this work, if someone can give me enough bits to > get it going. > > Thanks for your time, > > Josh Fryman > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From burke at thealmquists.net Sat Apr 4 14:20:53 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 09:20:53 -0500 Subject: [K12OSN] Feedback from a teacher on FC10/LTSP and Tuxpaint In-Reply-To: References: Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Does the LTSP 5 setup have all X traffic encrypted by chance? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAknXbMUACgkQxWV7OPa/g5FRqgCdH+i2aVgYN1ftPWDTpHE/oY5G 5aMAn3OoX1fjYNJMCQNv6e9eXBYPEm4X =3e2l -----END PGP SIGNATURE----- From asmo.koskinen at arkki.info Sat Apr 4 18:37:38 2009 From: asmo.koskinen at arkki.info (asmo.koskinen at arkki.info) Date: Sat, 4 Apr 2009 21:37:38 +0300 (EEST) Subject: [K12OSN] Single NIC Setup In-Reply-To: References: <49CB9AC3.1020902@cnc-usa.com> <49CC8E6D.4040000@arkki.info> Message-ID: <48578.85.157.204.13.1238870258.squirrel@webmail3.nebula.fi> > Having installed a single NIC setup of K12Linux (LTSP 5 on Fedora 10) > is there anything you think is missing from the wiki page that would > trip someone up? Something they need to know to get this working that > is not obvious from what has been written on the site? > >> >> I used that howto when I installed Fedora in KVM (Ubuntu). >> >> http://www.arkki.info/howto/Fedora10/Eee_FC10_00.png >> http://www.arkki.info/howto/Fedora10/ I do not remember anything unusual. First there was a problem with dhcpd.conf, but I figured out that. http://www.arkki.info/howto/Fedora10/Nfsmount_error_FC10.flv I just added server's ip number in dhcpd.conf file. [--] option root-path "192.168.1.110:/opt/ltsp/i386"; [--] Sorry, but I do not have that setup anymore. There are just some screenshots about installing LTSP5 in Fedora on KVM. http://www.arkki.info/howto/Fedora10/Kuvakaappaus-fedora-10%20Virtual%20Machine%20Console-0.png [--] http://www.arkki.info/howto/Fedora10/Kuvakaappaus-fedora-10%20Virtual%20Machine%20Console-8.png Best Regards Asmo Koskinen. From dahopkins429 at gmail.com Sat Apr 4 19:07:23 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Sat, 4 Apr 2009 15:07:23 -0400 Subject: [K12OSN] Feedback from a teacher on FC10/LTSP and Tuxpaint In-Reply-To: References: Message-ID: > Does the LTSP 5 setup have all X traffic encrypted by chance? I thought I had the encryption turned off via LDM_DIRECTX=YES Is there a way to just not use LDM at all and just use gdm? TBH I am not that worried about having anything intercepted between the server and the clients. This is a standard 2 NIC setup. Sincerely, Dave Hopkins From burke at thealmquists.net Sat Apr 4 20:28:34 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 15:28:34 -0500 Subject: [K12OSN] Feedback from a teacher on FC10/LTSP and Tuxpaint In-Reply-To: References: Message-ID: <9E768C62-1E5C-45DC-826D-610081C4EFEF@thealmquists.net> I was thinking that by default only the login screen was encrypted, but I'm not sure now. Plus there are two lts.conf files an empty one in /opt/ltsp/i386/etc/ and one in /var/lib/tftpboot/arch/ II guess I'm a little confused about that. On Apr 4, 2009, at 2:07 PM, David Hopkins wrote: >> Does the LTSP 5 setup have all X traffic encrypted by chance? > > I thought I had the encryption turned off via > > LDM_DIRECTX=YES > > Is there a way to just not use LDM at all and just use gdm? TBH I am > not that worried about having anything intercepted between the server > and the clients. This is a standard 2 NIC setup. > > Sincerely, > Dave Hopkins > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part URL: From burke at thealmquists.net Sat Apr 4 20:52:06 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 15:52:06 -0500 Subject: [K12OSN] Custom xorg.conf In-Reply-To: References: Message-ID: <6C35ACD3-CF11-456D-8A12-2E9A64C5A9E5@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 1, 2009, at 7:35 AM, David Hopkins wrote: >> Can you specify a resolution in the lts.conf file? > > Supposedly, but I tried XMODE and that didn't change anything ... > unless I'm looking at the wrong lts.conf? I'm modifying the one in > tftpboot/i386. There is also an lts.conf in /opt/ltsp/i386/etc. And, > finally, with ltsp4.2, there was an /etc/hosts file with entries for > all the thin clients. with ltsp5 that has also changed? Sort of a > change in topic, but I'll check tonight to see if the client knows its > name. > According to the lts.conf man page, you need X_MODE_0 = 1024x768 You may also need to have CONFIGURE_X = TRUE set. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEUEARECAAYFAknXyHYACgkQxWV7OPa/g5HCQwCfaocQPlBNOYSakqNolJH9tsUb ibwAmPsG+GRfWG4tHCu/LeyFiKUbrVw= =HvU2 -----END PGP SIGNATURE----- From burke at thealmquists.net Sat Apr 4 21:04:08 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 16:04:08 -0500 Subject: [K12OSN] OT: Testing FC10 results In-Reply-To: References: Message-ID: <8D80BB3F-F9D8-457B-B434-FE2C3215BFF3@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 3, 2009, at 8:59 AM, David Hopkins wrote: > This is sort-of off-topic, but the server I have installed with FC10, > LTSP5 continues to perform rather poorly with packages such as > tuxpaint, gcompris and the like. > So, could it be that the ltspbr0 bridge is a source of the issue? I don't think that should impact performance much at all. IIRC the chroot memory footprint in K12Linux is still higher than K12LTSP. That might be part of it, if your clients have less that 128MB of memory. > Or > using nbdswap for the clients? Actually, NBD swap is supposed to be faster than NFS swap, that's why the developers implemented it. > > Any ideas on how to tell? > I thought it might be the LDM_DIRECTX setting, but that is set to TRUE by default on Fedora, which means that X traffic isn't encrypted on K12LTSP by default. So it must be something else. Higher network traffic for some reason, like increased screen resolution or color depth? You said the server has plenty of memory and CPU, and I don't think the disk IO should change that much. > Sincerely, > Dave Hopkins > > _______________________________________________ > 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) iEYEARECAAYFAknXy0gACgkQxWV7OPa/g5GqYACfSWderwx+vm9GKoJ2pA+sSE5z BvEAnju8ctkn5GQpbsAUJUIiHet46VsI =iKCZ -----END PGP SIGNATURE----- From mrjohnlucas at gmail.com Sat Apr 4 21:53:53 2009 From: mrjohnlucas at gmail.com (John Lucas) Date: Sat, 04 Apr 2009 17:53:53 -0400 Subject: [K12OSN] Broken updates for K12LTSP ...!!!! In-Reply-To: References: Message-ID: <49D7D6F1.3020103@gmail.com> Arturo Lim?n wrote: > I have just installed a K12LTSP server. I have other 8 similar > installations running in schools. > > Trying to update my just installed system, I run into a suprise. Yum > tells me it cannot > find http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/.... > what is > true, because going into its web page, now it has changed to > 5.3: http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.3/.... > > > This is my yum output: > > [root at linux yum.repos.d]# yum update > Loading "installonlyn" plugin > Setting up Update Process > Setting up repositories > k12ltsp 100% |=========================| 1.1 kB > 00:00 > addons 100% |=========================| 951 B > 00:00 > webmin 100% |=========================| 951 B > 00:00 > macromedia 100% |=========================| 1.9 kB > 00:00 > updates 100% |=========================| 951 B > 00:00 > base 100% |=========================| 1.1 kB > 00:00 > extras 100% |=========================| 1.1 kB > 00:00 > Reading repository metadata in from local files > http://mirror.silyus.net/centos/5.3/updates/i386/repodata/primary.xml.gz: > [Errno 12] Timeout: > Trying other mirror. > primary.xml.gz 100% |=========================| 69 kB > 00:00 > updates : ################################################## 83/83 > Added 42 new packages, deleted 0 old in 0.77 seconds > Resolving Dependencies > --> Populating transaction set with selected packages. Please wait. > ---> Downloading header for chkconfig to pack into transaction set. > http://k12linux.mesd.k12.or.us/K12LTSP/5.0.0-EL-32bit/5.2/chkconfig-1.3.30.1-2.i386.rpm: > [Errno 14] HTTP Error 404: Not Found > Trying other mirror. > Error: failure: 5.2/chkconfig-1.3.30.1-2.i386.rpm from k12ltsp: [Errno > 256] No more mirrors to try. > > What do I have todo to make yum work with 5.3 and not with 5.2? > > Thanks in advance. > > Arturo Limon > Before performing a general update, update yum first: yum update yum That helped me last week with a similar problem, but this was before the release of CentOS v5.3. -- "History doesn't repeat itself; at best it rhymes." - Mark Twain | John Lucas MrJohnLucas at gmail.com | | St. Thomas, VI 00802 http://mrjohnlucas.googlepages.com/ | | 18.3?N, 65?W AST (UTC-4) | From peter at scheie.homedns.org Sat Apr 4 22:03:14 2009 From: peter at scheie.homedns.org (Peter Scheie) Date: Sat, 04 Apr 2009 17:03:14 -0500 Subject: [K12OSN] Feedback from a teacher on FC10/LTSP and Tuxpaint In-Reply-To: <9E768C62-1E5C-45DC-826D-610081C4EFEF@thealmquists.net> References: <9E768C62-1E5C-45DC-826D-610081C4EFEF@thealmquists.net> Message-ID: <49D7D922.4020301@scheie.homedns.org> LDM_DIRECTX=YES encrypts the login but not the X session. LTSP 5 does not use gdm, as it has no provision for encryption, and passing user IDs & PWs cleartext is not a good thing, even in a school setting. Setting LDM_DIRECTX=NO will encrypt the entire session, which some people need, but also requires faster hardware in the client. Peter Almquist Burke wrote: > I was thinking that by default only the login screen was encrypted, but > I'm not sure now. Plus there are two lts.conf files an empty one in > /opt/ltsp/i386/etc/ and one in /var/lib/tftpboot//arch// > II guess I'm a little confused about that. > > On Apr 4, 2009, at 2:07 PM, David Hopkins wrote: > >>> Does the LTSP 5 setup have all X traffic encrypted by chance? >> >> I thought I had the encryption turned off via >> >> LDM_DIRECTX=YES >> >> Is there a way to just not use LDM at all and just use gdm? TBH I am >> not that worried about having anything intercepted between the server >> and the clients. This is a standard 2 NIC setup. >> >> Sincerely, >> Dave Hopkins >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see > > > ------------------------------------------------------------------------ > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From dahopkins429 at gmail.com Sat Apr 4 22:05:46 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Sat, 4 Apr 2009 18:05:46 -0400 Subject: [K12OSN] OT: Testing FC10 results In-Reply-To: <8D80BB3F-F9D8-457B-B434-FE2C3215BFF3@thealmquists.net> References: <8D80BB3F-F9D8-457B-B434-FE2C3215BFF3@thealmquists.net> Message-ID: > >> So, could it be that the ltspbr0 bridge is a source of the issue? > > I don't think that should impact performance much at all. IIRC the chroot > memory footprint in K12Linux is still higher than K12LTSP. That might be > part of it, if your clients have less that 128MB of memory. All clients have at least 128Mb, most have 256Mb. >> Or >> using nbdswap for the clients? > > Actually, NBD swap is supposed to be faster than NFS swap, that's why the > developers implemented it. > >> >> Any ideas on how to tell? >> > > I thought it might be the LDM_DIRECTX setting, but that is set to TRUE by > default on Fedora, which means that X traffic isn't encrypted on K12LTSP by > default. So it must be something else. ?Higher network traffic for some > reason, like increased screen resolution or color depth? You said the server > has plenty of memory and CPU, and I don't think the disk IO should change > that much. In terms of bandwidth, this server is performing the same for disk I/O as the other servers running 4.2 (measured using netio). Screen resolution is likewise typical. I could force the color depth to 16 instead of 24. It is confusing. I am using ldap for authentication and wonder if something changed wrt to how authentication is working, or if dns has changed? The host table is empty for ltsp5? ltsp4.2 had an entry for every client. This is the one difference that I know of since the clients are using 172.x.y.z whereas 4.2 used 192.x.y.z. I can change this but don't really see why it would have an effect. Sincerely, Dave Hopkins From dahopkins429 at gmail.com Sat Apr 4 22:11:05 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Sat, 4 Apr 2009 18:11:05 -0400 Subject: [K12OSN] Feedback from a teacher on FC10/LTSP and Tuxpaint In-Reply-To: <49D7D922.4020301@scheie.homedns.org> References: <9E768C62-1E5C-45DC-826D-610081C4EFEF@thealmquists.net> <49D7D922.4020301@scheie.homedns.org> Message-ID: On Sat, Apr 4, 2009 at 6:03 PM, Peter Scheie wrote: > LDM_DIRECTX=YES encrypts the login but not the X session. ?LTSP 5 does not > use gdm, as it has no provision for encryption, and passing user IDs & PWs > cleartext is not a good thing, even in a school setting. ?Setting > LDM_DIRECTX=NO will encrypt the entire session, which some people need, but > also requires faster hardware in the client. So do the same gnome login session scripts run as before? I had users in groups which assigned them desktop icons (added or removed) based on group membership. With the new server, this has stopped working. If it isn't related to ldm, then I have something to fix. Sincerely, Dave Hopkins From burke at thealmquists.net Sat Apr 4 22:23:51 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 4 Apr 2009 17:23:51 -0500 Subject: [K12OSN] Feedback from a teacher on FC10/LTSP and Tuxpaint In-Reply-To: References: <9E768C62-1E5C-45DC-826D-610081C4EFEF@thealmquists.net> <49D7D922.4020301@scheie.homedns.org> Message-ID: <8A5A7BDA-993C-45ED-8775-BC2AC56BCC99@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 4, 2009, at 5:11 PM, David Hopkins wrote: > On Sat, Apr 4, 2009 at 6:03 PM, Peter Scheie > wrote: >> LDM_DIRECTX=YES encrypts the login but not the X session. LTSP 5 >> does not >> use gdm, as it has no provision for encryption, and passing user >> IDs & PWs >> cleartext is not a good thing, even in a school setting. Setting >> LDM_DIRECTX=NO will encrypt the entire session, which some people >> need, but >> also requires faster hardware in the client. > > So do the same gnome login session scripts run as before? I had users > in groups which assigned them desktop icons (added or removed) based > on group membership. With the new server, this has stopped working. > If it isn't related to ldm, then I have something to fix. > I think this is something you can specify in the lts.conf file with the rcfile parameter. > Sincerely, > Dave Hopkins > > _______________________________________________ > 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) iEYEARECAAYFAknX3fcACgkQxWV7OPa/g5G9SQCggCcK0wDkbjzoxbyKRrIKhlMV MUIAnibzpiLTH9Cluie4SXq2yLjDtAwH =rSqi -----END PGP SIGNATURE----- From rowens at ptd.net Sun Apr 5 01:20:02 2009 From: rowens at ptd.net (Rob Owens) Date: Sat, 4 Apr 2009 21:20:02 -0400 Subject: [K12OSN] Feedback from a teacher on FC10/LTSP and Tuxpaint In-Reply-To: References: Message-ID: <20090405012002.GA24840@aurora.owens.net> On Sat, Apr 04, 2009 at 03:07:23PM -0400, David Hopkins wrote: > > Does the LTSP 5 setup have all X traffic encrypted by chance? > > I thought I had the encryption turned off via > > LDM_DIRECTX=YES > > Is there a way to just not use LDM at all and just use gdm? TBH I am > not that worried about having anything intercepted between the server > and the clients. This is a standard 2 NIC setup. > If you want you can use "startx" as your screen script (SCREEN_07=startx). Don't forget to enable XDMCP on your server. You'll still use ldm, but after login all your traffic will be like it was in LTSP 4.2. My Debian system also has an "xdmcp" screen script. I haven't tried it, but you may want to experiment with it. You could also look at running LTSP 4.2 on certain thin clients, if that's the only way to improve performance. See this: http://wiki.ltsp.org/twiki/bin/view/Ltsp/Ltsp5SameServerLTSP42 -Rob From rowens at ptd.net Sun Apr 5 01:26:30 2009 From: rowens at ptd.net (Rob Owens) Date: Sat, 4 Apr 2009 21:26:30 -0400 Subject: [K12OSN] OT: Testing FC10 results In-Reply-To: References: <8D80BB3F-F9D8-457B-B434-FE2C3215BFF3@thealmquists.net> Message-ID: <20090405012630.GB24840@aurora.owens.net> On Sat, Apr 04, 2009 at 06:05:46PM -0400, David Hopkins wrote: > > > > I thought it might be the LDM_DIRECTX setting, but that is set to TRUE by > > default on Fedora, which means that X traffic isn't encrypted on K12LTSP by > > default. So it must be something else. ?Higher network traffic for some > > reason, like increased screen resolution or color depth? You said the server > > has plenty of memory and CPU, and I don't think the disk IO should change > > that much. > > In terms of bandwidth, this server is performing the same for disk I/O > as the other servers running 4.2 (measured using netio). Screen > resolution is likewise typical. I could force the color depth to 16 > instead of 24. It is confusing. I am using ldap for authentication > and wonder if something changed wrt to how authentication is working, > or if dns has changed? The host table is empty for ltsp5? ltsp4.2 > had an entry for every client. This is the one difference that I know > of since the clients are using 172.x.y.z whereas 4.2 used 192.x.y.z. > I can change this but don't really see why it would have an effect. > There's a possibility that the video driver you're getting with LTSP 5 is not the same as the driver in LTSP 4.2 (in fact, I think it's likely). Perhaps the new driver is causing the slowdown. Do all your thin clients use the same type of video card, or are they all different? My working LTSP 5 systems don't put anything special into /etc/hosts like the LTSP 4.2 systems needed. I'm not sure why, but it seems it's not needed in LTSP 5. Maybe it was a remote X thing (bare X, not X over ssh). -Rob From jthomas at bittware.com Sun Apr 5 03:14:18 2009 From: jthomas at bittware.com (j.w. thomas) Date: Sat, 04 Apr 2009 23:14:18 -0400 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <1443710F-1A29-406B-9630-99ACEF0635A8@thealmquists.net> References: <49D5720B.2020201@bittware.com> <1443710F-1A29-406B-9630-99ACEF0635A8@thealmquists.net> Message-ID: <49D8220A.4040307@bittware.com> Almquist Burke wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > On Apr 2, 2009, at 9:18 PM, j.w. thomas wrote: > >> Hi everyone, >> >> I recently set up a box and installed IPCop on it to serve as our >> firewall. It apparently runs a caching DNS server, but for the life >> of me, I cannot figure out where it gets ITS dns info. And as you >> might guess, I'm having some pretty severe DNS troubles. >> >> Lookups are taking a full minute. That means that any webpage with >> say... five different domains in it takes a minimum of five minutes to >> load. >> > > Are you using LDAP or some other network authentication system by chance? No LDAP, just straight up /etc/passwd & shadow. I have pointed ipcop at OpenDNS now, and it seems much more responsive now. Thanks for the help! -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Anything is possible if you don't know what you are talking about. From josh.fryman at gmail.com Sun Apr 5 03:54:42 2009 From: josh.fryman at gmail.com (Josh Fryman) Date: Sat, 4 Apr 2009 20:54:42 -0700 Subject: [K12OSN] Guide to enabling Local Apps support? In-Reply-To: <49D761A7.6010903@scheie.homedns.org> References: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> <49D761A7.6010903@scheie.homedns.org> Message-ID: <351233d20904042054u348ccf45o7bdfbf542da501b2@mail.gmail.com> Thanks to all for the suggestions. I'm going to try to make this step work tonight, and then I think I'll have a solid enough "demo" network that I can appease any comers. A few questions on the support ability of local-apps. Is it possible to choose when to run it local vs remote? The point Peter raised about frequently not needing flash-based content when browsing is very valid. If I could have Firefox run under two aliases -- one local with flash, one remote without flash -- then that would be a nice trick to use. A related note -- I've seen that GL programs absolutely crawl in LTSP. Is there a way around this other than local-apps? It's probably not the best place to ask, but a few other questions I've run into would be summed up in bullet form: - Is there a good Linux-based gradebook program? Or does everyone just use OOo Calc / Gnumeric? Google searching has failed to turn up anything concrete. - As an "exchange equivalent" backend server, is there a coherent thought in the community? I'm not sure if they current do use Exchange server, but I'm willing to bet they do. The two I've seen are: OpenExchange (http://www.open-xchange.com/en/oxpedia) eGroupWare http://www.egroupware.org/Home) Thanks again to everyone. Cheers, Josh Fryman From josh.fryman at gmail.com Sun Apr 5 06:21:23 2009 From: josh.fryman at gmail.com (Josh Fryman) Date: Sat, 4 Apr 2009 23:21:23 -0700 Subject: [K12OSN] Local-Apps setup for Firefox + Flash + Sound = Success Message-ID: <351233d20904042321n5b1579c1mfb5b6e48bb1b25db@mail.gmail.com> Hi all, Following onto my earlier post, I have updated the local client image, installed local firefox + flash + misc, and now have it running quite nicely with sound. From the earlier thread I was given a link to, it sounds like sound has been the stumbling block. My log of what commands I used are below. I simultaneously connected three clients to the server, ran local Firefox on each, and then streamed various flash content. I also ran a few episodes of The Daily Show and YouTube videos, all with no problem and great sound. The one caveat I've found is that firefox does not "quit" properly from local side when multimedia is running. You have to manually have a "local-apps xterm" up, and then "killall firefox" to get it to die. I haven't quite figured out why, yet. Also, I can report the same problem others have seen -- starting firefox for the first time in any session is quite slow -- depending on the machine, 10-50 seconds. Every restart after that, though, is 1-4 seconds. These clients have between 256M and 1024M local memory, and between P3M-733 and P4-2G processors. Ironically, the weakest processor and memory loads it the fastest. I still see some programs as excruciatingly slow off the server. KTouch, for example, is about 1 character per 2 seconds. (I type very, very fast. This is completely annoying.) Any GL-based program is also incredibly slow. Does anyone know how to improve those, without running them locally? Has anyone tried to use Skype as a local app under LTSP? Thanks, Josh Fryman [[ start with "K12Linux F10 Stable 4 x86-64" ISO on web site, install that, run through README to get final steps configured, add extra programs as desired ]] [[ boot thin clients to test basic stuff all works properly ]] [[ disconnect all thin clients ]] [[ on server machine, login as user, "su -", and then . . . ]] [[ my server is x86-64; the clients are i386 ]] [root at localhost ~]# setarch i386 chroot /opt/ltsp/i386/ bash-3.2# mount /proc bash-3.2# cat /etc/resolv.conf bash-3.2# ping www.google.com bash-3.2# mv /etc/resolv.conf /etc/resolv.conf.orig bash-3.2# cat >> resolv.conf nameserver 192.168.1.1 bash-3.2# ping www.google.com bash-3.2# yum install vim traceroute bash-3.2# rpm -ivh http://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-stable.noarch.rpm bash-3.2# rpm -ivh http://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-stable.noarch.rpm bash-3.2# rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-rpmfusion-* bash-3.2# yum update bash-3.2# yum install firefox bash-3.2# rpm -ivh http://linuxdownload.adobe.com/adobe-release/adobe-release-i386-1.0-1.noarch.rpm bash-3.2# rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-adobe-linux bash-3.2# yum install flash-plugin bash-3.2# yum install java-1.6.0-openjdk java-1.6.0-openjdk-plugin bash-3.2# umount /proc bash-3.2# exit [root at localhost ~]# iptables -L [root at localhost ~]# iptables -F [root at localhost ~]# iptables -t nat -A POSTROUTING -o eth1 -j MASQUERADE [[ NOTE: for my machine, "eth1" is the public Internet interface; "eth0" is tied to ltspbr0; also, 192.168.1.1 is my local nameserver cache on my home LAN. Replace these two fields in the above commands as needed. ]] [[ reboot all thin clients ]] [[ login as a user on a thin client ]] [[ open a shell; run "ltsp-localapps firefox" and enjoy ]] [[ if firefox fails to clean up after itself upon close, open a shell, run "ltsp-localapps xterm"; in the new xterm, run "killall firefox" ]] For those interested, the RPM sources used for flash, etc, are via this article: http://www.mjmwired.net/resources/mjm-fedora-f10.html From asmo.koskinen at arkki.info Sun Apr 5 06:44:44 2009 From: asmo.koskinen at arkki.info (Asmo Koskinen) Date: Sun, 05 Apr 2009 09:44:44 +0300 Subject: [K12OSN] Local-Apps setup for Firefox + Flash + Sound = Success In-Reply-To: <351233d20904042321n5b1579c1mfb5b6e48bb1b25db@mail.gmail.com> References: <351233d20904042321n5b1579c1mfb5b6e48bb1b25db@mail.gmail.com> Message-ID: <49D8535C.1090201@arkki.info> Josh Fryman kirjoitti: > Has anyone tried to use Skype as a local app under LTSP? No, but Mumble/Murmur with usb mic (Asus Eee 701/Zoom H2) Screenshot: http://www.arkki.info/howto/Mumble/Mumble_thin-client_01.png Picture: http://www.arkki.info/howto/Mumble/Mumble_thin-client_02.png Another test was Cheese with Asus Eee 701's own built-in webcam. Screenshot: http://www.arkki.info/howto/Wiki/LTSP5/ltsp-localapps_cheese_01.png ---- I used Ubuntu, but Debian, Ubuntu and Fedora use LTSP5/localapps same way. Best Regards Asmo Koskinen. From peter at scheie.homedns.org Sun Apr 5 17:00:17 2009 From: peter at scheie.homedns.org (Peter Scheie) Date: Sun, 05 Apr 2009 12:00:17 -0500 Subject: [K12OSN] Guide to enabling Local Apps support? In-Reply-To: <351233d20904042054u348ccf45o7bdfbf542da501b2@mail.gmail.com> References: <351233d20904032253x4123a8ffw2f25ebd8f742ae0e@mail.gmail.com> <49D761A7.6010903@scheie.homedns.org> <351233d20904042054u348ccf45o7bdfbf542da501b2@mail.gmail.com> Message-ID: <49D8E3A1.50804@scheie.homedns.org> Josh Fryman wrote: > A few questions on the support ability of local-apps. Is it possible > to choose when to run it local vs remote? The point Peter raised > about frequently not needing flash-based content when browsing is very > valid. If I could have Firefox run under two aliases -- one local > with flash, one remote without flash -- then that would be a nice > trick to use. > You can specify by MAC address which clients do local apps and which don't. Then cluster your local-app clients in the 'lab for watching video'. Peter From joseph.bishay at gmail.com Mon Apr 6 19:13:33 2009 From: joseph.bishay at gmail.com (Joseph Bishay) Date: Mon, 6 Apr 2009 15:13:33 -0400 Subject: [K12OSN] RAID Question Message-ID: Hello, I hope everyone is doing well. I have a question about some trouble I'm having with a RAID controller I'm setting up. I'm creating a backup machine to backup the LTSP server and I get my hands on an ATA IDE RAID Controller - http://www.adaptec.com/en-US/support/raid/ata_raid/AAR-1200A I have installed the raid controller, put in 2X400 GB drives, formatted, partitioned, and created ext3 filesystems, and configured the RAID to run as RAID 1 and created the array. Now 2 issues: 1) The RAID controller reports only 127 GB of space on each drive. There is no bios update for the RAID controller 2) Booting into CentOS actually finds both drives independently, not as a RAID -- one is /dev/hde and one is /dev/hdg -- both report correctly 400 GB and both, when mounted can act independently of each other. Creating a file on one affects only that one drive, not the other. Typing lspci shows me: 02:0b.0 Mass storage controller: HighPoint Technologies, Inc. HPT366/368/370/370A/372/372N (rev 03) that is the chip on the RAID controller so the hardware is there and seems to be working. Am I missing a step to get the drives to act as a RAID? I thought hardware raid was only suppose to report the unified array to the OS so it wouldn't see two drives? Thank you Joseph From dyoung at mesd.k12.or.us Mon Apr 6 20:06:58 2009 From: dyoung at mesd.k12.or.us (Dan Young) Date: Mon, 6 Apr 2009 13:06:58 -0700 Subject: [K12OSN] RAID Question In-Reply-To: References: Message-ID: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> On Mon, Apr 6, 2009 at 12:13 PM, Joseph Bishay wrote: > that is the chip on the RAID controller so the hardware is there and > seems to be working. > > Am I missing a step to get the drives to act as a RAID? I thought > hardware raid was only suppose to report the unified array to the OS > so it wouldn't see two drives? http://linuxmafia.com/faq/Hardware/sata.html#fakeraid You didn't mention your distro, but support for these quasi-RAID chips varies widely w/ your kernel and dmraid version. The dmraid tools are probably required here. Good luck! -- Dan Young Multnomah ESD - Technology Services 503-257-1562 From joseph.bishay at gmail.com Mon Apr 6 20:13:17 2009 From: joseph.bishay at gmail.com (Joseph Bishay) Date: Mon, 6 Apr 2009 16:13:17 -0400 Subject: [K12OSN] RAID Question In-Reply-To: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> Message-ID: Hello, On Mon, Apr 6, 2009 at 4:06 PM, Dan Young wrote: >> Am I missing a step to get the drives to act as a RAID? I thought >> hardware raid was only suppose to report the unified array to the OS >> so it wouldn't see two drives? > > http://linuxmafia.com/faq/Hardware/sata.html#fakeraid > > You didn't mention your distro, but support for these quasi-RAID chips > varies widely w/ your kernel and dmraid version. The dmraid tools are > probably required here. Good luck! I am running Centos 5.3 on this machine. The link you posted recommended not using the RAID controller at all but using software raid -- I was always under the impression that hardware was better in all circumstances. I will try using dmraid now. Thanks Joseph From tnelson at rockbochs.com Mon Apr 6 20:12:47 2009 From: tnelson at rockbochs.com (Tim Nelson) Date: Mon, 6 Apr 2009 15:12:47 -0500 (CDT) Subject: [K12OSN] RAID Question In-Reply-To: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> Message-ID: <14528377.1181239048767777.JavaMail.root@zmail.rockbochs.com> ----- "Dan Young" wrote: > On Mon, Apr 6, 2009 at 12:13 PM, Joseph Bishay > wrote: > > that is the chip on the RAID controller so the hardware is there > and > > seems to be working. > > > > Am I missing a step to get the drives to act as a RAID? I thought > > hardware raid was only suppose to report the unified array to the > OS > > so it wouldn't see two drives? > > http://linuxmafia.com/faq/Hardware/sata.html#fakeraid > > You didn't mention your distro, but support for these quasi-RAID > chips > varies widely w/ your kernel and dmraid version. The dmraid tools are > probably required here. Good luck! > > -- > Dan Young > Multnomah ESD - Technology Services > 503-257-1562 > In that case I'd be very inclined to remove the RAID config from the cards 'firmware' and just run pure Linux software RAID. Tim Nelson Systems/Network Support Rockbochs Inc. (218)727-4332 x105 From microman at cmosnetworks.com Tue Apr 7 03:47:07 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Mon, 06 Apr 2009 23:47:07 -0400 Subject: [K12OSN] RAID Question In-Reply-To: References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> Message-ID: <49DACCBB.7070703@cmosnetworks.com> _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! Joseph Bishay wrote: > Hello, > > > On Mon, Apr 6, 2009 at 4:06 PM, Dan Young wrote: > >>> Am I missing a step to get the drives to act as a RAID? I thought >>> hardware raid was only suppose to report the unified array to the OS >>> so it wouldn't see two drives? >>> >> http://linuxmafia.com/faq/Hardware/sata.html#fakeraid >> >> You didn't mention your distro, but support for these quasi-RAID chips >> varies widely w/ your kernel and dmraid version. The dmraid tools are >> probably required here. Good luck! >> > > I am running Centos 5.3 on this machine. The link you posted > recommended not using the RAID controller at all but using software > raid -- I was always under the impression that hardware was better in > all circumstances. > > I will try using dmraid now. > > Thanks > Joseph > Well, they're right...IF it's a *real* hardware RAID. Unfortunately, a lot of these aren't. A good example of a real hardware RAID is the LSI Logic 150-6 SATA, which I use. Virtually all SCSI RAID cards are real hardware RAID, with the exception of the LSI 8200 series (8300 and up are the real McCoy). In your case, unless your budget allows for buying a real card, I'd go with Linux's own software RAID. --TP -------------- next part -------------- An HTML attachment was scrubbed... URL: From gianugo.altieri at tiscali.it Tue Apr 7 19:12:05 2009 From: gianugo.altieri at tiscali.it (Gianugo Altieri) Date: Tue, 07 Apr 2009 21:12:05 +0200 Subject: [K12OSN] Introducing ourselves Message-ID: <49DBA585.20904@tiscali.it> > If the printer is able to work on the server then you can configure it > to work on client. You will needa client with USB2 sockets. I am > working four/five printers in my office like this. > Thanks for your help. I tried again the way you suggested, to no result. I cabled our HP1015 to the server and it immediately recognized it and added it to the printers list. But when I disconnected it from the server and connected to one of the clients' USB2 port, the printer, although still listed, appeared as "disconnected - check cable" and didn't work. I must say my K12 is plain vanilla, i.e. I made no special configuration. Did you change any setting for letting printers work? > Not very clear but if you want to open a session on server from client > you can do that by some thing like: > > Xnest -indirect serverip :1 > > Or from client do ssh -X serverip and then after log in run the > command. In simple terms the -X allows graphical pass through. > > If you need to run from server on the client something then you will > have to configure client to make sure that it allows remote logins and > also forwarding of XDMCP. > > HTH > This was really useful to me. I managed to ping the server, opened an ssh terminal session on it, and, by running "ssh -X" command as you suggested I could open graphic programs on my desktop. I didn't know of this possibility, and it's really useful. Probably "Xnest" is what I was looking for, to work on the server, but unfortunately I get a "Xnest: malloc() memory corruption... " followed by a deluge of hexadecimals and other stuff. The good news is that from a Windows machine running CygWin I can get perfect remote login screens on the same server. So server configuration appear correct, but client has some problems. I'll try to remove and download again Xnest to see if the problem disappears. Thanks again Gianugo Altieri From robark at gmail.com Wed Apr 8 06:34:56 2009 From: robark at gmail.com (Robert Arkiletian) Date: Tue, 7 Apr 2009 23:34:56 -0700 Subject: [K12OSN] Fl_TeacherTool Instructional videos for teachers Message-ID: These videos are made using Fl_TeacherTool 0.70 on K12LTSP 5EL Snapshots and right click menu items http://rark.blip.tv/#1971024 Broadcasting and Internet Access control http://rark.blip.tv/#1975162 Sending files, messages and controlling user processes http://rark.blip.tv/#1975099 Group filter http://rark.blip.tv/#1975204 Room filter http://rark.blip.tv/#1975241 -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From robark at gmail.com Wed Apr 8 06:43:47 2009 From: robark at gmail.com (Robert Arkiletian) Date: Tue, 7 Apr 2009 23:43:47 -0700 Subject: [K12OSN] Re: Fl_TeacherTool Instructional videos for teachers In-Reply-To: References: Message-ID: Let me try that again. On Tue, Apr 7, 2009 at 11:34 PM, Robert Arkiletian wrote: > These videos are made using Fl_TeacherTool 0.70 on K12LTSP 5EL > > Snapshots and right click menu items > http://rark.blip.tv/#1971024 http://blip.tv/file/1961314 > > Broadcasting and Internet Access control > http://rark.blip.tv/#1975162 http://blip.tv/file/1965218 > > Sending files, messages and controlling user processes > http://rark.blip.tv/#1975099 http://blip.tv/file/1965155 > > Group filter > http://rark.blip.tv/#1975204 http://blip.tv/file/1965260 > > Room filter > http://rark.blip.tv/#1975241 http://blip.tv/file/1965297 -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From gianugo.altieri at tiscali.it Wed Apr 8 15:27:08 2009 From: gianugo.altieri at tiscali.it (Gianugo Altieri) Date: Wed, 08 Apr 2009 17:27:08 +0200 Subject: [K12OSN] Introducing ourselves Message-ID: <49DCC24C.9020500@tiscali.it> Hi Terrell, thanks for your help. Sure I will follow your suggestion and, in future, buy printers with on-board Ethernet, but for the time being I'm stuck to the current HP LJ 1015 and CM 1312, and there's no budget for more... A possibility is to let a chubby terminal to act as a printing server. In fact, my Fedora 8 PC that shares the CM 1312 printer is visible by both server and thin clients and everyone can print on it. Yet, it still tastes of centralized printing, not exactly the same as enjoying local printers. Terrell Prud? Jr. ha scritto: > Gianugo Altieri wrote: >> I have a couple of wishes in my list, for K12. First of all, I >> would like to be able to connect printers to thin clients. In fact, >> for safety reasons, server resides in a closed room, and it would be >> impractical to connect a printer to it. A convenient location for >> printers would be beside clients, but it seems clients cannot "see" >> USB printers, or at least our HP laser. Strangely enough, clients >> recognize USB pen-drives, but not USB external DVD readers, so I >> can't figure out why some USB peripherals are working ad some aren't. > > Hello and welcome, Gianugo, > > An easier idea might be to get some network printers, i. e. printers > with network cards in them. Then, you can place the printers wherever > you want, including beside a client if you wish, but they would still > be "connected" to the K12 server through Ethernet. This is how we > have our printers set up at work, and I do the same at home. It's great. > > --TP > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From gianugo.altieri at tiscali.it Wed Apr 8 15:29:30 2009 From: gianugo.altieri at tiscali.it (Gianugo Altieri) Date: Wed, 08 Apr 2009 17:29:30 +0200 Subject: [K12OSN] Introducing ourselves Message-ID: <49DCC2DA.8070206@tiscali.it> Terrell, I did what you suggest (interesting indeed) to get remote login screen on K12 server from Fedora 8 chubby client. Since I use Gnome, I asked the F8 start-up screen to show me the menu for remote logins (what you do with KDE), and it did. After a while, the machine came up with a list of remote login servers, including itself (pseudo-remote, actually local) and the K12 server. So I asked for the latter, and, at first, it seemed it worked, it switched to a different graphic mode, it showed briefly a cursor with a "wait" icon, but then it suddenly dropped me to the local login screen, with no error message, and I have no idea of where to look for a log file. I repeated the same operation many times, to no result. I then tried rdesktop-xdmcp combination, and, after showing a wrist-watch icon for a while, it sent me out with a "xnest:malloc() ... " allocation error. By the way, I operated with both server and client firewalls off. Back in the times of Fedora 8, there was that handy program that allowed graphical configuration of local and remote logins, and worked great. Now, since F10-K12 don't have anything similar, I copied /etc/gdm/custom.conf from Fedora 8 to Fedora 10, but this isn't enough for get it running. "gdmgreeter" file is no longer available, so I tried "gdm-simple-greeter" and other greeters, with no success. Thanks for your help Gianugo Altieri Terrell Prud? Jr. ha scritto: > Gianugo Altieri wrote: >> Second problem is a fat client I wish to link. It runs Fedora 8 >> with some legacy software. I'm trying to let it "open a window" on >> the server for a graphical login, and also vice versa, i.e. let the >> server (and therefore the thin clients) open a graphical login on it. >> The problem is that I don't know how to do it. If I connect the fat >> client to LTSP LAN, it grabs an IP, but that's all. It doesn't even >> "ping" the server, let alone opening a remote desktop on it. Any >> suggestion? > > It should certainly be able to ping the server, *if* this fat client > is also on the LTSP client network. If not, i. e. it's on the main > company LAN and you're using the classic two-NIC LTSP setup, then you > have two choices: > > 1.) poke a hole in the K12 server's firewall rules for SSH and X11, or > 2.) place the Fedora 8 fat client on the LTSP client network. > > I would recommend the second choice, for two reasons. First, you > don't need to poke any holes in the built-in packet filter for X11 > (proper security is always good). Second, it's just lower maintenance. > > Now, once you can ping the K12 server, here's how I would handle > graphical logins. I will use KDE as an example, since that's what I > happen to run. From your K Menu, choose "Switch user", and choose > "Start New Session". You'll get a new graphical login screen, and one > of the options at the lower left are two options--"Session Type" and > "Menu". Choose "Menu". Of the choices, one of them is "Remote > login". That's the one you want. Just point that at your K12 > server's IP address, and boom, you have a K12 login prompt. > Meanwhile, your "original" Fedora 8 session is still running. To > switch between them, you use F7 (to go back) and F8 (to go forward). > > I'm not sure how easy it would be to have the clients be able to do > the same to the Fedora 8 box, because I've never done that. But > fundamentally, it should be possible. It might take some fiddling > with some of the configuration files under /opt/ltsp/i386...or if your > client has X11 with XDMCP built in, you should be able to configure > the client directly. What you want at that point is what's called an > "indirect" X11 query. Do some Googling for that--"indirect query", > XDMCP, and X11. > > --TP > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From microman at cmosnetworks.com Wed Apr 8 16:16:48 2009 From: microman at cmosnetworks.com (=?ISO-8859-15?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 08 Apr 2009 12:16:48 -0400 Subject: [K12OSN] Introducing ourselves In-Reply-To: <49DCC24C.9020500@tiscali.it> References: <49DCC24C.9020500@tiscali.it> Message-ID: <49DCCDF0.60300@cmosnetworks.com> Ah yes, budgetary issues do get in our way sometimes, don't they? :-) Fortunately, there are inexpensive devices called "print servers" that can turn any printer with a parallel or USB interface into a network printer. This is very useful for "older" printers that don't have on-board Ethernet. The device that I use is the TrendNet TE100 series. Since my printer is an HP LaserJet 6L (a parallel printer), I use the model TE100-P1P. For USB printers, you'd use the TE100-P1U. Either version should be less than US $50.00. I found my TE100-P1P for US $35 on sale last year, and the devices have proved to be very Linux-friendly. Here's a link. They have them in both wireless and wired Ethernet. I much, MUCH prefer the wired versions. http://www.trendnet.com/products/products.asp?cat=46 --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! Gianugo Altieri wrote: > > Hi Terrell, > > thanks for your help. Sure I will follow your suggestion and, in > future, buy printers with on-board Ethernet, but for the time being > I'm stuck to the current HP LJ 1015 and CM 1312, and there's no budget > for more... A possibility is to let a chubby terminal to act as a > printing server. In fact, my Fedora 8 PC that shares the CM 1312 > printer is visible by both server and thin clients and everyone can > print on it. Yet, it still tastes of centralized printing, not exactly > the same as enjoying local printers. > > Terrell Prud? Jr. ha scritto: >> Gianugo Altieri wrote: >>> I have a couple of wishes in my list, for K12. First of all, I >>> would like to be able to connect printers to thin clients. In fact, >>> for safety reasons, server resides in a closed room, and it would be >>> impractical to connect a printer to it. A convenient location for >>> printers would be beside clients, but it seems clients cannot "see" >>> USB printers, or at least our HP laser. Strangely enough, clients >>> recognize USB pen-drives, but not USB external DVD readers, so I >>> can't figure out why some USB peripherals are working ad some aren't. >> >> Hello and welcome, Gianugo, >> >> An easier idea might be to get some network printers, i. e. printers >> with network cards in them. Then, you can place the printers >> wherever you want, including beside a client if you wish, but they >> would still be "connected" to the K12 server through Ethernet. This >> is how we have our printers set up at work, and I do the same at >> home. It's great. >> >> --TP >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see -------------- next part -------------- An HTML attachment was scrubbed... URL: From Marc.Fromm at wwu.edu Wed Apr 8 17:01:51 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Wed, 8 Apr 2009 10:01:51 -0700 Subject: [K12OSN] k12ltsp extras repository error Message-ID: I upgraded my FC6 k12ltsp system to FC7. When I run yum update I get this error message: Loading "installonlyn" plugin Setting up Update Process http://download.fedora.redhat.com/pu...ta/repomd.xml: [Errno 14] HTTP Error 404: Date: Wed, 08 Apr 2009 16:02:35 GMT Server: Apache Content-Length: 248 Content-Type: text/html; charset=iso-8859-1 Trying other mirror. Error: Cannot open/read repomd.xml file for repository: extras The above error is generated by the k12ltsp-extras repository. Is it down? Is there a new repository to replace it? Thanks Marc -------------- next part -------------- An HTML attachment was scrubbed... URL: From nils at breun.nl Wed Apr 8 17:30:59 2009 From: nils at breun.nl (Nils Breunese) Date: Wed, 8 Apr 2009 19:30:59 +0200 Subject: [K12OSN] k12ltsp extras repository error In-Reply-To: References: Message-ID: <21047215-EF58-489F-B0E8-2C9DF36618E1@breun.nl> Marc Fromm wrote: > I upgraded my FC6 k12ltsp system to FC7. When I run yum update I get > this error message: > > Loading "installonlyn" plugin > Setting up Update Process > http://download.fedora.redhat.com/pu...ta/repomd.xml: [Errno 14] > HTTP Error 404: Date: Wed, 08 Apr 2009 16:02:35 GMT > Server: Apache > Content-Length: 248 > Content-Type: text/html; charset=iso-8859-1 > > Trying other mirror. > Error: Cannot open/read repomd.xml file for repository: extras > > The above error is generated by the k12ltsp-extras repository. > Is it down? Is there a new repository to replace it? Fedora Core 6 reached end of life on December 7 2007 and Fedora 7 reached end of life on June 13 2008, so neither are supported anymore. Currently only K12LTSP 5EL (based on CentOS 5 and LTSP 4.2) is still supported with updates. K12Linux is the new Fedora-based LTSP distribution, based on Fedora 10 and LTSP 5. Nils Breunese. From microman at cmosnetworks.com Wed Apr 8 17:58:21 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 08 Apr 2009 13:58:21 -0400 Subject: [K12OSN] k12ltsp extras repository error In-Reply-To: <21047215-EF58-489F-B0E8-2C9DF36618E1@breun.nl> References: <21047215-EF58-489F-B0E8-2C9DF36618E1@breun.nl> Message-ID: <49DCE5BD.1080909@cmosnetworks.com> Nils Breunese wrote: > Marc Fromm wrote: > >> I upgraded my FC6 k12ltsp system to FC7. When I run yum update I get >> this error message: >> >> Loading "installonlyn" plugin >> Setting up Update Process >> http://download.fedora.redhat.com/pu...ta/repomd.xml: [Errno 14] HTTP >> Error 404: Date: Wed, 08 Apr 2009 16:02:35 GMT >> Server: Apache >> Content-Length: 248 >> Content-Type: text/html; charset=iso-8859-1 >> >> Trying other mirror. >> Error: Cannot open/read repomd.xml file for repository: extras >> >> The above error is generated by the k12ltsp-extras repository. >> Is it down? Is there a new repository to replace it? > > Fedora Core 6 reached end of life on December 7 2007 and Fedora 7 > reached end of life on June 13 2008, so neither are supported anymore. > Currently only K12LTSP 5EL (based on CentOS 5 and LTSP 4.2) is still > supported with updates. K12Linux is the new Fedora-based LTSP > distribution, based on Fedora 10 and LTSP 5. > > Nils Breunese. > +1. Please upgrade to K12LTSP 5EL or K12Linux. --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.bishay at gmail.com Wed Apr 8 21:36:17 2009 From: joseph.bishay at gmail.com (Joseph Bishay) Date: Wed, 8 Apr 2009 17:36:17 -0400 Subject: [K12OSN] RAID Question In-Reply-To: <49DACCBB.7070703@cmosnetworks.com> References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> Message-ID: Hello, As per everyone's recommendation, I removed the fakeraid hardware and plugged the drives right into the motherboard and used the built-in Linux software RAID. Everything seems to be working very well! I appreciate everyone's recommendation on this issue. I have a follow-up question -- the RAID is made up of 2 refurbished drives we were able to obtain (cost issues). I'd like to find a command / program that I can use to test them / stress them while they are still in their warranty period (90 days). I've done some research and came across a lot of options: Bonnie++, badblocks, fsck, smartctl -t, etc etc. I'm not sure what is best to use -- I guess I want something that will check to see if there are any existing current issues and something to put the drives under some tests to see if any issues arise. The other side of that question is: will the drives being in RAID affect these tests at all? Should I run the tests on the RAID array or should I dismantle the array, run the tests, then re-enable the array? Thank you! Joseph From Marc.Fromm at wwu.edu Wed Apr 8 22:40:26 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Wed, 8 Apr 2009 15:40:26 -0700 Subject: [K12OSN] k12ltsp extras repository error In-Reply-To: <49DCE5BD.1080909@cmosnetworks.com> References: <21047215-EF58-489F-B0E8-2C9DF36618E1@breun.nl> <49DCE5BD.1080909@cmosnetworks.com> Message-ID: Can I upgrade directly from FC7 to K12Linux or does it require a fresh install? Thanks ________________________________ From: k12osn-bounces at redhat.com [mailto:k12osn-bounces at redhat.com] On Behalf Of "Terrell Prud? Jr." Sent: Wednesday, April 08, 2009 10:58 AM To: Support list for open source software in schools. Subject: Re: [K12OSN] k12ltsp extras repository error Nils Breunese wrote: Marc Fromm wrote: I upgraded my FC6 k12ltsp system to FC7. When I run yum update I get this error message: Loading "installonlyn" plugin Setting up Update Process http://download.fedora.redhat.com/pu...ta/repomd.xml: [Errno 14] HTTP Error 404: Date: Wed, 08 Apr 2009 16:02:35 GMT Server: Apache Content-Length: 248 Content-Type: text/html; charset=iso-8859-1 Trying other mirror. Error: Cannot open/read repomd.xml file for repository: extras The above error is generated by the k12ltsp-extras repository. Is it down? Is there a new repository to replace it? Fedora Core 6 reached end of life on December 7 2007 and Fedora 7 reached end of life on June 13 2008, so neither are supported anymore. Currently only K12LTSP 5EL (based on CentOS 5 and LTSP 4.2) is still supported with updates. K12Linux is the new Fedora-based LTSP distribution, based on Fedora 10 and LTSP 5. Nils Breunese. +1. Please upgrade to K12LTSP 5EL or K12Linux. --TP _______________________________ Do you GNU? Microsoft Free since 2003--the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From wtogami at redhat.com Wed Apr 8 22:56:51 2009 From: wtogami at redhat.com (Warren Togami) Date: Wed, 08 Apr 2009 18:56:51 -0400 Subject: [K12OSN] k12ltsp extras repository error In-Reply-To: References: <21047215-EF58-489F-B0E8-2C9DF36618E1@breun.nl> <49DCE5BD.1080909@cmosnetworks.com> Message-ID: <49DD2BB3.3090201@redhat.com> On 04/08/2009 06:40 PM, Marc Fromm wrote: > Can I upgrade directly from FC7 to K12Linux or does it require a fresh > install? Fresh. Warren From wtogami at redhat.com Wed Apr 8 23:08:32 2009 From: wtogami at redhat.com (Warren Togami) Date: Wed, 08 Apr 2009 19:08:32 -0400 Subject: [K12OSN] k12ltsp extras repository error In-Reply-To: <49DD2BB3.3090201@redhat.com> References: <21047215-EF58-489F-B0E8-2C9DF36618E1@breun.nl> <49DCE5BD.1080909@cmosnetworks.com> <49DD2BB3.3090201@redhat.com> Message-ID: <49DD2E70.7020800@redhat.com> On 04/08/2009 06:56 PM, Warren Togami wrote: > On 04/08/2009 06:40 PM, Marc Fromm wrote: >> Can I upgrade directly from FC7 to K12Linux or does it require a fresh >> install? > > Fresh. > > Warren > https://fedorahosted.org/k12linux/wiki/LiveServer Try the Live Server in demo mode to see if it suits your needs, before reinstalling your server. You might decide that K12LTSP EL5 better supports your thin client hardware. K12Linux has better support of newer thin client hardware. Warren From missive at hotmail.com Thu Apr 9 00:06:34 2009 From: missive at hotmail.com (Lee Harr) Date: Thu, 9 Apr 2009 04:36:34 +0430 Subject: [K12OSN] does recordmydesktop work on thin clients? Message-ID: I'm looking for a way for students to record things on their desktop sessions, but I'm not sure what is the best way to go. Has anyone got recordmydesktop working on thin clients? When I tried it, I got this error: $ recordmydesktop Initial recording window is set to: X:0 Y:0 Width:1024 Height:768 Adjusted recording window is set to: X:0 Y:0 Width:1024 Height:768 Your window manager appears to be KWin Initializing... Buffer size adjusted to 4096 from 4096 frames. Capturing! X Error: BadAccess (attempt to access private resource denied) I tried "xhost +" but get the same error. Anyone know how to get this working, or have a suggestion for another program to try? Thanks for your time. _________________________________________________________________ News, entertainment and everything you care about at Live.com. Get it now! http://www.live.com/getstarted.aspx From microman at cmosnetworks.com Thu Apr 9 05:58:42 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Thu, 09 Apr 2009 01:58:42 -0400 Subject: [K12OSN] RAID Question In-Reply-To: References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> Message-ID: <49DD8E92.7010104@cmosnetworks.com> Joseph Bishay wrote: > Hello, > > As per everyone's recommendation, I removed the fakeraid hardware and > plugged the drives right into the motherboard and used the built-in > Linux software RAID. Everything seems to be working very well! I > appreciate everyone's recommendation on this issue. > > I have a follow-up question -- the RAID is made up of 2 refurbished > drives we were able to obtain (cost issues). I'd like to find a > command / program that I can use to test them / stress them while they > are still in their warranty period (90 days). I've done some research > and came across a lot of options: Bonnie++, badblocks, fsck, smartctl > -t, etc etc. > > I'm not sure what is best to use -- I guess I want something that will > check to see if there are any existing current issues and something to > put the drives under some tests to see if any issues arise. > > The other side of that question is: will the drives being in RAID > affect these tests at all? Should I run the tests on the RAID array or > should I dismantle the array, run the tests, then re-enable the array? > > Thank you! > Joseph > Here's what I used to do on Windows NT boxes back in the day to stress-test a hard disk. The same idea would apply to any other OS. 1.) Install the hard disk into your system. 2.) Make two partitions of about equal size on the hard disk. Use fdisk or whatever you prefer. 3.) Format both partitions *simultaneously*. That means start one format command, and then start the second one very shortly thereafter. In UNIX parlance, that'd be "make filesystems on each partition simultaneously." What will happen when you start that second "mke2fs" command is that you will hear the read/write head going back 'n' forth like crazy. I would just do that over and over again for about a week, on each disk, on both partitions. This is, of course, scriptable, so you can set these "mke2fs" commands in a do-forever loop and just let it do its thing. No, there's nothing saying you can't install both hard disks and have both of them subjected to this treatment at the same time. Probably save you some time to do so. Yes, I am cruel. And I'm proud of it. :-) --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From burke at thealmquists.net Thu Apr 9 17:06:54 2009 From: burke at thealmquists.net (Almquist Burke) Date: Thu, 9 Apr 2009 12:06:54 -0500 Subject: [K12OSN] RAID Question In-Reply-To: References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 8, 2009, at 4:36 PM, Joseph Bishay wrote: > > I have a follow-up question -- the RAID is made up of 2 refurbished > drives we were able to obtain (cost issues). I'd like to find a > command / program that I can use to test them / stress them while they > are still in their warranty period (90 days). I've done some research > and came across a lot of options: Bonnie++, badblocks, fsck, smartctl > -t, etc etc. > > I'm not sure what is best to use -- I guess I want something that will > check to see if there are any existing current issues and something to > put the drives under some tests to see if any issues arise. If you are willing to part with about $80, there's always Spinrite from GRC.com. You'll need windows (I used wine) to run the self extractor and have it create either an iso or floppy image. If you are going to be checking/recovering hard drives with any frequency, then it's worth the investment, assuming you are ok with buying non- free software (it IS DRM free). -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAkneKy4ACgkQxWV7OPa/g5EtZwCfeNRIioKn+hiH3xVH83GRwdSQ n3YAnj+5EUqxsvHxX4VbnFteXYD9IaGV =ZN1F -----END PGP SIGNATURE----- From lesmikesell at gmail.com Thu Apr 9 17:22:48 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Thu, 09 Apr 2009 12:22:48 -0500 Subject: [K12OSN] RAID Question In-Reply-To: References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> Message-ID: <49DE2EE8.40109@gmail.com> Joseph Bishay wrote: > > I have a follow-up question -- the RAID is made up of 2 refurbished > drives we were able to obtain (cost issues). I'd like to find a > command / program that I can use to test them / stress them while they > are still in their warranty period (90 days). I've done some research > and came across a lot of options: Bonnie++, badblocks, fsck, smartctl > -t, etc etc. > > I'm not sure what is best to use -- I guess I want something that will > check to see if there are any existing current issues and something to > put the drives under some tests to see if any issues arise. > > The other side of that question is: will the drives being in RAID > affect these tests at all? Should I run the tests on the RAID array or > should I dismantle the array, run the tests, then re-enable the array? I consider hard drives to be about as predictable as light bulbs in terms of guessing when they are going to die. That is, they either work or they don't - and tomorrow it might be different. A simple surface scan (like 'cat /dev/sd? >/dev/null') will tell you if anything is bad today and (after that) 'smartctl -a' will give you the drive's own best guess about tomorrow. -- Les Mikesell lesmikesell at gmail.com From DLWillson at TheGeek.NU Thu Apr 9 17:23:48 2009 From: DLWillson at TheGeek.NU (David L. Willson) Date: Thu, 9 Apr 2009 11:23:48 -0600 (MDT) Subject: [K12OSN] RAID Question In-Reply-To: <49DACCBB.7070703@cmosnetworks.com> Message-ID: <28629977.96481239297828486.JavaMail.root@zimbra.thegeek.nu> > raid -- I was always under the impression that hardware was better in > all circumstances. If you're willing to become intermediate (or expert) with the mdadm family, software RAID is faster for cheaper, and more capabler. - The processor on a RAID controller is usually slower than your main CPU. - There is less memory on your RAID controller than on your motherboard. - Few RAID controllers support RAID6, still. Recoverability: In case of a controller "smoke out", your pretty hosed with hardware RAID, but with software RAID, you drop your drives into any recent Linux box with md tools and enough SATA ports, re-detect the array and get back to work. The best and highest combination of low cost, features, performance, reliability, manageability, and recoverability might be OpenFiler, in combination with a thorough understanding of mdadm, LVM, and their friends. From robark at gmail.com Thu Apr 9 17:46:20 2009 From: robark at gmail.com (Robert Arkiletian) Date: Thu, 9 Apr 2009 10:46:20 -0700 Subject: [K12OSN] RAID Question In-Reply-To: References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> Message-ID: On Wed, Apr 8, 2009 at 2:36 PM, Joseph Bishay wrote: > Hello, > > As per everyone's recommendation, I removed the fakeraid hardware and > plugged the drives right into the motherboard and used the built-in > Linux software RAID. Everything seems to be working very well! I > appreciate everyone's recommendation on this issue. > > I have a follow-up question -- the RAID is made up of 2 refurbished > drives we were able to obtain (cost issues). I'd like to find a > command / program that I can use to test them / stress them while they > are still in their warranty period (90 days). I've done some research > and came across a lot of options: Bonnie++, badblocks, fsck, smartctl > -t, etc etc. I use badblocks to do a destructive test which flips each bit 101 then 010 then 111 then 000. This can take 24 hours. After that I test the disk controller by formatting them and then I grab a Big iso (5GB) copy it to each drive. Then take an MD5SUM of each iso (make sure they are the same) Now copy each iso to the other drive at the same time. So each drive has reads and writes simultaneously. Then check the MD5SUM of each again. If it passes I then setup software raid 1. You can use a live CD to do all this. > > I'm not sure what is best to use -- I guess I want something that will > check to see if there are any existing current issues and something to > put the drives under some tests to see if any issues arise. > > The other side of that question is: will the drives being in RAID > affect these tests at all? Should I run the tests on the RAID array or > should I dismantle the array, run the tests, then re-enable the array? > > Thank you! > Joseph > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From ltsp at dunc-it.com Thu Apr 9 17:50:00 2009 From: ltsp at dunc-it.com (C. Duncan Hudson) Date: Thu, 09 Apr 2009 13:50:00 -0400 Subject: [K12OSN] No Frames until Desktop Effects fail Message-ID: <49DE3548.8000509@dunc-it.com> I have several terminals that are successfully using desktop effects - they're all running Diskless Workstations 1500 series terminals. I have several other terminals, all Diskless Workstation 1200 terminals, that don't appear to be able to take advantage of desktop effects. I'm fine with that - if the hardware can't support it then no bigee, but I'm not fine with how things are working. When the 1200s boot all of the apps load frameless (can't resize, no min / max /close buttons). In order to get the frames back I have to run Desktop Effects (System, Preferences, Look and Feel) and select Enable Desktop Effects. After the terminal attempts to enable effects, and fails, then my frames get created. This is really a pain, as it has to be done after each reboot. Is there a way to make the selection permanent? Thanks in advance, Dunc From lesmikesell at gmail.com Thu Apr 9 18:16:37 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Thu, 09 Apr 2009 13:16:37 -0500 Subject: [K12OSN] RAID Question In-Reply-To: References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> Message-ID: <49DE3B85.6030706@gmail.com> Robert Arkiletian wrote: > After that I test > the disk controller by formatting them... This may be a superstition going back to the old days of having tracking information laid down in the formatting step - and controllers with slight differences, but I always do a low-level format on scsi drives when I move them to a machine with a different controller type or if I don't know where they were before. It probably really isn't necessary any more but it does give the drive a chance to remap any questionable blocks transparently. With IDE and SATA, I'm not sure if this is even possible. -- Les Mikesell lesmikesell at gmail.com From jkinney at localnetsolutions.com Thu Apr 9 18:32:16 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Thu, 09 Apr 2009 14:32:16 -0400 Subject: [K12OSN] RAID Question In-Reply-To: <49DD8E92.7010104@cmosnetworks.com> References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> <49DD8E92.7010104@cmosnetworks.com> Message-ID: <1239301936.20968.9.camel@merlin.localnetsolutions.com> On Thu, 2009-04-09 at 01:58 -0400, "Terrell Prud? Jr." wrote: > 1.) Install the hard disk into your system. > 2.) Make two partitions of about equal size on the hard disk. Use > fdisk or whatever you prefer. > 3.) Format both partitions *simultaneously*. That means start one > format command, and then start the second one very shortly thereafter. > In UNIX parlance, that'd be "make filesystems on each partition > simultaneously." > > What will happen when you start that second "mke2fs" command is that > you will hear the read/write head going back 'n' forth like crazy. I > would just do that over and over again for about a week, on each disk, > on both partitions. This is, of course, scriptable, so you can set > these "mke2fs" commands in a do-forever loop and just let it do its > thing. "dd" is a serious stress tester. It will choke up an error if it can't read or write. 2 drives: sda sdb. Create a single partition on each drive the full size of each drive or the smaller of the two drives (each partition should be exactly the same size) #put data on drive - don't use /dev/random as it'll take 20+ years! dd if=/dev/urandom of=/dev/sda1 && dd if=/dev/sda1 of=/dev/sdb1 dd if=/dev/sdb1 of=/dev/sda1 # repeat as desired - this is good if you have ability to read drive temperatures as well. -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From julius at turtle.com Thu Apr 9 19:49:34 2009 From: julius at turtle.com (Julius Szelagiewicz) Date: Thu, 9 Apr 2009 15:49:34 -0400 (EDT) Subject: [K12OSN] RAID Question In-Reply-To: <49DE3B85.6030706@gmail.com> References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> <49DE3B85.6030706@gmail.com> Message-ID: <34640.216.216.171.238.1239306574.squirrel@216.216.171.238> > Robert Arkiletian wrote: >> After that I test >> the disk controller by formatting them... > > This may be a superstition going back to the old days of having tracking > information laid down in the formatting step - and controllers with > slight differences, but I always do a low-level format on scsi drives > when I move them to a machine with a different controller type or if I > don't know where they were before. It probably really isn't necessary > any more but it does give the drive a chance to remap any questionable > blocks transparently. With IDE and SATA, I'm not sure if this is even > possible. > > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > Yes, it is possible. Spinrite from Gibson Research is an unbelievably good product. I've been using it in its evolving incantations forever (since my first IBM PC XT) and it works. It is especially good at fixing track layout to mach the drift of the heads. Worth the money and then some. julius From lesmikesell at gmail.com Thu Apr 9 19:19:15 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Thu, 09 Apr 2009 14:19:15 -0500 Subject: [K12OSN] RAID Question In-Reply-To: <34640.216.216.171.238.1239306574.squirrel@216.216.171.238> References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> <49DE3B85.6030706@gmail.com> <34640.216.216.171.238.1239306574.squirrel@216.216.171.238> Message-ID: <49DE4A33.6080409@gmail.com> Julius Szelagiewicz wrote: > >>> After that I test >>> the disk controller by formatting them... >> This may be a superstition going back to the old days of having tracking >> information laid down in the formatting step - and controllers with >> slight differences, but I always do a low-level format on scsi drives >> when I move them to a machine with a different controller type or if I >> don't know where they were before. It probably really isn't necessary >> any more but it does give the drive a chance to remap any questionable >> blocks transparently. With IDE and SATA, I'm not sure if this is even >> possible. >> >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see >> > Yes, it is possible. Spinrite from Gibson Research is an unbelievably good > product. I've been using it in its evolving incantations forever (since my > first IBM PC XT) and it works. It is especially good at fixing track > layout to mach the drift of the heads. Worth the money and then some. > julius But is that relevant any more? I thought current drives had permanent servo tracks embedded separately from the data and the heads followed them instead of having their own stepper concept. If you have to spend money, I'm inclined to just toss the old questionable stuff and replace it since things have gotten so much better in the last few years. -- Les Mikesell lesmikesell at gmail.com From julius at turtle.com Thu Apr 9 21:24:40 2009 From: julius at turtle.com (Julius Szelagiewicz) Date: Thu, 9 Apr 2009 17:24:40 -0400 (EDT) Subject: [K12OSN] RAID Question In-Reply-To: <49DE4A33.6080409@gmail.com> References: <994441ae0904061306x5b1cf389i135379df7e987e89@mail.gmail.com> <49DACCBB.7070703@cmosnetworks.com> <49DE3B85.6030706@gmail.com> <34640.216.216.171.238.1239306574.squirrel@216.216.171.238> <49DE4A33.6080409@gmail.com> Message-ID: <49100.216.216.171.238.1239312280.squirrel@216.216.171.238> > Julius Szelagiewicz wrote: >> >>>> After that I test >>>> the disk controller by formatting them... >>> This may be a superstition going back to the old days of having >>> tracking >>> information laid down in the formatting step - and controllers with >>> slight differences, but I always do a low-level format on scsi drives >>> when I move them to a machine with a different controller type or if I >>> don't know where they were before. It probably really isn't necessary >>> any more but it does give the drive a chance to remap any questionable >>> blocks transparently. With IDE and SATA, I'm not sure if this is even >>> possible. >>> >>> https://www.redhat.com/mailman/listinfo/k12osn >>> For more info see >>> >> Yes, it is possible. Spinrite from Gibson Research is an unbelievably >> good >> product. I've been using it in its evolving incantations forever (since >> my >> first IBM PC XT) and it works. It is especially good at fixing track >> layout to mach the drift of the heads. Worth the money and then some. >> julius > > But is that relevant any more? I thought current drives had permanent > servo tracks embedded separately from the data and the heads followed > them instead of having their own stepper concept. If you have to spend > money, I'm inclined to just toss the old questionable stuff and replace > it since things have gotten so much better in the last few years. > >> Les, the drives got much, much better, but they still fail. A move from controller to controller should be always painless, but sometimes things happen. I've used Spinrite most often for data recovery, but I also have a few old systems that I've just run it on and they got better and are still active. The beauty there is that the need to get the boxes unracked and opened is immediately obviated. I do configure NAS on those systems, but I never had any problems with the drives that Spinrite fixed and deemed OK. julius From thunsucker at cnc-usa.com Fri Apr 10 01:29:37 2009 From: thunsucker at cnc-usa.com (Trey Hunsucker) Date: Thu, 09 Apr 2009 20:29:37 -0500 Subject: [K12OSN] nbd errors Message-ID: <49DEA101.6060507@cnc-usa.com> hello, I kept getting nbd and ioctl errors until I convereted back to nfs, are there any other work arounds? Trey From francois.patte at mi.parisdescartes.fr Sat Apr 11 09:51:50 2009 From: francois.patte at mi.parisdescartes.fr (=?ISO-8859-1?Q?Fran=E7ois_Patte?=) Date: Sat, 11 Apr 2009 11:51:50 +0200 Subject: [K12OSN] time on thin client Message-ID: <49E06836.4080502@mi.parisdescartes.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Bonjour, I am unable to change the time on a TC. Either on console on TC or in the chroot on the server the date command returns: sam. avril 11 02:49:53 PDT 2009 I want CEST I changed the ZONE in /etc/sysconfig/clock (in the chroot) but TC is death! Thanks for helping. - -- Fran?ois Patte UFR de math?matiques et informatique Universit? Paris Descartes 45, rue des Saints P?res F-75270 Paris Cedex 06 T?l. +33 (0)1 4286 2145 http://www.math-info.univ-paris5.fr/~patte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkngaDYACgkQdE6C2dhV2JWGGACgluAWONnRbeUJqnSxA+EmAMBu uboAoKKkVrvwyF5Mh0GLXXqM20JaM01w =Arp0 -----END PGP SIGNATURE----- From francois.patte at mi.parisdescartes.fr Sat Apr 11 09:57:34 2009 From: francois.patte at mi.parisdescartes.fr (=?ISO-8859-1?Q?Fran=E7ois_Patte?=) Date: Sat, 11 Apr 2009 11:57:34 +0200 Subject: [K12OSN] language problems Message-ID: <49E0698E.7020203@mi.parisdescartes.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Bonjour, I have (at least) 2 language problems on my TC: 1- keyboard remains us on the console TC in spite of the config in lts.conf: CONSOLE_KEYMAP=fr 2- language of the greeter remains in english in spite of LANG=fr_FR.utf8 (in lts.conf) and Language=fr_FR.utf8 (in /etc/ltsp/ldm-global-dmrc) What to do? Thanks - -- Fran?ois Patte UFR de math?matiques et informatique Universit? Paris Descartes 45, rue des Saints P?res F-75270 Paris Cedex 06 T?l. +33 (0)1 4286 2145 http://www.math-info.univ-paris5.fr/~patte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkngaY4ACgkQdE6C2dhV2JU8OwCgt72HzmWUy2E+nEApkPYGKCCq sRkAoMI1SKKUejL68UMYQQG/VPPbRSQ3 =la2L -----END PGP SIGNATURE----- From francois.patte at mi.parisdescartes.fr Sat Apr 11 09:59:56 2009 From: francois.patte at mi.parisdescartes.fr (=?ISO-8859-1?Q?Fran=E7ois_Patte?=) Date: Sat, 11 Apr 2009 11:59:56 +0200 Subject: [K12OSN] log tranferred to the server Message-ID: <49E06A1C.8010405@mi.parisdescartes.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Bonjour, How to make the log of the TC available in /var/log/messages on the server? I have this in lts.conf file: SYSLOG_HOST=172.31.100.254 It doesn't work.... Thanks - -- Fran?ois Patte UFR de math?matiques et informatique Universit? Paris Descartes 45, rue des Saints P?res F-75270 Paris Cedex 06 T?l. +33 (0)1 4286 2145 http://www.math-info.univ-paris5.fr/~patte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkngahwACgkQdE6C2dhV2JX9dwCeN1YPB4SbTaj6wAHMmGgfnkgG jFYAoJ9+SL5DCVuUAad1hJkL09hSgVrB =Qwqk -----END PGP SIGNATURE----- From burke at thealmquists.net Sat Apr 11 18:05:26 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 11 Apr 2009 13:05:26 -0500 Subject: [K12OSN] time on thin client In-Reply-To: <49E06836.4080502@mi.parisdescartes.fr> References: <49E06836.4080502@mi.parisdescartes.fr> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Bonjour, > > I am unable to change the time on a TC. > > Either on console on TC or in the chroot on the server the date > command > returns: > > sam. avril 11 02:49:53 PDT 2009 > > I want CEST > > I changed the ZONE in /etc/sysconfig/clock (in the chroot) but TC > is death! > > Thanks for helping. Is the date the thin clients see from the server correct? > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAkng2+YACgkQxWV7OPa/g5EjogCeJE+R+QcZDntE2QAF2G8ibf2C gGwAnjeBYnKemJM1gHSfBS2yE+3NqMXA =nTcu -----END PGP SIGNATURE----- From burke at thealmquists.net Sat Apr 11 18:07:29 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 11 Apr 2009 13:07:29 -0500 Subject: [K12OSN] language problems In-Reply-To: <49E0698E.7020203@mi.parisdescartes.fr> References: <49E0698E.7020203@mi.parisdescartes.fr> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > Bonjour, > > I have (at least) 2 language problems on my TC: > > 1- keyboard remains us on the console TC in spite of the config in > lts.conf: > > CONSOLE_KEYMAP=fr > > 2- language of the greeter remains in english in spite of > > LANG=fr_FR.utf8 (in lts.conf) > > and > > Language=fr_FR.utf8 (in /etc/ltsp/ldm-global-dmrc) > > What to do? > > What distro are you using of LTSP: K12Linux (Fedora 10), Unbuntu, or K12LTSP (Centos)? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAkng3GEACgkQxWV7OPa/g5G+SACfWC7ABVQuGp2jBMIUrsD6Mhya BZ4An05pEcZ4foHh/Snc1eLsqyY7s5hs =Hpsl -----END PGP SIGNATURE----- From francois.patte at mi.parisdescartes.fr Sat Apr 11 18:52:47 2009 From: francois.patte at mi.parisdescartes.fr (=?ISO-8859-1?Q?Fran=E7ois_Patte?=) Date: Sat, 11 Apr 2009 20:52:47 +0200 Subject: [K12OSN] language problems In-Reply-To: References: <49E0698E.7020203@mi.parisdescartes.fr> Message-ID: <49E0E6FF.101@mi.parisdescartes.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Le 11/04/2009 20:07, Almquist Burke a ?crit : > >> Bonjour, > >> I have (at least) 2 language problems on my TC: > >> 1- keyboard remains us on the console TC in spite of the config in >> lts.conf: > >> CONSOLE_KEYMAP=fr > >> 2- language of the greeter remains in english in spite of > >> LANG=fr_FR.utf8 (in lts.conf) > >> and > >> Language=fr_FR.utf8 (in /etc/ltsp/ldm-global-dmrc) > >> What to do? > > > > What distro are you using of LTSP: K12Linux (Fedora 10), K12Linux (Fedora 10) - -- Fran?ois Patte UFR de math?matiques et informatique Universit? Paris Descartes 45, rue des Saints P?res F-75270 Paris Cedex 06 T?l. +33 (0)1 4286 2145 http://www.math-info.univ-paris5.fr/~patte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkng5v8ACgkQdE6C2dhV2JWoMQCghCROSorOZEfFe4Ty/cXtElgf Y6YAnRJjzFeB+kwmFG2NplcPmtNnDyIR =G30o -----END PGP SIGNATURE----- From francois.patte at mi.parisdescartes.fr Sat Apr 11 18:53:32 2009 From: francois.patte at mi.parisdescartes.fr (=?ISO-8859-1?Q?Fran=E7ois_Patte?=) Date: Sat, 11 Apr 2009 20:53:32 +0200 Subject: [K12OSN] time on thin client In-Reply-To: References: <49E06836.4080502@mi.parisdescartes.fr> Message-ID: <49E0E72C.3070708@mi.parisdescartes.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Le 11/04/2009 20:05, Almquist Burke a ?crit : >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 > >> Bonjour, > >> I am unable to change the time on a TC. > >> Either on console on TC or in the chroot on the server the date command >> returns: > >> sam. avril 11 02:49:53 PDT 2009 > >> I want CEST > >> I changed the ZONE in /etc/sysconfig/clock (in the chroot) but TC is >> death! > >> Thanks for helping. > > > Is the date the thin clients see from the server correct? yes. - -- Fran?ois Patte UFR de math?matiques et informatique Universit? Paris Descartes 45, rue des Saints P?res F-75270 Paris Cedex 06 T?l. +33 (0)1 4286 2145 http://www.math-info.univ-paris5.fr/~patte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkng5ywACgkQdE6C2dhV2JUSKQCgrUlndgLroZSYfqQtd+NDzCD+ 2wYAn177oYUXmPrHZPkv1amJWF2HrE4q =uJGZ -----END PGP SIGNATURE----- From burke at thealmquists.net Sat Apr 11 22:00:02 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 11 Apr 2009 17:00:02 -0500 Subject: [K12OSN] time on thin client In-Reply-To: <49E0E72C.3070708@mi.parisdescartes.fr> References: <49E06836.4080502@mi.parisdescartes.fr> <49E0E72C.3070708@mi.parisdescartes.fr> Message-ID: <05A67C7A-A05F-48F0-963F-6B2131624B6F@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm confused why are you are trying to set the time/date on the terminals. By default, isn't the chroot exported to the clients as read-only? When you log into a terminal, you are actually looking at a GUI provided from the server, including the time applet, unless I'm misunderstanding you. So the time your users are seeing should be coming from the server, unless you set up shell access on the thin client itself (like a local app, not an xterm). On Apr 11, 2009, at 1:53 PM, Fran?ois Patte wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Le 11/04/2009 20:05, Almquist Burke a ?crit : >>> -----BEGIN PGP SIGNED MESSAGE----- >>> Hash: SHA1 >> >>> Bonjour, >> >>> I am unable to change the time on a TC. >> >>> Either on console on TC or in the chroot on the server the date >>> command >>> returns: >> >>> sam. avril 11 02:49:53 PDT 2009 >> >>> I want CEST >> >>> I changed the ZONE in /etc/sysconfig/clock (in the chroot) but TC is >>> death! >> >>> Thanks for helping. >> >> >> Is the date the thin clients see from the server correct? > > yes. > > > - -- > Fran?ois Patte > UFR de math?matiques et informatique > Universit? Paris Descartes > 45, rue des Saints P?res > F-75270 Paris Cedex 06 > T?l. +33 (0)1 4286 2145 > http://www.math-info.univ-paris5.fr/~patte > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkng5ywACgkQdE6C2dhV2JUSKQCgrUlndgLroZSYfqQtd+NDzCD+ > 2wYAn177oYUXmPrHZPkv1amJWF2HrE4q > =uJGZ > -----END PGP SIGNATURE----- > > _______________________________________________ > 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) iEYEARECAAYFAknhEuIACgkQxWV7OPa/g5HF+wCcDvUksIN8DKq7uJe9knCersWW TuUAn3gc+WxGQB6gi5TP5umJI3rsylZN =Uy3K -----END PGP SIGNATURE----- From burke at thealmquists.net Sat Apr 11 22:00:59 2009 From: burke at thealmquists.net (Almquist Burke) Date: Sat, 11 Apr 2009 17:00:59 -0500 Subject: [K12OSN] language problems In-Reply-To: <49E0698E.7020203@mi.parisdescartes.fr> References: <49E0698E.7020203@mi.parisdescartes.fr> Message-ID: <41396F5D-8315-41BE-93B6-74821DE07B13@thealmquists.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 11, 2009, at 4:57 AM, Fran?ois Patte wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Bonjour, > > I have (at least) 2 language problems on my TC: > > 1- keyboard remains us on the console TC in spite of the config in > lts.conf: > > CONSOLE_KEYMAP=fr > > 2- language of the greeter remains in english in spite of > > LANG=fr_FR.utf8 (in lts.conf) > > and > > Language=fr_FR.utf8 (in /etc/ltsp/ldm-global-dmrc) > > What to do? If you are using K12Linux based on Fedora 10, this might apply to you. https://fedorahosted.org/k12linux/wiki/LanguageSetup -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAknhExsACgkQxWV7OPa/g5F8ywCfb/s00pe2N+5nmq133uT46igz ug0AnR03noT5pCDI+bhD8JftkB8dIyXX =opdC -----END PGP SIGNATURE----- From francois.patte at mi.parisdescartes.fr Sun Apr 12 07:05:19 2009 From: francois.patte at mi.parisdescartes.fr (=?ISO-8859-1?Q?Fran=E7ois_Patte?=) Date: Sun, 12 Apr 2009 09:05:19 +0200 Subject: [K12OSN] time on thin client In-Reply-To: <05A67C7A-A05F-48F0-963F-6B2131624B6F@thealmquists.net> References: <49E06836.4080502@mi.parisdescartes.fr> <49E0E72C.3070708@mi.parisdescartes.fr> <05A67C7A-A05F-48F0-963F-6B2131624B6F@thealmquists.net> Message-ID: <49E192AF.50004@mi.parisdescartes.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Le 12/04/2009 00:00, Almquist Burke a ?crit : > I'm confused why are you are trying to set the time/date on the > terminals. By default, isn't the chroot exported to the clients as > read-only? When you log into a terminal, you are actually looking at a > GUI provided from the server, including the time applet, unless I'm > misunderstanding you. So the time your users are seeing should be coming > from the server Yes when someone is logged on a TC but the time displayed on the greeter when nobody is logged on is pacific daylight time, which is important for people in California but we don't care too much here in Paris! (minus 9 hours!) I would like a greeter on time and possibly in French too (see my other posts) Thanks for helping. Regards. - -- Fran?ois Patte UFR de math?matiques et informatique Universit? Paris Descartes 45, rue des Saints P?res F-75270 Paris Cedex 06 T?l. +33 (0)1 4286 2145 http://www.math-info.univ-paris5.fr/~patte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAknhkq8ACgkQdE6C2dhV2JUj5gCgtZDldFxSQSsoOZP9re9fge7a sPQAoMVkDng1zU9iE8h//FnfbTUKAu9b =2FP3 -----END PGP SIGNATURE----- From francois.patte at mi.parisdescartes.fr Sun Apr 12 08:17:14 2009 From: francois.patte at mi.parisdescartes.fr (=?ISO-8859-1?Q?Fran=E7ois_Patte?=) Date: Sun, 12 Apr 2009 10:17:14 +0200 Subject: [K12OSN] language problems [HALF SOLVED] In-Reply-To: <41396F5D-8315-41BE-93B6-74821DE07B13@thealmquists.net> References: <49E0698E.7020203@mi.parisdescartes.fr> <41396F5D-8315-41BE-93B6-74821DE07B13@thealmquists.net> Message-ID: <49E1A38A.90705@mi.parisdescartes.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Le 12/04/2009 00:00, Almquist Burke a ?crit : > > On Apr 11, 2009, at 4:57 AM, Fran?ois Patte wrote: > > If you are using K12Linux based on Fedora 10, this might apply to you. > https://fedorahosted.org/k12linux/wiki/LanguageSetup Thanks for the link. It is suggested to set LANG=fr_FR.UTF_8 in /var/lib/tftpboot/ltsp/i386/lts.conf That was already done. The second suggestion is: yum groupinstall "Japanese Support" answer: "nothing to do". I found how to solve greeter language: put LANG="fr_FR.UTF-8" in /etc/sysconfig/i18n in the chroot Remains: how to have an azerty keyboard in console on TC. Best regards. - -- Fran?ois Patte UFR de math?matiques et informatique Universit? Paris Descartes 45, rue des Saints P?res F-75270 Paris Cedex 06 T?l. +33 (0)1 4286 2145 http://www.math-info.univ-paris5.fr/~patte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAknho4oACgkQdE6C2dhV2JUGNgCdH+yplJmvDReStgWU0RmT6GgH G9sAni9t0rJvblGs/0L3C09qdACiGxE4 =+MlU -----END PGP SIGNATURE----- From proyecto.edulin at gmail.com Tue Apr 14 22:30:56 2009 From: proyecto.edulin at gmail.com (Alberto Castillo) Date: Tue, 14 Apr 2009 19:30:56 -0300 Subject: [K12OSN] Help VNC crash w/firefox Message-ID: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> Hi everybody! I write from Argentina. I recently build a micro-distro that runs with i486 w/4MB ram (yes... just four megas). This micro-distro have a Xclient from a vnc-client (svncviewer). All apps run nice and with a good performance, but when I try to navigate with firefox, the vnc connection closes and the Xclient crashes and freeze!!!... Here a only that I can get of /var/log/messages: (I login w/user 'educa1') ***************************************************************** Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 from=192.168.0.100 Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n 2.14.0), pid 8890 usuario ?educa1? Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de configuraci?n de s?lo lectura en la posici?n 0 Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n escribible en la posici?n 1 Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de configuraci?n de s?lo lectura en la posici?n 2 Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, /org/bluez/applet) registered Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader E-Gate 0 0 Not Found Apr 14 09:24:25 server last message repeated 4 times Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n escribible en la posici?n 0 Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 pid=8789 duration=108(sec) Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? en uso, cerr?ndolo. Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando ******************************************************************** As you can see, xinetd suddenly closes the vnc connection. When I put more ram the crash dissapear (4MB more, total=8MB), but I really need to use this machines (near of 1500) of 4MB ram, the machines are a IBM valuepoint-425SX model 6381. A few days ago, I can not neither open firefox (the free mem that bios tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). Next, I disable the ROM in RAM and I get near of 320Kb, and then I can navigate only in some pages, that have a few text nothing more... I did read that is a bug of Xorg, and I did update it, but nothing.... I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try to get some repos (especially non-free) Any can help me? or give me any idea? Thankyou a lot! Alberto Castillo From rowens at ptd.net Wed Apr 15 01:34:57 2009 From: rowens at ptd.net (Rob Owens) Date: Tue, 14 Apr 2009 21:34:57 -0400 Subject: [K12OSN] Help VNC crash w/firefox In-Reply-To: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> References: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> Message-ID: <20090415013457.GB32661@aurora.owens.net> I recall a lot of discussion about 6 months (or maybe a year) ago about Firefox causing thin clients to crash when many tabs were open. I think it boiled down to images that Firefox loads get cached by the X server, and once the X server is out of RAM it crashes. You have only 4 MB for your X server, so it is crashing instantly (images on your home page use up all 4 MB, perhaps). Have you tried other browsers? I'm not sure if it's a Firefox thing or if all browsers will behave the same. -Rob On Tue, Apr 14, 2009 at 07:30:56PM -0300, Alberto Castillo wrote: > Hi everybody! > I write from Argentina. > > I recently build a micro-distro that runs with i486 w/4MB ram (yes... > just four megas). > This micro-distro have a Xclient from a vnc-client (svncviewer). > All apps run nice and with a good performance, but when I try to > navigate with firefox, the vnc connection closes and the Xclient > crashes and freeze!!!... > > Here a only that I can get of /var/log/messages: (I login w/user 'educa1') > > ***************************************************************** > Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 > from=192.168.0.100 > Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n > 2.14.0), pid 8890 usuario ?educa1? > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de > configuraci?n de s?lo lectura en la posici?n 0 > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n > escribible en la posici?n 1 > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de > configuraci?n de s?lo lectura en la posici?n 2 > Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, > /org/bluez/applet) registered > Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader > E-Gate 0 0 Not Found > Apr 14 09:24:25 server last message repeated 4 times > Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n > escribible en la posici?n 0 > Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 > pid=8789 duration=108(sec) > Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? > en uso, cerr?ndolo. > Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando > ******************************************************************** > > As you can see, xinetd suddenly closes the vnc connection. > When I put more ram the crash dissapear (4MB more, total=8MB), but I > really need to use this machines (near of 1500) of 4MB ram, the > machines are a IBM valuepoint-425SX model 6381. > A few days ago, I can not neither open firefox (the free mem that bios > tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). > Next, I disable the ROM in RAM and I get near of 320Kb, and then I can > navigate only in some pages, that have a few text nothing more... > I did read that is a bug of Xorg, and I did update it, but nothing.... > I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try > to get some repos (especially non-free) > Any can help me? or give me any idea? > Thankyou a lot! > Alberto Castillo > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From microman at cmosnetworks.com Wed Apr 15 01:49:09 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Tue, 14 Apr 2009 21:49:09 -0400 Subject: [K12OSN] Help VNC crash w/firefox In-Reply-To: <20090415013457.GB32661@aurora.owens.net> References: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> <20090415013457.GB32661@aurora.owens.net> Message-ID: <49E53D15.3000000@cmosnetworks.com> As I recall, the X11 issue happens regardless of which browser you use, be it Firefox, Konqueror, Epiphany, CucamongaBrowser, etc. But he's using VNC, so this might not necessarily apply anywhere near to the same degree. My guess is that VNC is probably much like the early Citrix RDP clients that could run in relatively low DRAM. Those PS/ValuePoint 6381's use 72-pin SIMMs. What I would suggest Alberto do, if he wants to use those ValuePoints, is either of the following: 1.) Pick up some 72-pin SIMM memory, enough to upgrade to at least 8MB (more is better, of course). Or.... 2.) Cannibalize DRAM from some of those 1,500 and make 750 of them with 8MB DRAM, since he's discovered that 8MB works. Either way, it looks like he's going to have to upgrade the DRAM on those boxes. --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! Rob Owens wrote: > I recall a lot of discussion about 6 months (or maybe a year) ago about > Firefox causing thin clients to crash when many tabs were open. I think it > boiled down to images that Firefox loads get cached by the X server, and > once the X server is out of RAM it crashes. You have only 4 MB for your X > server, so it is crashing instantly (images on your home page use up all 4 > MB, perhaps). > > Have you tried other browsers? I'm not sure if it's a Firefox thing or if > all browsers will behave the same. > > -Rob > > On Tue, Apr 14, 2009 at 07:30:56PM -0300, Alberto Castillo wrote: > >> Hi everybody! >> I write from Argentina. >> >> I recently build a micro-distro that runs with i486 w/4MB ram (yes... >> just four megas). >> This micro-distro have a Xclient from a vnc-client (svncviewer). >> All apps run nice and with a good performance, but when I try to >> navigate with firefox, the vnc connection closes and the Xclient >> crashes and freeze!!!... >> >> Here a only that I can get of /var/log/messages: (I login w/user 'educa1') >> >> ***************************************************************** >> Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 >> from=192.168.0.100 >> Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n >> 2.14.0), pid 8890 usuario ?educa1? >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de >> configuraci?n de s?lo lectura en la posici?n 0 >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n >> escribible en la posici?n 1 >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de >> configuraci?n de s?lo lectura en la posici?n 2 >> Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, >> /org/bluez/applet) registered >> Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader >> E-Gate 0 0 Not Found >> Apr 14 09:24:25 server last message repeated 4 times >> Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n >> escribible en la posici?n 0 >> Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 >> pid=8789 duration=108(sec) >> Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? >> en uso, cerr?ndolo. >> Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando >> ******************************************************************** >> >> As you can see, xinetd suddenly closes the vnc connection. >> When I put more ram the crash dissapear (4MB more, total=8MB), but I >> really need to use this machines (near of 1500) of 4MB ram, the >> machines are a IBM valuepoint-425SX model 6381. >> A few days ago, I can not neither open firefox (the free mem that bios >> tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). >> Next, I disable the ROM in RAM and I get near of 320Kb, and then I can >> navigate only in some pages, that have a few text nothing more... >> I did read that is a bug of Xorg, and I did update it, but nothing.... >> I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try >> to get some repos (especially non-free) >> Any can help me? or give me any idea? >> Thankyou a lot! >> Alberto Castillo >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see >> > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From simpsond at leopards.k12.ar.us Wed Apr 15 13:12:12 2009 From: simpsond at leopards.k12.ar.us (Doug Simpson) Date: Wed, 15 Apr 2009 08:12:12 -0500 Subject: [K12OSN] Help VNC crash w/firefox In-Reply-To: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> References: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> Message-ID: <49E596DC.550C.0078.0@leopards.k12.ar.us> Would it be possible to obtain a copy of this distro? Does it boot from CD or run from hard drive? I have a use for this. Let me know and thank you! Doug Doug Simpson Technology Specialist De Queen Public Schools De Queen, AR simpsond at leopards.k12.ar.us "A Dollar Saved is a Dollar Earned" >>> Alberto Castillo 4/14/2009 5:30 PM >>> Hi everybody! I write from Argentina. I recently build a micro-distro that runs with i486 w/4MB ram (yes... just four megas). This micro-distro have a Xclient from a vnc-client (svncviewer). All apps run nice and with a good performance, but when I try to navigate with firefox, the vnc connection closes and the Xclient crashes and freeze!!!... Here a only that I can get of /var/log/messages: (I login w/user 'educa1') ***************************************************************** Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 from=192.168.0.100 Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n 2.14.0), pid 8890 usuario ?educa1? Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de configuraci?n de s?lo lectura en la posici?n 0 Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n escribible en la posici?n 1 Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de configuraci?n de s?lo lectura en la posici?n 2 Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, /org/bluez/applet) registered Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader E-Gate 0 0 Not Found Apr 14 09:24:25 server last message repeated 4 times Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n escribible en la posici?n 0 Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 pid=8789 duration=108(sec) Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? en uso, cerr?ndolo. Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando ******************************************************************** As you can see, xinetd suddenly closes the vnc connection. When I put more ram the crash dissapear (4MB more, total=8MB), but I really need to use this machines (near of 1500) of 4MB ram, the machines are a IBM valuepoint-425SX model 6381. A few days ago, I can not neither open firefox (the free mem that bios tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). Next, I disable the ROM in RAM and I get near of 320Kb, and then I can navigate only in some pages, that have a few text nothing more... I did read that is a bug of Xorg, and I did update it, but nothing.... I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try to get some repos (especially non-free) Any can help me? or give me any idea? Thankyou a lot! Alberto Castillo _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see From proyecto.edulin at gmail.com Wed Apr 15 16:51:51 2009 From: proyecto.edulin at gmail.com (Alberto Castillo) Date: Wed, 15 Apr 2009 13:51:51 -0300 Subject: [K12OSN] Help VNC crash w/firefox In-Reply-To: <49E53D15.3000000@cmosnetworks.com> References: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> <20090415013457.GB32661@aurora.owens.net> <49E53D15.3000000@cmosnetworks.com> Message-ID: <12b235150904150951p27d82e1ag4d95b85375651570@mail.gmail.com> 2009/4/14 "Terrell Prud? Jr." : > As I recall, the X11 issue happens regardless of which browser you use, be > it Firefox, Konqueror, Epiphany, CucamongaBrowser, etc. > > But he's using VNC, so this might not necessarily apply anywhere near to the > same degree.? My guess is that VNC is probably much like the early Citrix > RDP clients that could run in relatively low DRAM. > > Those PS/ValuePoint 6381's use 72-pin SIMMs.? What I would suggest Alberto > do, if he wants to use those ValuePoints, is either of the following: > > 1.)? Pick up some 72-pin SIMM memory, enough to upgrade to at least 8MB > (more is better, of course).? Or.... > Last night I can stabilize the client and no more freezes... (for now...) I just put more inodes in the roofs image (1000 inodes) and I eliminate my mini-busybox experiment and I recompile with NOTHING but ash (sh), echo, date, ifconfig, insmod, uname and another link to some binary tool... [in fact, if you try to use this micro distro outside of GUI client (graphic), its a trash... do not offer nothing, but is the only way that I can know to run 4MB ram machines...] Netx I rebuild all rootfs image adjusting the new size, and I did enter to sites like WIKIPEDIA and ADOBE.com! without freeze!!! Just only the render of web pages is slow, but is the any problem! because the opneoffice apps for example works nice! > 2.)? Cannibalize DRAM from some of those 1,500 and make 750 of them with 8MB > DRAM, since he's discovered that 8MB works. In the past here a people build a minidistro that runs w/8mb, based in the hint of Csaba Henk of LFS. Next, I based this microdist in the same hint, but I reduce everything I can and modif some aspects, nothing more. > > Either way, it looks like he's going to have to upgrade the DRAM on those > boxes. > In some time, this machines run to the creep... but for the moment we can use for kinder garden and primary schools... > --TP > _______________________________ > Do you GNU? > Microsoft Free since 2003--the ultimate antivirus protection! > > > Rob Owens wrote: > > I recall a lot of discussion about 6 months (or maybe a year) ago about > Firefox causing thin clients to crash when many tabs were open. I think it > boiled down to images that Firefox loads get cached by the X server, and > once the X server is out of RAM it crashes. You have only 4 MB for your X > server, so it is crashing instantly (images on your home page use up all 4 > MB, perhaps). > > Have you tried other browsers? I'm not sure if it's a Firefox thing or if > all browsers will behave the same. > > -Rob > > On Tue, Apr 14, 2009 at 07:30:56PM -0300, Alberto Castillo wrote: > > > Hi everybody! > I write from Argentina. > > I recently build a micro-distro that runs with i486 w/4MB ram (yes... > just four megas). > This micro-distro have a Xclient from a vnc-client (svncviewer). > All apps run nice and with a good performance, but when I try to > navigate with firefox, the vnc connection closes and the Xclient > crashes and freeze!!!... > > Here a only that I can get of /var/log/messages: (I login w/user 'educa1') > > ***************************************************************** > Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 > from=192.168.0.100 > Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n > 2.14.0), pid 8890 usuario ?educa1? > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de > configuraci?n de s?lo lectura en la posici?n 0 > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n > escribible en la posici?n 1 > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de > configuraci?n de s?lo lectura en la posici?n 2 > Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, > /org/bluez/applet) registered > Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader > E-Gate 0 0 Not Found > Apr 14 09:24:25 server last message repeated 4 times > Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n > escribible en la posici?n 0 > Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 > pid=8789 duration=108(sec) > Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? > en uso, cerr?ndolo. > Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando > ******************************************************************** > > As you can see, xinetd suddenly closes the vnc connection. > When I put more ram the crash dissapear (4MB more, total=8MB), but I > really need to use this machines (near of 1500) of 4MB ram, the > machines are a IBM valuepoint-425SX model 6381. > A few days ago, I can not neither open firefox (the free mem that bios > tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). > Next, I disable the ROM in RAM and I get near of 320Kb, and then I can > navigate only in some pages, that have a few text nothing more... > I did read that is a bug of Xorg, and I did update it, but nothing.... > I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try > to get some repos (especially non-free) > Any can help me? or give me any idea? > Thankyou a lot! > Alberto Castillo > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From proyecto.edulin at gmail.com Wed Apr 15 16:54:08 2009 From: proyecto.edulin at gmail.com (Alberto Castillo) Date: Wed, 15 Apr 2009 13:54:08 -0300 Subject: [K12OSN] Help VNC crash w/firefox In-Reply-To: <20090415013457.GB32661@aurora.owens.net> References: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> <20090415013457.GB32661@aurora.owens.net> Message-ID: <12b235150904150954r420927d2kdd36f0a2afb818a1@mail.gmail.com> 2009/4/14 Rob Owens : > I recall a lot of discussion about 6 months (or maybe a year) ago about > Firefox causing thin clients to crash when many tabs were open. ?I think it > boiled down to images that Firefox loads get cached by the X server, and > once the X server is out of RAM it crashes. ?You have only 4 MB for your X > server, so it is crashing instantly (images on your home page use up all 4 > MB, perhaps). > > Have you tried other browsers? ?I'm not sure if it's a Firefox thing or if > all browsers will behave the same. I try with some browsers and the result is the same: firefox dillo opera and others that I can not remember > > -Rob > > On Tue, Apr 14, 2009 at 07:30:56PM -0300, Alberto Castillo wrote: >> Hi everybody! >> I write from Argentina. >> >> I recently build a micro-distro that runs with i486 w/4MB ram (yes... >> just four megas). >> This micro-distro have a Xclient from a vnc-client (svncviewer). >> All apps run nice and with a good performance, but when I try to >> navigate with firefox, the vnc connection closes and the Xclient >> crashes and freeze!!!... >> >> Here a only that I can get of /var/log/messages: (I login w/user 'educa1') >> >> ***************************************************************** >> Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 >> from=192.168.0.100 >> Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n >> 2.14.0), pid 8890 usuario ?educa1? >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de >> configuraci?n de s?lo lectura en la posici?n 0 >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n >> escribible en la posici?n 1 >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de >> configuraci?n de s?lo lectura en la posici?n 2 >> Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, >> /org/bluez/applet) registered >> Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader >> E-Gate 0 0 Not Found >> Apr 14 09:24:25 server last message repeated 4 times >> Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n >> escribible en la posici?n 0 >> Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 >> pid=8789 duration=108(sec) >> Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? >> en uso, cerr?ndolo. >> Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando >> ******************************************************************** >> >> As you can see, xinetd suddenly closes the vnc connection. >> When I put more ram the crash dissapear (4MB more, total=8MB), but I >> really need to use this machines (near of 1500) of 4MB ram, the >> machines are a IBM valuepoint-425SX model 6381. >> A few days ago, I can not neither open firefox (the free mem that bios >> tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). >> Next, I disable the ROM in RAM and I get near of 320Kb, and then I can >> navigate only in some pages, that have a few text nothing more... >> I did read that is a bug of Xorg, and I did update it, but nothing.... >> I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try >> to get some repos (especially non-free) >> Any can help me? or give me any idea? >> Thankyou a lot! >> Alberto Castillo >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From proyecto.edulin at gmail.com Wed Apr 15 16:54:28 2009 From: proyecto.edulin at gmail.com (Alberto Castillo) Date: Wed, 15 Apr 2009 13:54:28 -0300 Subject: [K12OSN] Help VNC crash w/firefox In-Reply-To: <20090415013457.GB32661@aurora.owens.net> References: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> <20090415013457.GB32661@aurora.owens.net> Message-ID: <12b235150904150954m756845c1u646cbf51db0f8c6a@mail.gmail.com> 2009/4/14 Rob Owens : > I recall a lot of discussion about 6 months (or maybe a year) ago about > Firefox causing thin clients to crash when many tabs were open. ?I think it > boiled down to images that Firefox loads get cached by the X server, and > once the X server is out of RAM it crashes. ?You have only 4 MB for your X > server, so it is crashing instantly (images on your home page use up all 4 > MB, perhaps). > > Have you tried other browsers? ?I'm not sure if it's a Firefox thing or if > all browsers will behave the same. I try with some browsers and the result is the same: firefox dillo opera and others that I can not remember > > -Rob > > On Tue, Apr 14, 2009 at 07:30:56PM -0300, Alberto Castillo wrote: >> Hi everybody! >> I write from Argentina. >> >> I recently build a micro-distro that runs with i486 w/4MB ram (yes... >> just four megas). >> This micro-distro have a Xclient from a vnc-client (svncviewer). >> All apps run nice and with a good performance, but when I try to >> navigate with firefox, the vnc connection closes and the Xclient >> crashes and freeze!!!... >> >> Here a only that I can get of /var/log/messages: (I login w/user 'educa1') >> >> ***************************************************************** >> Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 >> from=192.168.0.100 >> Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n >> 2.14.0), pid 8890 usuario ?educa1? >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de >> configuraci?n de s?lo lectura en la posici?n 0 >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n >> escribible en la posici?n 1 >> Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de >> configuraci?n de s?lo lectura en la posici?n 2 >> Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, >> /org/bluez/applet) registered >> Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader >> E-Gate 0 0 Not Found >> Apr 14 09:24:25 server last message repeated 4 times >> Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n >> ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n >> escribible en la posici?n 0 >> Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 >> pid=8789 duration=108(sec) >> Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? >> en uso, cerr?ndolo. >> Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando >> ******************************************************************** >> >> As you can see, xinetd suddenly closes the vnc connection. >> When I put more ram the crash dissapear (4MB more, total=8MB), but I >> really need to use this machines (near of 1500) of 4MB ram, the >> machines are a IBM valuepoint-425SX model 6381. >> A few days ago, I can not neither open firefox (the free mem that bios >> tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). >> Next, I disable the ROM in RAM and I get near of 320Kb, and then I can >> navigate only in some pages, that have a few text nothing more... >> I did read that is a bug of Xorg, and I did update it, but nothing.... >> I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try >> to get some repos (especially non-free) >> Any can help me? or give me any idea? >> Thankyou a lot! >> Alberto Castillo >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From proyecto.edulin at gmail.com Wed Apr 15 17:01:30 2009 From: proyecto.edulin at gmail.com (Alberto Castillo) Date: Wed, 15 Apr 2009 14:01:30 -0300 Subject: [K12OSN] Help VNC crash w/firefox In-Reply-To: <49E596DC.550C.0078.0@leopards.k12.ar.us> References: <12b235150904141530g37dfc4a8xac252b198f05915b@mail.gmail.com> <49E596DC.550C.0078.0@leopards.k12.ar.us> Message-ID: <12b235150904151001u3d8e004et8311335697d99d03@mail.gmail.com> Hi Doug! 2009/4/15 Doug Simpson : > Would it be possible to obtain a copy of this distro? Does it boot from > CD or run from hard drive? > well, in this moment, this micro distro run in a diskett, but I will try to setup to run from a tftp server (like ltsp). That the space is very important, I can not use a dhcp client in this distro to set up the network interface, so, I define in static way on the fly with ifconfig, then we need create one distro for each client, is no problem! just a bit slow in the implementation, but is just only one time... Sorry for my bad english > I have a use for this. Great! I just finish and send you 10 personalized copies for 10 clients! and if I this these days run via tftp, I tell you how make it! > > Let me know and thank you! Thanks to you!!! > > Doug > > Doug Simpson > Technology Specialist > De Queen Public Schools > De Queen, AR > simpsond at leopards.k12.ar.us > "A Dollar Saved is a Dollar Earned" > > >>>> Alberto Castillo 4/14/2009 5:30 PM >>> > Hi everybody! > I write from Argentina. > > I recently build a micro-distro that runs with i486 w/4MB ram (yes... > just four megas). > This micro-distro have a Xclient from a vnc-client (svncviewer). > All apps run nice and with a good performance, but when I try to > navigate with firefox, the vnc connection closes and the Xclient > crashes and freeze!!!... > > Here a only that I can get of /var/log/messages: (I login w/user > 'educa1') > > ***************************************************************** > Apr 14 09:23:58 server xinetd[7823]: START: vnc-800x600x16 pid=8789 > from=192.168.0.100 > Apr 14 09:24:23 server gconfd (educa1-8890): comenzando (versi?n > 2.14.0), pid 8890 usuario ?educa1? > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readonly:/etc/gconf/gconf.xml.mandatory? a una fuente de > configuraci?n de s?lo lectura en la posici?n 0 > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n > escribible en la posici?n 1 > Apr 14 09:24:23 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readonly:/etc/gconf/gconf.xml.defaults? a una fuente de > configuraci?n de s?lo lectura en la posici?n 2 > Apr 14 09:24:25 server hcid[2127]: Default passkey agent (:1.525, > /org/bluez/applet) registered > Apr 14 09:24:25 server pcscd: winscard.c:304:SCardConnect() Reader > E-Gate 0 0 Not Found > Apr 14 09:24:25 server last message repeated 4 times > Apr 14 09:24:25 server gconfd (educa1-8890): Se resolvi? la direcci?n > ?xml:readwrite:/home/educa1/.gconf? a una fuente de configuraci?n > escribible en la posici?n 0 > Apr 14 09:25:46 server xinetd[7823]: EXIT: vnc-800x600x16 status=0 > pid=8789 duration=108(sec) > Apr 14 09:25:53 server gconfd (educa1-8890): El servidor GConf no est? > en uso, cerr?ndolo. > Apr 14 09:25:53 server gconfd (educa1-8890): Finalizando > ******************************************************************** > > As you can see, xinetd suddenly closes the vnc connection. > When I put more ram the crash dissapear (4MB more, total=8MB), but I > really need to use this machines (near of 1500) of 4MB ram, the > machines are a IBM valuepoint-425SX model 6381. > A few days ago, I can not neither open firefox (the free mem that bios > tell me are 3072Kb, kernel 509Kb-bzipped, rootfs-290kb-gz). > Next, I disable the ROM in RAM and I get near of 320Kb, and then I can > navigate only in some pages, that have a few text nothing more... > I did read that is a bug of Xorg, and I did update it, but nothing.... > I have been thinking come back to fedora 5 -4 -3 -2 -1 or RH9, and try > to get some repos (especially non-free) > Any can help me? or give me any idea? > Thankyou a lot! > Alberto Castillo > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From timlegge at gmail.com Thu Apr 16 01:27:19 2009 From: timlegge at gmail.com (Timothy Legge) Date: Wed, 15 Apr 2009 22:27:19 -0300 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: <49D5720B.2020201@bittware.com> References: <49D5720B.2020201@bittware.com> Message-ID: On Thu, Apr 2, 2009 at 11:18 PM, j.w. thomas wrote: > I recently set up a box and installed IPCop on it to serve as our firewall. > It apparently runs a caching DNS server, but for the life of me, I cannot > figure out where it gets ITS dns info. And as you might guess, I'm having > some pretty severe DNS troubles. You need to logon to the console as admin. DNS is set there. It may be possible to setup dns via the web interface but it seems to me that when I started using OpenDNS with ipcop I had to use the admin console. You can enable ssh on the web page if you need to do it remotely. Tim From thewhitmers at gmail.com Thu Apr 16 12:20:05 2009 From: thewhitmers at gmail.com (David Whitmer) Date: Thu, 16 Apr 2009 08:20:05 -0400 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: References: <49D5720B.2020201@bittware.com> Message-ID: On Wed, Apr 15, 2009 at 9:27 PM, Timothy Legge wrote: > On Thu, Apr 2, 2009 at 11:18 PM, j.w. thomas wrote: > > I recently set up a box and installed IPCop on it to serve as our > firewall. > > It apparently runs a caching DNS server, but for the life of me, I > cannot > > figure out where it gets ITS dns info. And as you might guess, I'm > having > > some pretty severe DNS troubles. > > You need to logon to the console as admin. DNS is set there. It may > be possible to setup dns via the web interface but it seems to me that > when I started using OpenDNS with ipcop I had to use the admin > console. You can enable ssh on the web page if you need to do it > remotely. > > Tim > I second that process. A while ago when I configured our school's IPCop to use OpenDNS, I could only do that though the console. David -------------- next part -------------- An HTML attachment was scrubbed... URL: From jthomas at bittware.com Thu Apr 16 12:37:19 2009 From: jthomas at bittware.com (j.w. thomas) Date: Thu, 16 Apr 2009 08:37:19 -0400 Subject: [K12OSN] IPCop DNS Issues In-Reply-To: References: <49D5720B.2020201@bittware.com> Message-ID: <49E7267F.4080101@bittware.com> David Whitmer wrote: > > > On Wed, Apr 15, 2009 at 9:27 PM, Timothy Legge > wrote: > > On Thu, Apr 2, 2009 at 11:18 PM, j.w. thomas > wrote: > > I recently set up a box and installed IPCop on it to serve as our > firewall. > > It apparently runs a caching DNS server, but for the life of me, > I cannot > > figure out where it gets ITS dns info. And as you might guess, > I'm having > > some pretty severe DNS troubles. > > You need to logon to the console as admin. DNS is set there. It may > be possible to setup dns via the web interface but it seems to me that > when I started using OpenDNS with ipcop I had to use the admin > console. You can enable ssh on the web page if you need to do it > remotely. > > Tim > > > > I second that process. > > A while ago when I configured our school's IPCop to use OpenDNS, I could > only do that though the console. > > David Thanks to everyone who chimed in on this one. I switched us over to OpenDNS, and everything works like a charm now. -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Sometimes experience is the only teacher that works - Mike Rosing From proyecto.edulin at gmail.com Thu Apr 16 13:03:30 2009 From: proyecto.edulin at gmail.com (Alberto Castillo) Date: Thu, 16 Apr 2009 10:03:30 -0300 Subject: [K12OSN] Help VNC crash w/firefox [SOLVED] Message-ID: <12b235150904160603p16efd0f2qeb62989c34c81d91@mail.gmail.com> Hi everybody! I just solved my problem, I just recompile busybox and make more room from ramdisk, I disable rom speculars in Bios and now every is fine! I can navegate in sites like wikipedia and sites with high contents of splashs screens of flash, like ADOBE. Cheers! Works fine in IBM valuepoint 425SX and EPSON Action3000 (CyRyx486DX-40Mhz) both w/4mb RAM From Marc.Fromm at wwu.edu Thu Apr 16 16:26:09 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Thu, 16 Apr 2009 09:26:09 -0700 Subject: [K12OSN] boot failure Message-ID: Nothing new has changed on the server, but the client machines are not booting. The screen reports information like My ip address seems to be 8CA0D8E5 ###.###.###.### The system then goes through a series of: Trying to load pxelinux.cfg/8CA0D8E5 Trying to load pxelinux.cfg/8CA0D8E . . . Trying to load pxelinux.cfg/8 Trying to load pxelinux.cfg/default The system then hangs. Marc -------------- next part -------------- An HTML attachment was scrubbed... URL: From chan at sacredsf.org Thu Apr 16 17:12:33 2009 From: chan at sacredsf.org (Hoover Chan) Date: Thu, 16 Apr 2009 10:12:33 -0700 (PDT) Subject: [K12OSN] directory services in a linux centric cross platform workstation environment Message-ID: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> The subject line is quite a mouthful but that's pretty much the question I'm mulling over right now. I've been bombarded by proponents of Active Directory for Windows and Open Directory for the Macintosh with all the different neat network applications that become easy to implement and manage if only I switch the servers to the appropriate platform. Does anyone here have a mostly Macintosh and Windows workstation environment that integrates well with an OpenLDAP type Linux centric server? If so, how hard is it to build? A lot of my colleagues at different schools locally seem to come from using AD as their core directory service and then "bend" everything to adapt to that. I'd hate to be forced into building a Windows server just for AD... Thanks in advance. -------------------------------------------------- Hoover Chan chan at sacredsf.org Technology Director Schools of the Sacred Heart 2222 Broadway St. San Francisco, CA 94115 From Marc.Fromm at wwu.edu Thu Apr 16 18:45:56 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Thu, 16 Apr 2009 11:45:56 -0700 Subject: [K12OSN] boot failure update Message-ID: I posted the following problem Nothing new has changed on the server, but the client machines are not booting. The screen reports information like My ip address seems to be 8CA0D8E5 ###.###.###.### The system then goes through a series of: Trying to load pxelinux.cfg/8CA0D8E5 Trying to load pxelinux.cfg/8CA0D8E . . . Trying to load pxelinux.cfg/8 Trying to load pxelinux.cfg/default The system then hangs. Update: Only 2 of the 5 ltsp clients are having the above problem. Also, after hanging for a while at default a message is finally posted stating "could not find kernel image : linux" All five systems are configured the same on the ltsp server. Marc -------------- next part -------------- An HTML attachment was scrubbed... URL: From microman at cmosnetworks.com Thu Apr 16 19:30:47 2009 From: microman at cmosnetworks.com (=?UTF-8?B?IlRlcnJlbGwgUHJ1ZMOpIEpyLiI=?=) Date: Thu, 16 Apr 2009 15:30:47 -0400 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> References: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> Message-ID: <49E78767.2010402@cmosnetworks.com> You might want to look at Fedora/RedHat Directory Server. I just took Red Hat's class in it last month and have been practicing with it. It is pretty cool and will indeed authenticate MS Windows boxes. Even has a GUI interface! :-) Macintoshes, I understand, "just work" in an LDAP environment. Yes, I'd stay away from the MS Craptive Directory if I were you. You're thinking right-on here. --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! Hoover Chan wrote: > The subject line is quite a mouthful but that's pretty much the question I'm mulling over right now. > > I've been bombarded by proponents of Active Directory for Windows and Open Directory for the Macintosh with all the different neat network applications that become easy to implement and manage if only I switch the servers to the appropriate platform. Does anyone here have a mostly Macintosh and Windows workstation environment that integrates well with an OpenLDAP type Linux centric server? If so, how hard is it to build? > > A lot of my colleagues at different schools locally seem to come from using AD as their core directory service and then "bend" everything to adapt to that. I'd hate to be forced into building a Windows server just for AD... > > Thanks in advance. > > > -------------------------------------------------- > Hoover Chan chan at sacredsf.org > Technology Director > Schools of the Sacred Heart > 2222 Broadway St. > San Francisco, CA 94115 > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bmead at lane.k12.or.us Thu Apr 16 19:34:44 2009 From: bmead at lane.k12.or.us (Bob Mead) Date: Thu, 16 Apr 2009 12:34:44 -0700 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> References: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> Message-ID: <49E78854.1070601@lane.k12.or.us> We use openLDAP as our main authentication tool. Our school is mostly win-doze desktops with some macs, yet our server infrastructure is mainly linux. We have only two win servers (required for a few specific apps). All of our clients (some 1300+ computers district-wide) authenticate through the LDAP server. I wasn't here for the build out, but I can say that it works. For our win clients we have samba to serve win-style shares and we use roaming profiles to allow transparency on the network. The problem we have is that our users tend to generate huge .profile directories and this slows down the login process to a crawl [or worse]. Once we fix that - it works seamlessly. As a side note, even our ltsp terminals and servers use the ldap server to authenticate. Upon re-reading your post below, are you trying to build an ltsp server that is also your main openLDAP authentication server (in lieu of an AD or other auth. server)? If this is so, then I would guess that your installation is going to be somewhat complex. HTH ~bob Hoover Chan wrote: > The subject line is quite a mouthful but that's pretty much the question I'm mulling over right now. > > I've been bombarded by proponents of Active Directory for Windows and Open Directory for the Macintosh with all the different neat network applications that become easy to implement and manage if only I switch the servers to the appropriate platform. Does anyone here have a mostly Macintosh and Windows workstation environment that integrates well with an OpenLDAP type Linux centric server? If so, how hard is it to build? > > A lot of my colleagues at different schools locally seem to come from using AD as their core directory service and then "bend" everything to adapt to that. I'd hate to be forced into building a Windows server just for AD... > > Thanks in advance. > > > -------------------------------------------------- > Hoover Chan chan at sacredsf.org > Technology Director > Schools of the Sacred Heart > 2222 Broadway St. > San Francisco, CA 94115 > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > -------------- next part -------------- A non-text attachment was scrubbed... Name: bmead.vcf Type: text/x-vcard Size: 209 bytes Desc: not available URL: From chan at sacredsf.org Thu Apr 16 19:42:21 2009 From: chan at sacredsf.org (Hoover Chan) Date: Thu, 16 Apr 2009 12:42:21 -0700 (PDT) Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <5417166.813621239910658951.JavaMail.root@zimbra.sacredsf.org> Message-ID: <871859.813731239910941915.JavaMail.root@zimbra.sacredsf.org> re: huge .profile directories slowing things to a crawl I experienced that but ran out of time to figure it out to a satisfactory conclusion. A pointer to how to solve it? re: planning to use ltsp as the core of the new directory service No, I'm not planning to do that but thought that the collected audience here would be knowledgeable and sympathetic toward a FOSS solution to this rather than resorting to buying a Windows Active Directory or Apple Macintosh Open Directory server. Most of the workstations and laptops here on the academic side of the house are Macintoshes. Easily about 85% of our campus are Macs. The remaining are Windows for the administration (we use something called Blackbaud for administrative computing which is Windows specific). I have one Linux workstation... Thanks. -------------------------------------------------- Hoover Chan chan at sacredsf.org Technology Director Schools of the Sacred Heart 2222 Broadway St. San Francisco, CA 94115 ----- "Bob Mead" wrote: > We use openLDAP as our main authentication tool. Our school is mostly > > win-doze desktops with some macs, yet our server infrastructure is > mainly linux. We have only two win servers (required for a few > specific > apps). All of our clients (some 1300+ computers district-wide) > authenticate through the LDAP server. I wasn't here for the build out, > > but I can say that it works. For our win clients we have samba to > serve > win-style shares and we use roaming profiles to allow transparency on > > the network. The problem we have is that our users tend to generate > huge > .profile directories and this slows down the login process to a crawl > > [or worse]. Once we fix that - it works seamlessly. As a side note, > even > our ltsp terminals and servers use the ldap server to authenticate. > > Upon re-reading your post below, are you trying to build an ltsp > server > that is also your main openLDAP authentication server (in lieu of an > AD > or other auth. server)? If this is so, then I would guess that your > installation is going to be somewhat complex. > > HTH > ~bob > > Hoover Chan wrote: > > The subject line is quite a mouthful but that's pretty much the > question I'm mulling over right now. > > > > I've been bombarded by proponents of Active Directory for Windows > and Open Directory for the Macintosh with all the different neat > network applications that become easy to implement and manage if only > I switch the servers to the appropriate platform. Does anyone here > have a mostly Macintosh and Windows workstation environment that > integrates well with an OpenLDAP type Linux centric server? If so, how > hard is it to build? > > > > A lot of my colleagues at different schools locally seem to come > from using AD as their core directory service and then "bend" > everything to adapt to that. I'd hate to be forced into building a > Windows server just for AD... > > > > Thanks in advance. > > > > > > -------------------------------------------------- > > Hoover Chan chan at sacredsf.org > > Technology Director > > Schools of the Sacred Heart > > 2222 Broadway St. > > San Francisco, CA 94115 > > > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see > > > > From bmead at lane.k12.or.us Thu Apr 16 21:01:07 2009 From: bmead at lane.k12.or.us (Bob Mead) Date: Thu, 16 Apr 2009 14:01:07 -0700 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <871859.813731239910941915.JavaMail.root@zimbra.sacredsf.org> References: <871859.813731239910941915.JavaMail.root@zimbra.sacredsf.org> Message-ID: <49E79C93.7060204@lane.k12.or.us> An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: bmead.vcf Type: text/x-vcard Size: 209 bytes Desc: not available URL: From bmead at lane.k12.or.us Thu Apr 16 22:52:26 2009 From: bmead at lane.k12.or.us (Bob Mead) Date: Thu, 16 Apr 2009 15:52:26 -0700 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <49E79C93.7060204@lane.k12.or.us> References: <871859.813731239910941915.JavaMail.root@zimbra.sacredsf.org> <49E79C93.7060204@lane.k12.or.us> Message-ID: <49E7B6AA.5000405@lane.k12.or.us> An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: bmead.vcf Type: text/x-vcard Size: 209 bytes Desc: not available URL: From missive at hotmail.com Thu Apr 16 23:57:23 2009 From: missive at hotmail.com (Lee Harr) Date: Fri, 17 Apr 2009 04:27:23 +0430 Subject: [K12OSN] Re: does recordmydesktop work on thin clients? Message-ID: > I'm looking for a way for students to record things > on their desktop sessions, but I'm not sure what is > the best way to go. Any suggestions? I have a feeling I could use some kind of vnc recorder, but that seems like it would be pretty complex. recordmydesktop looks nice, but I'm not sure it will work on a thin. Anyone have experience with this? Here is my specs.txt: recordMyDesktop = 0.3.8.1 Width = 1024 Height = 768 Filename = /home/lee/out.ogv FPS = 15.000000 NoSound = 1 Frequency = 22050 Channels = 1 BufferSize = 4096 SoundFrameSize = -1081074260 PeriodSize = 134598644 UsedJack = 0 v_bitrate = 45000 v_quality = 63 s_quality = 10 ZeroCompression = 1 _________________________________________________________________ Drag n? drop?Get easy photo sharing with Windows Live? Photos. http://www.microsoft.com/windows/windowslive/products/photos.aspx From mel at melwade.com Fri Apr 17 02:46:30 2009 From: mel at melwade.com (Mel Wade) Date: Thu, 16 Apr 2009 19:46:30 -0700 Subject: [K12OSN] Firefox Proxy Settings Message-ID: <43080f460904161946h57d293b5w753677efa12c79b8@mail.gmail.com> K12LTSP 5EL I've set the proxy settings in the /usr/lib64/xulrunner-1.9/greprefs/all.js file. This has set the proxy setting for the client. Users can still change the settings. If a user does change them, how can I change them back or does that happen automatically in a new session? -- Mel Wade "A common mistake people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools." - Douglas Adams http://www.melwade.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From robark at gmail.com Fri Apr 17 04:23:05 2009 From: robark at gmail.com (Robert Arkiletian) Date: Thu, 16 Apr 2009 21:23:05 -0700 Subject: [K12OSN] Re: does recordmydesktop work on thin clients? In-Reply-To: References: Message-ID: 2009/4/16 Lee Harr : > >> I'm looking for a way for students to record things >> on their desktop sessions, but I'm not sure what is >> the best way to go. > > Any suggestions? > I don't think it will work. Even if you manage to somehow get the video working what about the microphone? There is no mic support in ltsp. I don't think recordmydesktop or any other similar software was developed with ltsp in mind. > > I have a feeling I could use some kind of vnc recorder, but > that seems like it would be pretty complex. > > > recordmydesktop looks nice, but I'm not sure it will work > on a thin. Anyone have experience with this? > > Here is my specs.txt: > > recordMyDesktop = 0.3.8.1 > Width = 1024 > Height = 768 > Filename = /home/lee/out.ogv > FPS = 15.000000 > NoSound = 1 > Frequency = 22050 > Channels = 1 > BufferSize = 4096 > SoundFrameSize = -1081074260 > PeriodSize = 134598644 > UsedJack = 0 > v_bitrate = 45000 > v_quality = 63 > s_quality = 10 > ZeroCompression = 1 > > > _________________________________________________________________ > Drag n? drop?Get easy photo sharing with Windows Live? Photos. > > http://www.microsoft.com/windows/windowslive/products/photos.aspx > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From rowens at ptd.net Fri Apr 17 12:48:34 2009 From: rowens at ptd.net (Rob Owens) Date: Fri, 17 Apr 2009 08:48:34 -0400 Subject: [K12OSN] boot failure update In-Reply-To: References: Message-ID: <20090417124834.GA12113@aurora.owens.net> Bad switch or wiring maybe? If nothing changed on the server (no updates or anything), then I'd have to suspect hardware. -Rob On Thu, Apr 16, 2009 at 11:45:56AM -0700, Marc Fromm wrote: > I posted the following problem > > Nothing new has changed on the server, but the client machines are not booting. > The screen reports information like > > My ip address seems to be 8CA0D8E5 ###.###.###.### > The system then goes through a series of: > Trying to load pxelinux.cfg/8CA0D8E5 > Trying to load pxelinux.cfg/8CA0D8E > . . . > Trying to load pxelinux.cfg/8 > Trying to load pxelinux.cfg/default > > The system then hangs. > > Update: > Only 2 of the 5 ltsp clients are having the above problem. > Also, after hanging for a while at default a message is finally posted stating "could not find kernel image : linux" > > All five systems are configured the same on the ltsp server. > > Marc > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From micha at arava.co.il Fri Apr 17 14:31:43 2009 From: micha at arava.co.il (Micha Silver) Date: Fri, 17 Apr 2009 17:31:43 +0300 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <49E7B6AA.5000405@lane.k12.or.us> References: <871859.813731239910941915.JavaMail.root@zimbra.sacredsf.org> <49E79C93.7060204@lane.k12.or.us> <49E7B6AA.5000405@lane.k12.or.us> Message-ID: <49E892CF.9010507@arava.co.il> Hello Bob: Bob Mead wrote: > > > Bob Mead wrote: >> >> >> Hoover Chan wrote: >>> re: huge .profile directories slowing things to a crawl >>> >>> I experienced that but ran out of time to figure it out to a satisfactory conclusion. A pointer to how to solve it? >>> >>> >> > Sorry, lets try this again - here! > >> I've read your post, and I'm still confused. If you have a minute... Once you've moved both the user's .profile directory and the .SHSD directory, doesn't she lose everything in her profile at the next login? How would windows (or the samba server) "know" that all the user's stuff is now in a new location? Thanks, Micha From pxeboot at gmail.com Fri Apr 17 17:20:07 2009 From: pxeboot at gmail.com (Conrad Lawes) Date: Fri, 17 Apr 2009 13:20:07 -0400 Subject: [K12OSN] boot failure update In-Reply-To: <20090417124834.GA12113@aurora.owens.net> References: <20090417124834.GA12113@aurora.owens.net> Message-ID: Are all your thin clients running on identical hardware? If not, could be an issue with the PXE boot ROM. I have seen similar behavior with the SIS 9000 boot ROM. On Fri, Apr 17, 2009 at 8:48 AM, Rob Owens wrote: > Bad switch or wiring maybe? If nothing changed on the server (no updates > or anything), then I'd have to suspect hardware. > > -Rob > > On Thu, Apr 16, 2009 at 11:45:56AM -0700, Marc Fromm wrote: > > I posted the following problem > > > > Nothing new has changed on the server, but the client machines are not > booting. > > The screen reports information like > > > > My ip address seems to be 8CA0D8E5 ###.###.###.### > > The system then goes through a series of: > > Trying to load pxelinux.cfg/8CA0D8E5 > > Trying to load pxelinux.cfg/8CA0D8E > > . . . > > Trying to load pxelinux.cfg/8 > > Trying to load pxelinux.cfg/default > > > > The system then hangs. > > > > Update: > > Only 2 of the 5 ltsp clients are having the above problem. > > Also, after hanging for a while at default a message is finally posted > stating "could not find kernel image : linux" > > > > All five systems are configured the same on the ltsp server. > > > > Marc > > > > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- Regards, Conrad Lawes -------------- next part -------------- An HTML attachment was scrubbed... URL: From pxeboot at gmail.com Fri Apr 17 18:20:43 2009 From: pxeboot at gmail.com (Conrad Lawes) Date: Fri, 17 Apr 2009 14:20:43 -0400 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <49E79C93.7060204@lane.k12.or.us> References: <871859.813731239910941915.JavaMail.root@zimbra.sacredsf.org> <49E79C93.7060204@lane.k12.or.us> Message-ID: I'm not usually one to stick up for Microsoft, but when it comes to various implementation of LDAP, Microsoft has implemented the most user-friendly and intuitive directory service on the market today. I can't comment on Macintosh Open Directory; however, I am fairly knowledgeable about OpenLDAP since I have deployed and administered it for 3 years before migrating to Active Directory. It took me over 2 weeks to fully configure OpenLDAP for my needs. I really wish the Linux community could develop tools to make OpenLDAP easier to deploy and manage. Novell's Suse NDS is very robust and stable but is very difficult to deploy. Xandros Linuxis the only distro I found that shields the newbie system admin for all the grunt work involved in OpenLDAP. I have been waiting for Samba 4 since 2005 and still no sign of it. The smbldap-installer tools that make it easy to deploy OpenLDAP is now an orphan. There many commercial products that allow seamless integration between Linux and Mac clients and Active Directory. If cost is a factor, then check out LikeWise Open . It's free and supports Linux and Mac clients. IMHO, AD is now the defacto standard for directory services. On Thu, Apr 16, 2009 at 5:01 PM, Bob Mead wrote: > > > Hoover Chan wrote: > > re: huge .profile directories slowing things to a crawl > > I experienced that but ran out of time to figure it out to a satisfactory conclusion. A pointer to how to solve it? > > > > I wrote a blog post on how to fix these here. > > I echo TP's sentiment, that you are on the right foot to stay away from AD. > I am going to check out the Fedora/RedHat solution as our ldap server is > aging and could probably use an update. For now, we are just using a > standard openLDAP implementation with samba on linux servers of various > flavors. > > re: planning to use ltsp as the core of the new directory service > > No, I'm not planning to do that but thought that the collected audience here would be knowledgeable and sympathetic toward a FOSS solution to this rather than resorting to buying a Windows Active Directory or Apple Macintosh Open Directory server. > > Most of the workstations and laptops here on the academic side of the house are Macintoshes. Easily about 85% of our campus are Macs. The remaining are Windows for the administration (we use something called Blackbaud for administrative computing which is Windows specific). I have one Linux workstation... > > Thanks. > > -------------------------------------------------- > Hoover Chan chan at sacredsf.org > Technology Director > Schools of the Sacred Heart > 2222 Broadway St. > San Francisco, CA 94115 > > > ----- "Bob Mead" wrote: > > > > We use openLDAP as our main authentication tool. Our school is mostly > > win-doze desktops with some macs, yet our server infrastructure is > mainly linux. We have only two win servers (required for a few > specific > apps). All of our clients (some 1300+ computers district-wide) > authenticate through the LDAP server. I wasn't here for the build out, > > but I can say that it works. For our win clients we have samba to > serve > win-style shares and we use roaming profiles to allow transparency on > > the network. The problem we have is that our users tend to generate > huge > .profile directories and this slows down the login process to a crawl > > [or worse]. Once we fix that - it works seamlessly. As a side note, > even > our ltsp terminals and servers use the ldap server to authenticate. > > Upon re-reading your post below, are you trying to build an ltsp > server > that is also your main openLDAP authentication server (in lieu of an > AD > or other auth. server)? If this is so, then I would guess that your > installation is going to be somewhat complex. > > HTH > ~bob > > Hoover Chan wrote: > > > The subject line is quite a mouthful but that's pretty much the > > > question I'm mulling over right now. > > > I've been bombarded by proponents of Active Directory for Windows > > > and Open Directory for the Macintosh with all the different neat > network applications that become easy to implement and manage if only > I switch the servers to the appropriate platform. Does anyone here > have a mostly Macintosh and Windows workstation environment that > integrates well with an OpenLDAP type Linux centric server? If so, how > hard is it to build? > > > A lot of my colleagues at different schools locally seem to come > > > from using AD as their core directory service and then "bend" > everything to adapt to that. I'd hate to be forced into building a > Windows server just for AD... > > > Thanks in advance. > > > -------------------------------------------------- > Hoover Chan chan at sacredsf.org > Technology Director > Schools of the Sacred Heart > 2222 Broadway St. > San Francisco, CA 94115 > > _______________________________________________ > K12OSN mailing listK12OSN at redhat.comhttps://www.redhat.com/mailman/listinfo/k12osn > For more info see > > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- Regards, Conrad Lawes -------------- next part -------------- An HTML attachment was scrubbed... URL: From DLWillson at TheGeek.NU Fri Apr 17 20:39:32 2009 From: DLWillson at TheGeek.NU (David L. Willson) Date: Fri, 17 Apr 2009 14:39:32 -0600 (MDT) Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: Message-ID: <14935804.5001240000772484.JavaMail.root@zimbra.thegeek.nu> Have you tried Fedora DS or Red Hat DS? That other guy spoke pretty well of them, and I'm pretty proud of the pure Linux back end at one of my client sites. I'd hate to give it up if there's a Free LDAP that doesn't suck. I'll agree with you that Microsoft AD generally doesn't suck, but I really see some importance in the "Free-ness". David L. Willson Trainer, Engineer, Enthusiast MCT, MCSE, Linux+ tel://720.333.LANS Freeing people from the tyranny (or whatevery) of Microsofty-ness, one at a time. ----- "Conrad Lawes" wrote: > I'm not usually one to stick up for Microsoft, but ... From cgrossko at wusd.org Fri Apr 17 21:22:40 2009 From: cgrossko at wusd.org (Cody Grosskopf) Date: Fri, 17 Apr 2009 14:22:40 -0700 Subject: [K12OSN] directory services in a Linux centric cross platformworkstation environment In-Reply-To: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> References: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> Message-ID: <49E890B0020000BC00011C90@webmail.wusd.org> eDirectory coupled with OES2 and Domain Services for Windows will handle just about all of that. The best part is you don't have to install Windows Server. >>> Hoover Chan 4/16/2009 10:12 AM >>> The subject line is quite a mouthful but that's pretty much the question I'm mulling over right now. I've been bombarded by proponents of Active Directory for Windows and Open Directory for the Macintosh with all the different neat network applications that become easy to implement and manage if only I switch the servers to the appropriate platform. Does anyone here have a mostly Macintosh and Windows workstation environment that integrates well with an OpenLDAP type Linux centric server? If so, how hard is it to build? A lot of my colleagues at different schools locally seem to come from using AD as their core directory service and then "bend" everything to adapt to that. I'd hate to be forced into building a Windows server just for AD... Thanks in advance. -------------------------------------------------- Hoover Chan chan at sacredsf.org Technology Director Schools of the Sacred Heart 2222 Broadway St. San Francisco, CA 94115 _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see -------------- next part -------------- An HTML attachment was scrubbed... URL: From Marc.Fromm at wwu.edu Fri Apr 17 22:41:20 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Fri, 17 Apr 2009 15:41:20 -0700 Subject: [K12OSN] boot failure update In-Reply-To: References: <20090417124834.GA12113@aurora.owens.net> Message-ID: The clients are running on identical hardware. I think what is going on is our building recently became DHCP and fro some reason these two ltsp clients are picking up the campus DHCP broadcast and it is interfering. ________________________________ From: k12osn-bounces at redhat.com [mailto:k12osn-bounces at redhat.com] On Behalf Of Conrad Lawes Sent: Friday, April 17, 2009 10:20 AM To: Support list for open source software in schools. Subject: Re: [K12OSN] boot failure update Are all your thin clients running on identical hardware? If not, could be an issue with the PXE boot ROM. I have seen similar behavior with the SIS 9000 boot ROM. On Fri, Apr 17, 2009 at 8:48 AM, Rob Owens > wrote: Bad switch or wiring maybe? If nothing changed on the server (no updates or anything), then I'd have to suspect hardware. -Rob On Thu, Apr 16, 2009 at 11:45:56AM -0700, Marc Fromm wrote: > I posted the following problem > > Nothing new has changed on the server, but the client machines are not booting. > The screen reports information like > > My ip address seems to be 8CA0D8E5 ###.###.###.### > The system then goes through a series of: > Trying to load pxelinux.cfg/8CA0D8E5 > Trying to load pxelinux.cfg/8CA0D8E > . . . > Trying to load pxelinux.cfg/8 > Trying to load pxelinux.cfg/default > > The system then hangs. > > Update: > Only 2 of the 5 ltsp clients are having the above problem. > Also, after hanging for a while at default a message is finally posted stating "could not find kernel image : linux" > > All five systems are configured the same on the ltsp server. > > Marc > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see -- Regards, Conrad Lawes -------------- next part -------------- An HTML attachment was scrubbed... URL: From rowens at ptd.net Fri Apr 17 23:00:18 2009 From: rowens at ptd.net (Rob Owens) Date: Fri, 17 Apr 2009 19:00:18 -0400 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> References: <31527707.810221239901953827.JavaMail.root@zimbra.sacredsf.org> Message-ID: <20090417230018.GA15444@aurora.owens.net> I've used openldap/samba in the past, but not in a production environment. I used the smbldap-installer script on Debian Etch, and have used both LAM and Webmin to manage it. It took a lot of learning to set it up (which I am glad I did), but it's pretty easy to let a newbie Linux admin manage things with either of those two GUIs. I never did much with it besides creating users and groups, though, so I can't comment on the ease of any other LDAP features. -Rob On Thu, Apr 16, 2009 at 10:12:33AM -0700, Hoover Chan wrote: > The subject line is quite a mouthful but that's pretty much the question I'm mulling over right now. > > I've been bombarded by proponents of Active Directory for Windows and Open Directory for the Macintosh with all the different neat network applications that become easy to implement and manage if only I switch the servers to the appropriate platform. Does anyone here have a mostly Macintosh and Windows workstation environment that integrates well with an OpenLDAP type Linux centric server? If so, how hard is it to build? > > A lot of my colleagues at different schools locally seem to come from using AD as their core directory service and then "bend" everything to adapt to that. I'd hate to be forced into building a Windows server just for AD... > > Thanks in advance. > > > -------------------------------------------------- > Hoover Chan chan at sacredsf.org > Technology Director > Schools of the Sacred Heart > 2222 Broadway St. > San Francisco, CA 94115 > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From microman at cmosnetworks.com Sat Apr 18 02:46:02 2009 From: microman at cmosnetworks.com (=?UTF-8?B?IlRlcnJlbGwgUHJ1ZMOpIEpyLiI=?=) Date: Fri, 17 Apr 2009 22:46:02 -0400 Subject: [K12OSN] directory services in a linux centric cross platform workstation environment In-Reply-To: <14935804.5001240000772484.JavaMail.root@zimbra.thegeek.nu> References: <14935804.5001240000772484.JavaMail.root@zimbra.thegeek.nu> Message-ID: <49E93EEA.9040406@cmosnetworks.com> Here's a tip from "that other guy." :-) The GUI administration tool is a Java app. RH is working on making it work with GNU Classpath, so hopefully by RHEL 6 (meaning CentOS 6 as well), the entire stack will be Free. But it nonetheless does work. One other thing: I took the class, and that gave me a good jump-start regarding how LDAP works. But be forewarned that there is a learning curve with pretty much any UNIX-based LDAP out there. That said, it's worth it. Definitely. --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! David L. Willson wrote: > Have you tried Fedora DS or Red Hat DS? That other guy spoke pretty well of them, and I'm pretty proud of the pure Linux back end at one of my client sites. I'd hate to give it up if there's a Free LDAP that doesn't suck. I'll agree with you that Microsoft AD generally doesn't suck, but I really see some importance in the "Free-ness". > > David L. Willson > Trainer, Engineer, Enthusiast > MCT, MCSE, Linux+ > tel://720.333.LANS > Freeing people from the tyranny (or whatevery) of Microsofty-ness, one at a time. > > ----- "Conrad Lawes" wrote: > > >> I'm not usually one to stick up for Microsoft, but ... >> > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mel at melwade.com Sun Apr 19 18:14:54 2009 From: mel at melwade.com (Mel Wade) Date: Sun, 19 Apr 2009 11:14:54 -0700 Subject: [K12OSN] FireFox not responding Message-ID: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> We continually get users who have messages saying that FireFox is already running and won't let them login. If I delete the .mozilla folder in their Home directory everything is fine. How can I keep this from happening in the future? -- Mel Wade "A common mistake people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools." - Douglas Adams http://www.melwade.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From brcisna at eazylivin.net Sun Apr 19 19:11:03 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Sun, 19 Apr 2009 14:11:03 -0500 Subject: [K12OSN] Re: does recordmydesktop work on thin clients? Message-ID: <1240168263.26410.6.camel@localhost.localdomain> Hi Lee, I never could get recordmydesktop to work on a TC. vnc2swf will work for recording TC desktop sessions, albeit without sound. Here is a link to vnc2swf.(Only the C version will work,not the pyvnc2swf). http://www.unixuser.org/~euske/vnc2swf/ It would be easy for students to use,once they see the drill on the keyboard shortcuts.Just as it sounds the the recorded session is saved in .swf format. Comes out very clear and useable. Take Care, Barry Cisna From brcisna at eazylivin.net Sun Apr 19 19:17:28 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Sun, 19 Apr 2009 14:17:28 -0500 Subject: [K12OSN] Firefox Proxy Settings Message-ID: <1240168648.26410.12.camel@localhost.localdomain> Mel, the next login session the Firefox proxy settings will be in effect again. The downside if the little angels figured out how to remove them the previous time,they will remove this session as well,other than it takes a little leg work on their part. Here is a link that keeps your locked in settings 'greyed out' AKA: tamper proof.. http://www.stbernard.com/ip5kb/iPrism/WhatsNew/50SupportFiles/SupportFiles/IP0460.htm Hope this helps. Barry Cisna From brcisna at eazylivin.net Sun Apr 19 19:49:28 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Sun, 19 Apr 2009 14:49:28 -0500 Subject: [K12OSN] Re: FireFox not responding Message-ID: <1240170568.26410.23.camel@localhost.localdomain> Mel, I never could figure out how to make this delima more automated. This may make things a tad bit easier for you. As root on the server you can run; ps aux|grep firefox-bin this will of course return the firefox running processes. If say user jimsmith is having this message you simply look for jimsmith and write down the PID#. Say his process number is 12345. Next just do a; kill -9 12345 Now jimsmith can log onto the server and use firefox with no hiccups. This will at least get you away from having to drill through home folders deleting the .mozilla/firefox dot folder. There should be some way of automating this with a shell script but I am not smart enough to figure it out. the only way I could figure it out was to kill ALL firefox processes at a user login which would result in all kinds of teachers !@#$%^^,you know the drill. Actually at school the only time we ever get this is if over the weekend we have a power outage(which happens quite regular at one of the older schools) and someone was left logged in. So this is very rare we end up with a hanged firefox session for a user. Hope this helps some. Barry Cisna From missive at hotmail.com Sun Apr 19 20:30:11 2009 From: missive at hotmail.com (Lee Harr) Date: Mon, 20 Apr 2009 01:00:11 +0430 Subject: [K12OSN] Re: FireFox not responding Message-ID: > We continually get users who have messages saying that FireFox is already > running and won't let them login. Do you mean that it will not let them log in to the terminal at all? Or will it just not let them start firefox? > If I delete the .mozilla folder in their > Home directory everything is fine. How can I keep this from happening in > the future? We have a problem with firefox not starting quite regularly. I am pretty sure that the problem is flash crashing and leaving firefox in an odd state. If that is the problem, there's no need to delete .mozilla When it happens, I train students to use the process monitor to kill the runaway firefox. If they cannot even log in, you have a more serious problem. _________________________________________________________________ Show them the way! Add maps and directions to your party invites. http://www.microsoft.com/windows/windowslive/products/events.aspx From abo at lunger.org Sun Apr 19 20:41:32 2009 From: abo at lunger.org (Karl Lunger) Date: Sun, 19 Apr 2009 22:41:32 +0200 Subject: [K12OSN] FireFox not responding In-Reply-To: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> References: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> Message-ID: <49EB8C7C.5040806@lunger.org> Hi There are 2 things: 1. The processes in the RAM that can remove with kill -9 PID The PIDs you can see with ps aux | grep firefox Whene this not was the problem, then is a lock file in the firefox directory .mozilla/firefox/ with the name "lock". The user can remuve it and firefox can restart. find ~/.mozilla/firefox/ -name lock | xargs rm Happy Hacking! ~Karl Mel Wade schrieb: > We continually get users who have messages saying that FireFox is > already running and won't let them login. If I delete the .mozilla > folder in their Home directory everything is fine. How can I keep > this from happening in the future? > > -- > Mel Wade > "A common mistake people make when trying to design something > completely foolproof is to underestimate the ingenuity of complete > fools." - Douglas Adams > http://www.melwade.com > > ------------------------------------------------------------------------ > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From mel at melwade.com Sun Apr 19 20:52:46 2009 From: mel at melwade.com (Mel Wade) Date: Sun, 19 Apr 2009 13:52:46 -0700 Subject: [K12OSN] FireFox not responding In-Reply-To: <49EB8C7C.5040806@lunger.org> References: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> <49EB8C7C.5040806@lunger.org> Message-ID: <43080f460904191352n2cffbef6q63491a650300aa48@mail.gmail.com> I find it interesting that I have not had this problem before until I reinstalled K12LTSP on my server. I have three servers. The problem only happens on two of them. -- Mel Wade "A common mistake people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools." - Douglas Adams http://www.melwade.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From brcisna at eazylivin.net Tue Apr 21 01:46:45 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Mon, 20 Apr 2009 20:46:45 -0500 Subject: [K12OSN] boot failure update Message-ID: <1240278405.25494.7.camel@localhost.localdomain> Hi Marc, I would guess you are right in regards to having two dhcp servers running on the same subnet. An easy way to determine this is plug a laptop into the switch feeding your ltsp TC's. If it is a winders laptop do an 'ipconfig /release + ipconfig /renew' 5 times. If it is a linux laptop just do an 'service network restart' and then ifconfig 5 times. See what is showing as the dhcp server/gateway etc. Also if you leave the cat5 wires in the TC's, switch the other end of the non-booting TC's into a different port in the switch,and see if they boot now. They may or may not boot after switching the port on the switch. Take Care, Barry Cisna From dahopkins429 at gmail.com Tue Apr 21 17:23:48 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 21 Apr 2009 13:23:48 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP Message-ID: Let me first say that this is going to sound like a rant in places. Not much I can do about that, but ... FC10+LTSP5 has not been performing well at all. I am currently at a loss to explain why. However, since I have to have sound working, LTSP5 seems to be the only way to ensure that sound works correctly. I have CentOS+LTSP4.2 and that works well for everything except sound. So, the only option I see is to get a distribution that is using LTSP5 working. Again, just to be clear, I am using identical hardware for the comparison and using the same login accounts, same file server, same dns, same authentication server, etc. All hardware is 32bit, both server and clients. (I don't even want to deal/worry with the 64bit server/32bit client possible issues at the moment). Now, here is what I and the elementary school tech teacher observed today. The following is her write-up. "Things did not go so well this morning. When all 10 computers were in use at the same time, the delay between mouse and screen was significant. . . The point of the lesson was to improve mouse skills--not possible when there is a lag between their mouse movements and action on their monitors. We muddled through the first group of 10 students, and when the 2nd group began the exercise, I allowed the first 10 students to open Tux Paint. I thought because Tux Paint is running local, this would work. Big Mistake! The delay for everyone increased dramatically, making it virtually impossible to complete the mouse task in Starfall. When I tried to "QUIT" foxfire on 3 computers, it took 3-4 minutes to return to the desktop. Although I was circulating the room, trying to assist students, I glanced at the load several times--I never saw it rise above 6. It mostly hovered between 4 and 5. It took more than 5 minutes to successfully close the website from 10 computers. During that time, I had 10 students just waiting. When my second class arrived, I did not even try to use the website. We used Tux Paint today. However, shortly after we got started, I "banned" students from selecting a new piece of paper . . . The few who had tried feature had their monitor hung-up for more than a minute. That task used to respond immediately. There is also a terrific feature that allows students to select any color from the rainbow . . . but choosing that feature takes more than 1 full minute to accomplish." This is on a system where with CentOS+LTSP4.2 I could run 25 systems simultaneously without issues. She was trying to use 10. Notice that the load average never exceeded 6. This is dual hyperthreaded Xeon so a load average of 4 would mean 100% utilization although that is a bit misleading as load averages of 6-8 perform quite well on all my other systems. Also, the system was never using swap. In fact, memory usage never exceeded 5GB. So, where is the bottleneck? The starfall activity is flash-based (it was the Earth Day activity). I know that FF3+flash is going to load the system. But, This issue is not as severe with FF2+Flash 9 except that you don't get sound half the time. FF3+Flash10 seems to really slow down. Also, it seems that network traffic is significantly higher with FC10+LTSP5 using ldm than with gdm. Can I switch back to gdm as the default manager or is ldm it? I have the LDM_DIRECTX set to TRUE so that ssh is only used for login/logout. And, login/logout now takes 30+ secs compared to about 2 seconds for CentOS+LTSP4.2. For the local apps, launching FF3 can take over a minute. And then it will be sluggish, even when the local hardware isn't using swap either. I have this suspicion that it is a network bandwidth issue. The only difference there is that LTSP5 uses the ltsbr0 bridge setup while LTSP4.2 does not. To test this, I should be able to delete the bridge and set up LTSP5 in the same dual NIC scenario as with LTSP4.2, correct? Though I am not sure I have the skills to do so without breaking something else. It might be as easy as deleting the ltsbr0 entry and then defining the IP address for the currently-slaved NIC to be what the ltsbr0 was defined as. I haven't had a chance to look at the stats from the switch (Amer.com, SS2R24G4i ) but since I never changed the switch, only the OS, I don't see why there would suddenly be an issue. As for the Tuxpaint issue. That is truly baffling. I have the same version of Tuxpaint running on an older server and it is very responsive. There is a hardware difference for the server ... the one that runs very well has CPU's with only 70% the speed of the newer server. The other difference is again CentOS+LTSP4.2 (using gdm) vs FC10+LTSP5 (and ldm). So, something looks like it 'just isn't right' except I'm not getting any disk I/O errors, I'm not getting a huge spike in the load ... the system just isn't responsive. At this point the teacher has really reached her limit as have I. A single login with a single client works fine. Add a few more and I get the above. I want LTSP5 to work but I can't stay with it given the current performance issues. And I have to start planning now for next fall. If upgrading to FC10+LTSP5 means all my current hardware is not acceptable, then I have a huge issue. I know that all my current hardware works with FC10+LTSP5, but the performance I'm seeing is horrible. I have been advocating/using K12LTSP since 2003, I really want this to work, but right now to say I am depressed with FC10+LTSP5 would be an understatement. So ... help? I'll be back at the school tonight to try and determine what might be happening. And once there, sitting behind the state firewalls, access to IRC is blocked as is all other chat capabilities. Sincerely, Dave Hopkins Newark Charter School From scott at hosef.org Tue Apr 21 20:29:34 2009 From: scott at hosef.org (R. Scott Belford) Date: Tue, 21 Apr 2009 10:29:34 -1000 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: Message-ID: Deep Sigh. I know your pain, David, and I commend you for trying so hard to press forward for the sake of your school, students, and the K12LTSP legacy. Thanks for laying all this out. I hope that there is adequate response and follow-up from those familiar with and able to address the Fedora+LTSP5 part of this. When dealing with the tragic migration from an LTSP4.2 to a LTSP5 installation a few years ago, I eventually made my server tri-boot. I was using Edubuntu, so, one boot option was 7.04, and the other was 7.10. The third boot option was whatever else I was testing for the sake of salvation and sanity. In your case, you may want to consider sandboxing your migration, or at least creating some sane reboot options. If you can still fall back to the EL release of the K12LTSP, you can sleep while discovering the best way to include sound.Others will advise that you create custom images for subsets of your clients. This works well, but I am not sure how deep you want to go. I do know that Asmo on the Edubuntu mailing list has been a one-man bug-squashing and testing machine, and you may find some progress with this as a boot option. He has had mostly newer hardware, but has also had success. David aka 'nubae' has had some success with OpenSuse. I have enjoyed DRBL on Debian. I wish I could offer more, like a clear solution to your problem, but, in my observation the LTSP4.2 to LTSP5 migration, including supporting the hardware we all know and love, is yet to be easy enough for the culture we groomed and supported with the glorious K12LTSP releases. --scott On Tue, Apr 21, 2009 at 7:23 AM, David Hopkins wrote: > Let me first say that this is going to sound like a rant in places. > Not much I can do about that, but ... FC10+LTSP5 has not been > performing well at all. ?I am currently at a loss to explain why. > However, since I have to have sound working, LTSP5 seems to be the > only way to ensure that sound works correctly. I have CentOS+LTSP4.2 > and that works well for everything except sound. So, the only option I > see is to get a distribution that is using LTSP5 working. Again, just > to be clear, I am using identical hardware for the comparison and > using the same login accounts, same file server, same dns, same > authentication server, etc. All hardware is 32bit, both server and > clients. (I don't even want to deal/worry with the 64bit server/32bit > client possible issues at the moment). > > Now, here is what I and the elementary school tech teacher observed > today. The following is her write-up. > > "Things did not go so well this morning. ?When all 10 computers were > in use at the same time, the delay between mouse and screen was > significant. . . ?The point of the lesson was to improve mouse > skills--not possible when there is a lag between their mouse movements > and action on their monitors. ? We muddled through the first group of > 10 students, and when the 2nd group began the exercise, I allowed the > first 10 students to open Tux Paint. ?I thought because ?Tux Paint is > running local, this would work. ?Big Mistake! ?The delay for everyone > increased dramatically, making it virtually impossible to complete the > mouse task in Starfall. ? When I tried to "QUIT" foxfire on 3 > computers, it took 3-4 minutes to return to the desktop. ? Although I > was circulating the room, trying to assist students, I glanced at the > load several times--I never saw it rise above 6. ?It mostly hovered > between 4 and 5. ?It took more than 5 minutes to successfully close > the website from 10 computers. ?During that time, I had 10 students > just waiting. > > When my second class arrived, I did not even try to use the website. > We used Tux Paint today. ?However, shortly after we got started, I > "banned" students from selecting a new piece of paper . . . ?The few > who had tried feature ?had their monitor hung-up for more than a > minute. ?That task used to respond immediately. ? There is also a > terrific feature that allows students to select any color from the > rainbow . . . ?but choosing that feature takes more than 1 full minute > to accomplish." > > This is on a system where with CentOS+LTSP4.2 I could run 25 systems > simultaneously without issues. ?She was trying to use 10. > > Notice that the load average never exceeded 6. This is dual > hyperthreaded Xeon so a load average of 4 would mean 100% utilization > although that is a bit misleading as load averages of 6-8 perform > quite well on all my other systems. Also, the system was never using > swap. In fact, memory usage never exceeded 5GB. > > So, where is the bottleneck? ?The starfall activity is flash-based (it > was the Earth Day activity). ?I know that FF3+flash is going to load > the system. ?But, This issue is not as severe with FF2+Flash 9 except > that you don't get sound half the time. ?FF3+Flash10 seems to really > slow down. ?Also, it seems that network traffic is significantly > higher with FC10+LTSP5 using ldm than with gdm. ?Can I switch back to > gdm as the default manager or is ldm it? ?I have the LDM_DIRECTX set > to TRUE so that ssh is only used for login/logout. ?And, login/logout > now takes 30+ secs compared to about 2 seconds for CentOS+LTSP4.2. > > For the local apps, launching FF3 can take over a minute. And then it > will be sluggish, even when the local hardware isn't using swap > either. > > I have this suspicion that it is a network bandwidth issue. The only > difference there is that LTSP5 uses the ltsbr0 bridge setup while > LTSP4.2 does not. ?To test this, I should be able to delete the bridge > and set up LTSP5 in the same dual NIC scenario as with LTSP4.2, > correct? ?Though I am not sure I have the skills to do so without > breaking something else. ?It might be as easy as deleting the ltsbr0 > entry and then defining the IP address for the currently-slaved NIC to > be what the ltsbr0 was defined as. > > I haven't had a chance to look at the stats from the switch (Amer.com, > SS2R24G4i ) but since I never changed the switch, only the OS, I don't > see why there would suddenly be an issue. > > As for the Tuxpaint issue. ?That is truly baffling. ?I have the same > version of Tuxpaint running on an older server and it is very > responsive. There is a hardware difference for the server ... the one > that runs very well has CPU's with only 70% the speed of the newer > server. ?The other difference is again CentOS+LTSP4.2 (using gdm) vs > FC10+LTSP5 (and ldm). > > So, something looks like it 'just isn't right' except I'm not getting > any disk I/O errors, I'm not getting a huge spike in the load ... the > system just isn't responsive. > > At this point the teacher has really reached her limit as have I. ?A > single login with a single client works fine. ?Add a few more and I > get the above. I want LTSP5 to work but I can't stay with it given the > current performance issues. ?And I have to start planning now for next > fall. ?If upgrading to FC10+LTSP5 means all my current hardware is not > acceptable, then I have a huge issue. ?I know that all my current > hardware works with FC10+LTSP5, but the performance I'm seeing is > horrible. ?I have been advocating/using K12LTSP since 2003, I really > want this to work, but right now to say I am depressed with FC10+LTSP5 > would be an understatement. > > So ... help? ?I'll be back at the school tonight to try and determine > what might be happening. And once there, sitting behind the state > firewalls, access to IRC is blocked as is all other chat capabilities. > > Sincerely, > Dave Hopkins > Newark Charter School > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From jkinney at localnetsolutions.com Tue Apr 21 20:54:09 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Tue, 21 Apr 2009 16:54:09 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: Message-ID: <1240347249.13306.16.camel@merlin.localnetsolutions.com> The basic applications are not changed much between the two versions of OS you are running. The KERNEL is radically different, however. The CentOS version has a totally different scheduler than the F10 version. Plus a bazillion other changes that have a huge effect on LTSP kernel usage. Due to sound changes, it may be challenge to get the CentOS kernel src.rpm and compile it for the F10 OS environment but it is certainly worth a try. Also, sadly, there is not a non-kernel-hackers version of tuning params that I have found. lwn.net should be consulted for details on how each kernel has evolved as well as distro kernel specs (yes: read the spec files for each kernel src.rpm to see what patches were backported/migrated/merged/etc as sometimes those cause problems). Note also that CentOS kernels are by default tuned more for big installation stuff than is the Fedora kernel. Again, I strongly suspect you are having a serious kernel issue and recommend compiling your own. The bridge interface is also a serious slowdown hog. It doesn't seem to support gigabit traffic at all. More like 230-400 Mbps. So the idea of ditching the bridge and going straight, physical nic is the best, fastest route to a speed up. Dig on the Brandon elementary pta site for a write up by William Fragakis on how to turn of the bridge process. Lastly: you mentioned hyperthreading cpu's. If they are _really_ hyperthreaded and not full multi-core, turn _off_ the hyperthreading. The F10 kernel threading will thrash royally doing stupid task swapping as it seems to not understand that the second cpu is a fake one. There is a kernel /proc flag to help this but I can't find it right now (again). Also be sure to turn off EVERY process not actively used. Some are more of a drain on cpu throughput (avahi, a zillion python applets running desktop applets for notification - system monitor should be removed!! so students won't load it up - it's a HOG) On Tue, 2009-04-21 at 13:23 -0400, David Hopkins wrote: > Let me first say that this is going to sound like a rant in places. > Not much I can do about that, but ... FC10+LTSP5 has not been > performing well at all. I am currently at a loss to explain why. > However, since I have to have sound working, LTSP5 seems to be the > only way to ensure that sound works correctly. I have CentOS+LTSP4.2 > and that works well for everything except sound. So, the only option I > see is to get a distribution that is using LTSP5 working. Again, just > to be clear, I am using identical hardware for the comparison and > using the same login accounts, same file server, same dns, same > authentication server, etc. All hardware is 32bit, both server and > clients. (I don't even want to deal/worry with the 64bit server/32bit > client possible issues at the moment). > > Now, here is what I and the elementary school tech teacher observed > today. The following is her write-up. > > "Things did not go so well this morning. When all 10 computers were > in use at the same time, the delay between mouse and screen was > significant. . . The point of the lesson was to improve mouse > skills--not possible when there is a lag between their mouse movements > and action on their monitors. We muddled through the first group of > 10 students, and when the 2nd group began the exercise, I allowed the > first 10 students to open Tux Paint. I thought because Tux Paint is > running local, this would work. Big Mistake! The delay for everyone > increased dramatically, making it virtually impossible to complete the > mouse task in Starfall. When I tried to "QUIT" foxfire on 3 > computers, it took 3-4 minutes to return to the desktop. Although I > was circulating the room, trying to assist students, I glanced at the > load several times--I never saw it rise above 6. It mostly hovered > between 4 and 5. It took more than 5 minutes to successfully close > the website from 10 computers. During that time, I had 10 students > just waiting. > > When my second class arrived, I did not even try to use the website. > We used Tux Paint today. However, shortly after we got started, I > "banned" students from selecting a new piece of paper . . . The few > who had tried feature had their monitor hung-up for more than a > minute. That task used to respond immediately. There is also a > terrific feature that allows students to select any color from the > rainbow . . . but choosing that feature takes more than 1 full minute > to accomplish." > > This is on a system where with CentOS+LTSP4.2 I could run 25 systems > simultaneously without issues. She was trying to use 10. > > Notice that the load average never exceeded 6. This is dual > hyperthreaded Xeon so a load average of 4 would mean 100% utilization > although that is a bit misleading as load averages of 6-8 perform > quite well on all my other systems. Also, the system was never using > swap. In fact, memory usage never exceeded 5GB. > > So, where is the bottleneck? The starfall activity is flash-based (it > was the Earth Day activity). I know that FF3+flash is going to load > the system. But, This issue is not as severe with FF2+Flash 9 except > that you don't get sound half the time. FF3+Flash10 seems to really > slow down. Also, it seems that network traffic is significantly > higher with FC10+LTSP5 using ldm than with gdm. Can I switch back to > gdm as the default manager or is ldm it? I have the LDM_DIRECTX set > to TRUE so that ssh is only used for login/logout. And, login/logout > now takes 30+ secs compared to about 2 seconds for CentOS+LTSP4.2. > > For the local apps, launching FF3 can take over a minute. And then it > will be sluggish, even when the local hardware isn't using swap > either. > > I have this suspicion that it is a network bandwidth issue. The only > difference there is that LTSP5 uses the ltsbr0 bridge setup while > LTSP4.2 does not. To test this, I should be able to delete the bridge > and set up LTSP5 in the same dual NIC scenario as with LTSP4.2, > correct? Though I am not sure I have the skills to do so without > breaking something else. It might be as easy as deleting the ltsbr0 > entry and then defining the IP address for the currently-slaved NIC to > be what the ltsbr0 was defined as. > > I haven't had a chance to look at the stats from the switch (Amer.com, > SS2R24G4i ) but since I never changed the switch, only the OS, I don't > see why there would suddenly be an issue. > > As for the Tuxpaint issue. That is truly baffling. I have the same > version of Tuxpaint running on an older server and it is very > responsive. There is a hardware difference for the server ... the one > that runs very well has CPU's with only 70% the speed of the newer > server. The other difference is again CentOS+LTSP4.2 (using gdm) vs > FC10+LTSP5 (and ldm). > > So, something looks like it 'just isn't right' except I'm not getting > any disk I/O errors, I'm not getting a huge spike in the load ... the > system just isn't responsive. > > At this point the teacher has really reached her limit as have I. A > single login with a single client works fine. Add a few more and I > get the above. I want LTSP5 to work but I can't stay with it given the > current performance issues. And I have to start planning now for next > fall. If upgrading to FC10+LTSP5 means all my current hardware is not > acceptable, then I have a huge issue. I know that all my current > hardware works with FC10+LTSP5, but the performance I'm seeing is > horrible. I have been advocating/using K12LTSP since 2003, I really > want this to work, but right now to say I am depressed with FC10+LTSP5 > would be an understatement. > > So ... help? I'll be back at the school tonight to try and determine > what might be happening. And once there, sitting behind the state > firewalls, access to IRC is blocked as is all other chat capabilities. > > Sincerely, > Dave Hopkins > Newark Charter School > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From joseph.bishay at gmail.com Tue Apr 21 21:35:36 2009 From: joseph.bishay at gmail.com (Joseph Bishay) Date: Tue, 21 Apr 2009 17:35:36 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <1240347249.13306.16.camel@merlin.localnetsolutions.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> Message-ID: Hello, On Tue, Apr 21, 2009 at 4:54 PM, James P. Kinney III wrote: > Lastly: you mentioned hyperthreading cpu's. If they are _really_ > hyperthreaded and not full multi-core, turn _off_ the hyperthreading. > The F10 kernel threading will thrash royally doing stupid task swapping > as it seems to not understand that the second cpu is a fake one. There > is a kernel /proc flag to help this but I can't find it right now > (again). This is a really interesting suggestion. I am using a true hyper-threading Pentium 4 with Fedora 10. Is there a simple benchmark I could run on the LTSP server to compare before and after turning the hyper-threading on and off? Something more objective than 'it feels faster'? :) > Also be sure to turn off EVERY process not actively used. Some are more > of a drain on cpu throughput (avahi, a zillion python applets running > desktop applets for notification - system monitor should be removed!! so > students won't load it up - it's a HOG) Is there a definitive list of processes that should be kept on, and everything else to be turned off? Thank you Joseph From microman at cmosnetworks.com Tue Apr 21 21:50:21 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Tue, 21 Apr 2009 17:50:21 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: Message-ID: <49EE3F9D.6080508@cmosnetworks.com> I know you're at your wits' end here, and so is that teacher. This is not good. Remind us what sound card is in these thin clients, would you? --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! David Hopkins wrote: > Let me first say that this is going to sound like a rant in places. > Not much I can do about that, but ... FC10+LTSP5 has not been > performing well at all. I am currently at a loss to explain why. > However, since I have to have sound working, LTSP5 seems to be the > only way to ensure that sound works correctly. I have CentOS+LTSP4.2 > and that works well for everything except sound. So, the only option I > see is to get a distribution that is using LTSP5 working. Again, just > to be clear, I am using identical hardware for the comparison and > using the same login accounts, same file server, same dns, same > authentication server, etc. All hardware is 32bit, both server and > clients. (I don't even want to deal/worry with the 64bit server/32bit > client possible issues at the moment). > > Now, here is what I and the elementary school tech teacher observed > today. The following is her write-up. > > "Things did not go so well this morning. When all 10 computers were > in use at the same time, the delay between mouse and screen was > significant. . . The point of the lesson was to improve mouse > skills--not possible when there is a lag between their mouse movements > and action on their monitors. We muddled through the first group of > 10 students, and when the 2nd group began the exercise, I allowed the > first 10 students to open Tux Paint. I thought because Tux Paint is > running local, this would work. Big Mistake! The delay for everyone > increased dramatically, making it virtually impossible to complete the > mouse task in Starfall. When I tried to "QUIT" foxfire on 3 > computers, it took 3-4 minutes to return to the desktop. Although I > was circulating the room, trying to assist students, I glanced at the > load several times--I never saw it rise above 6. It mostly hovered > between 4 and 5. It took more than 5 minutes to successfully close > the website from 10 computers. During that time, I had 10 students > just waiting. > > When my second class arrived, I did not even try to use the website. > We used Tux Paint today. However, shortly after we got started, I > "banned" students from selecting a new piece of paper . . . The few > who had tried feature had their monitor hung-up for more than a > minute. That task used to respond immediately. There is also a > terrific feature that allows students to select any color from the > rainbow . . . but choosing that feature takes more than 1 full minute > to accomplish." > > This is on a system where with CentOS+LTSP4.2 I could run 25 systems > simultaneously without issues. She was trying to use 10. > > Notice that the load average never exceeded 6. This is dual > hyperthreaded Xeon so a load average of 4 would mean 100% utilization > although that is a bit misleading as load averages of 6-8 perform > quite well on all my other systems. Also, the system was never using > swap. In fact, memory usage never exceeded 5GB. > > So, where is the bottleneck? The starfall activity is flash-based (it > was the Earth Day activity). I know that FF3+flash is going to load > the system. But, This issue is not as severe with FF2+Flash 9 except > that you don't get sound half the time. FF3+Flash10 seems to really > slow down. Also, it seems that network traffic is significantly > higher with FC10+LTSP5 using ldm than with gdm. Can I switch back to > gdm as the default manager or is ldm it? I have the LDM_DIRECTX set > to TRUE so that ssh is only used for login/logout. And, login/logout > now takes 30+ secs compared to about 2 seconds for CentOS+LTSP4.2. > > For the local apps, launching FF3 can take over a minute. And then it > will be sluggish, even when the local hardware isn't using swap > either. > > I have this suspicion that it is a network bandwidth issue. The only > difference there is that LTSP5 uses the ltsbr0 bridge setup while > LTSP4.2 does not. To test this, I should be able to delete the bridge > and set up LTSP5 in the same dual NIC scenario as with LTSP4.2, > correct? Though I am not sure I have the skills to do so without > breaking something else. It might be as easy as deleting the ltsbr0 > entry and then defining the IP address for the currently-slaved NIC to > be what the ltsbr0 was defined as. > > I haven't had a chance to look at the stats from the switch (Amer.com, > SS2R24G4i ) but since I never changed the switch, only the OS, I don't > see why there would suddenly be an issue. > > As for the Tuxpaint issue. That is truly baffling. I have the same > version of Tuxpaint running on an older server and it is very > responsive. There is a hardware difference for the server ... the one > that runs very well has CPU's with only 70% the speed of the newer > server. The other difference is again CentOS+LTSP4.2 (using gdm) vs > FC10+LTSP5 (and ldm). > > So, something looks like it 'just isn't right' except I'm not getting > any disk I/O errors, I'm not getting a huge spike in the load ... the > system just isn't responsive. > > At this point the teacher has really reached her limit as have I. A > single login with a single client works fine. Add a few more and I > get the above. I want LTSP5 to work but I can't stay with it given the > current performance issues. And I have to start planning now for next > fall. If upgrading to FC10+LTSP5 means all my current hardware is not > acceptable, then I have a huge issue. I know that all my current > hardware works with FC10+LTSP5, but the performance I'm seeing is > horrible. I have been advocating/using K12LTSP since 2003, I really > want this to work, but right now to say I am depressed with FC10+LTSP5 > would be an understatement. > > So ... help? I'll be back at the school tonight to try and determine > what might be happening. And once there, sitting behind the state > firewalls, access to IRC is blocked as is all other chat capabilities. > > Sincerely, > Dave Hopkins > Newark Charter School > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From robark at gmail.com Tue Apr 21 22:01:42 2009 From: robark at gmail.com (Robert Arkiletian) Date: Tue, 21 Apr 2009 15:01:42 -0700 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <1240347249.13306.16.camel@merlin.localnetsolutions.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> Message-ID: On Tue, Apr 21, 2009 at 1:54 PM, James P. Kinney III wrote: ... > Note also that CentOS kernels are by default tuned more for big > installation stuff than is the Fedora kernel. > > Again, I strongly suspect you are having a serious kernel issue and > recommend compiling your own. I recommend against that. There are MANY settings in configuring and compiling a kernel. You could end up making things worse and spend a lot of time doing it. I think (for the most part) the default settings should be fine. > > The bridge interface is also a serious slowdown hog. It doesn't seem to > support gigabit traffic at all. More like 230-400 Mbps. So the idea of > ditching the bridge and going straight, physical nic is the best, > fastest route to a speed up. Dig on the Brandon elementary pta site for > a write up by William Fragakis on how to turn of the bridge process. *****If this is true, then this needs to be fixed****** Does anyone have any evidence/experience that the virtual bridge causes a slowdown? > > Lastly: you mentioned hyperthreading cpu's. If they are _really_ > hyperthreaded and not full multi-core, turn _off_ the hyperthreading. > The F10 kernel threading will thrash royally doing stupid task swapping > as it seems to not understand that the second cpu is a fake one. There > is a kernel /proc flag to help this but I can't find it right now > (again). The bios should provide you with an option to turn it off. But I have tested HT and I get a slight benefit with ltsp. > > Also be sure to turn off EVERY process not actively used. Some are more > of a drain on cpu throughput (avahi, a zillion python applets running > desktop applets for notification - system monitor should be removed!! so > students won't load it up - it's a HOG) > Good point. That's why I use Icewm. -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From rowens at ptd.net Tue Apr 21 22:16:45 2009 From: rowens at ptd.net (Rob Owens) Date: Tue, 21 Apr 2009 18:16:45 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: Message-ID: <20090421221645.GC31988@aurora.owens.net> At least you have a teacher who is giving you useful feedback. I used CentOS+LTSP4.2 for a while and sound worked (not sure if it worked in flash apps, though). I used the alsa package that Gadi put together, here: http://www.ltsp.org/twiki/bin/view/Ltsp/WorkInProgress#esd_ALSA_sound_on_LTSP_4_2 I'm not sure if it will work on Fedora because of the bridge interface, but see my notes about running LTSP 4.2 and LTSP 5 on the same server, here: http://www.ltsp.org/twiki/bin/view/Ltsp/Ltsp5SameServerLTSP42 At home I use Debian Lenny and LTSP 5. Tuxpaint works, as does sound. Of course, I don't have many thin clients in use simultaneously. If you can't Fedora working, and you can't get sound working on CentOS, you might consider giving Lenny a try. Ubuntu and OpenSuse have LTSP 5, too, although I can't make any recommendations about them since I don't use them. I'm not trying to lure you away from your distro of choice, but you need something to work and quick. I'd recommend you at least try another distro in a virtual machine or something, and tinker with it during the time you can't work on the Fedora server (such as while waiting for answers from this list). Good luck! -Rob On Tue, Apr 21, 2009 at 01:23:48PM -0400, David Hopkins wrote: > Let me first say that this is going to sound like a rant in places. > Not much I can do about that, but ... FC10+LTSP5 has not been > performing well at all. I am currently at a loss to explain why. > However, since I have to have sound working, LTSP5 seems to be the > only way to ensure that sound works correctly. I have CentOS+LTSP4.2 > and that works well for everything except sound. So, the only option I > see is to get a distribution that is using LTSP5 working. Again, just > to be clear, I am using identical hardware for the comparison and > using the same login accounts, same file server, same dns, same > authentication server, etc. All hardware is 32bit, both server and > clients. (I don't even want to deal/worry with the 64bit server/32bit > client possible issues at the moment). > > Now, here is what I and the elementary school tech teacher observed > today. The following is her write-up. > > "Things did not go so well this morning. When all 10 computers were > in use at the same time, the delay between mouse and screen was > significant. . . The point of the lesson was to improve mouse > skills--not possible when there is a lag between their mouse movements > and action on their monitors. We muddled through the first group of > 10 students, and when the 2nd group began the exercise, I allowed the > first 10 students to open Tux Paint. I thought because Tux Paint is > running local, this would work. Big Mistake! The delay for everyone > increased dramatically, making it virtually impossible to complete the > mouse task in Starfall. When I tried to "QUIT" foxfire on 3 > computers, it took 3-4 minutes to return to the desktop. Although I > was circulating the room, trying to assist students, I glanced at the > load several times--I never saw it rise above 6. It mostly hovered > between 4 and 5. It took more than 5 minutes to successfully close > the website from 10 computers. During that time, I had 10 students > just waiting. > > When my second class arrived, I did not even try to use the website. > We used Tux Paint today. However, shortly after we got started, I > "banned" students from selecting a new piece of paper . . . The few > who had tried feature had their monitor hung-up for more than a > minute. That task used to respond immediately. There is also a > terrific feature that allows students to select any color from the > rainbow . . . but choosing that feature takes more than 1 full minute > to accomplish." > > This is on a system where with CentOS+LTSP4.2 I could run 25 systems > simultaneously without issues. She was trying to use 10. > > Notice that the load average never exceeded 6. This is dual > hyperthreaded Xeon so a load average of 4 would mean 100% utilization > although that is a bit misleading as load averages of 6-8 perform > quite well on all my other systems. Also, the system was never using > swap. In fact, memory usage never exceeded 5GB. > > So, where is the bottleneck? The starfall activity is flash-based (it > was the Earth Day activity). I know that FF3+flash is going to load > the system. But, This issue is not as severe with FF2+Flash 9 except > that you don't get sound half the time. FF3+Flash10 seems to really > slow down. Also, it seems that network traffic is significantly > higher with FC10+LTSP5 using ldm than with gdm. Can I switch back to > gdm as the default manager or is ldm it? I have the LDM_DIRECTX set > to TRUE so that ssh is only used for login/logout. And, login/logout > now takes 30+ secs compared to about 2 seconds for CentOS+LTSP4.2. > > For the local apps, launching FF3 can take over a minute. And then it > will be sluggish, even when the local hardware isn't using swap > either. > > I have this suspicion that it is a network bandwidth issue. The only > difference there is that LTSP5 uses the ltsbr0 bridge setup while > LTSP4.2 does not. To test this, I should be able to delete the bridge > and set up LTSP5 in the same dual NIC scenario as with LTSP4.2, > correct? Though I am not sure I have the skills to do so without > breaking something else. It might be as easy as deleting the ltsbr0 > entry and then defining the IP address for the currently-slaved NIC to > be what the ltsbr0 was defined as. > > I haven't had a chance to look at the stats from the switch (Amer.com, > SS2R24G4i ) but since I never changed the switch, only the OS, I don't > see why there would suddenly be an issue. > > As for the Tuxpaint issue. That is truly baffling. I have the same > version of Tuxpaint running on an older server and it is very > responsive. There is a hardware difference for the server ... the one > that runs very well has CPU's with only 70% the speed of the newer > server. The other difference is again CentOS+LTSP4.2 (using gdm) vs > FC10+LTSP5 (and ldm). > > So, something looks like it 'just isn't right' except I'm not getting > any disk I/O errors, I'm not getting a huge spike in the load ... the > system just isn't responsive. > > At this point the teacher has really reached her limit as have I. A > single login with a single client works fine. Add a few more and I > get the above. I want LTSP5 to work but I can't stay with it given the > current performance issues. And I have to start planning now for next > fall. If upgrading to FC10+LTSP5 means all my current hardware is not > acceptable, then I have a huge issue. I know that all my current > hardware works with FC10+LTSP5, but the performance I'm seeing is > horrible. I have been advocating/using K12LTSP since 2003, I really > want this to work, but right now to say I am depressed with FC10+LTSP5 > would be an understatement. > > So ... help? I'll be back at the school tonight to try and determine > what might be happening. And once there, sitting behind the state > firewalls, access to IRC is blocked as is all other chat capabilities. > > Sincerely, > Dave Hopkins > Newark Charter School > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From wtogami at redhat.com Tue Apr 21 23:42:05 2009 From: wtogami at redhat.com (Warren Togami) Date: Tue, 21 Apr 2009 19:42:05 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: Message-ID: <49EE59CD.6090503@redhat.com> On 04/21/2009 01:23 PM, David Hopkins wrote: > > As for the Tuxpaint issue. That is truly baffling. I have the same > version of Tuxpaint running on an older server and it is very > responsive. There is a hardware difference for the server ... the one > that runs very well has CPU's with only 70% the speed of the newer > server. The other difference is again CentOS+LTSP4.2 (using gdm) vs > FC10+LTSP5 (and ldm). > I totally can't reproduce these problems with Tuxpaint. What is the hardware of the clients, especially video card? Is it exactly the same problem with clients of entirely different graphic chipsets? Warren From joebaker at dcresearch.com Wed Apr 22 01:09:58 2009 From: joebaker at dcresearch.com (Joseph W. Baker) Date: Tue, 21 Apr 2009 19:09:58 -0600 (GMT-06:00) Subject: [K12OSN] FireFox not responding In-Reply-To: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> Message-ID: <15849745.2971240362598225.JavaMail.root@mail2> Try pkill -u username firefox -Joe Baker ----- Original Message ----- From: "Mel Wade" To: "Support list for open source software in schools." Sent: Sunday, April 19, 2009 11:14:54 AM GMT -07:00 U.S. Mountain Time (Arizona) Subject: [K12OSN] FireFox not responding We continually get users who have messages saying that FireFox is already running and won't let them login. If I delete the .mozilla folder in their Home directory everything is fine. How can I keep this from happening in the future? -- Mel Wade "A common mistake people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools." - Douglas Adams http://www.melwade.com _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see From jkinney at localnetsolutions.com Wed Apr 22 01:16:40 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Tue, 21 Apr 2009 21:16:40 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> Message-ID: <1240363000.13306.20.camel@merlin.localnetsolutions.com> On Tue, 2009-04-21 at 17:35 -0400, Joseph Bishay wrote: > Hello, > > On Tue, Apr 21, 2009 at 4:54 PM, James P. Kinney III > wrote: > > Lastly: you mentioned hyperthreading cpu's. If they are _really_ > > hyperthreaded and not full multi-core, turn _off_ the hyperthreading. > > The F10 kernel threading will thrash royally doing stupid task swapping > > as it seems to not understand that the second cpu is a fake one. There > > is a kernel /proc flag to help this but I can't find it right now > > (again). > > This is a really interesting suggestion. I am using a true > hyper-threading Pentium 4 with Fedora 10. Is there a simple benchmark > I could run on the LTSP server to compare before and after turning the > hyper-threading on and off? Something more objective than 'it feels > faster'? :) Other than loading up a room full of TC's and watching the performance drag, not that I have used. It may be possible to launch multiple runs of the same app across the room full with a timer running and helpers doing the same procedure with HT on and off. > > > Also be sure to turn off EVERY process not actively used. Some are more > > of a drain on cpu throughput (avahi, a zillion python applets running > > desktop applets for notification - system monitor should be removed!! so > > students won't load it up - it's a HOG) > > Is there a definitive list of processes that should be kept on, and > everything else to be turned off? That list is too variable based on system usage. The default setup provides maximum availability of capabilities. > > Thank you > Joseph > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From jkinney at localnetsolutions.com Wed Apr 22 01:39:36 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Tue, 21 Apr 2009 21:39:36 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> Message-ID: <1240364376.13306.39.camel@merlin.localnetsolutions.com> On Tue, 2009-04-21 at 15:01 -0700, Robert Arkiletian wrote: > On Tue, Apr 21, 2009 at 1:54 PM, James P. Kinney III > wrote: > ... > > Note also that CentOS kernels are by default tuned more for big > > installation stuff than is the Fedora kernel. > > > > Again, I strongly suspect you are having a serious kernel issue and > > recommend compiling your own. > > I recommend against that. There are MANY settings in configuring and > compiling a kernel. You could end up making things worse and spend a > lot of time doing it. I think (for the most part) the default settings > should be fine. I mostly agree as the process is very "admin-y". It requires the kernel builder to know what they are doing. I've built many a clunker :-) and quite a few that didn't boot (learned about mkinitrd). The upside of learning the process is the admin can tune a kernel to provide exactly the needed parts and no more so that user space has all the rest of the system resources. Granted RAM is way less expensive than it used to be but some process wake up and poll things and that eats clock cycles. > > > > > The bridge interface is also a serious slowdown hog. It doesn't seem to > > support gigabit traffic at all. More like 230-400 Mbps. So the idea of > > ditching the bridge and going straight, physical nic is the best, > > fastest route to a speed up. Dig on the Brandon elementary pta site for > > a write up by William Fragakis on how to turn of the bridge process. > > *****If this is true, then this needs to be fixed****** > Does anyone have any evidence/experience that the virtual bridge > causes a slowdown? My tests were with a pre-release version and it was a synthetic network throughput speed test. My memory of the numbers are probably pretty close to the final results but I apparently buried the test procedure and actual data somewhere stupid on my archive system. Grr. I can recall my reason for testing was the idea that a fake interface requires more system overhead than a raw dump to hardware. So a clock tick to send data to bridge plus another to make the redirect point to the physical plus a third to actually send the data. Effectively it's a chained pipe. So a bridge theoretically should take twice the time to exit to the NIC as a straight NIC. But I'm not 100% that's how it works. Virtual interfaces (eth0:1) act exactly like physical ones in terms of throughput if the real one is not passing data. > > > > > Lastly: you mentioned hyperthreading cpu's. If they are _really_ > > hyperthreaded and not full multi-core, turn _off_ the hyperthreading. > > The F10 kernel threading will thrash royally doing stupid task swapping > > as it seems to not understand that the second cpu is a fake one. There > > is a kernel /proc flag to help this but I can't find it right now > > (again). > > The bios should provide you with an option to turn it off. But I have > tested HT and I get a slight benefit with ltsp. It is certainly possible to get a performance gain with HT. It is very system specific. As long as a kernel task can be run without being swapped out, HT will show a speedup. Otherwise, loading a cpu task back to either core will make the HT cpu act like a single cpu. A thrashing situation under heavy load (2+ with large memory needs) and the process swap begins to dominate the clock cycles. > > > > > Also be sure to turn off EVERY process not actively used. Some are more > > of a drain on cpu throughput (avahi, a zillion python applets running > > desktop applets for notification - system monitor should be removed!! so > > students won't load it up - it's a HOG) > > > > Good point. That's why I use Icewm. Yep. Gnome is a hog and so is KDE. But when a student loads firefox, the entire system is now ready to run gnome and it "looks" better. KDE is lighter than gnome but still hits the same system usage on loading firefox. My long time ago tests with xfce showed greater system usage after loading firefox that gnome or kde after loading firefox. > > -- > Robert Arkiletian > Eric Hamber Secondary, Vancouver, Canada > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From lesmikesell at gmail.com Wed Apr 22 04:02:03 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 21 Apr 2009 23:02:03 -0500 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: Message-ID: <49EE96BB.5090500@gmail.com> David Hopkins wrote: > > I have this suspicion that it is a network bandwidth issue. The only > difference there is that LTSP5 uses the ltsbr0 bridge setup while > LTSP4.2 does not. To test this, I should be able to delete the bridge > and set up LTSP5 in the same dual NIC scenario as with LTSP4.2, > correct? Though I am not sure I have the skills to do so without > breaking something else. It might be as easy as deleting the ltsbr0 > entry and then defining the IP address for the currently-slaved NIC to > be what the ltsbr0 was defined as. > > I haven't had a chance to look at the stats from the switch (Amer.com, > SS2R24G4i ) but since I never changed the switch, only the OS, I don't > see why there would suddenly be an issue. Be sure to test the obvious thing - this sounds like what would happen if you had a duplex mismatch between the server and switch. If the switch is managed and configured to force full duplex, you have to have the matching setting in the OS configuration. You are probably better off letting both negotiate. If either ifconfig or the switch port stats show any errors, look for the problem in the network first. -- Les Mikesell lesmikesell at gmail.com From dahopkins429 at gmail.com Wed Apr 22 12:48:18 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 22 Apr 2009 08:48:18 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <1240347249.13306.16.camel@merlin.localnetsolutions.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> Message-ID: On Tue, Apr 21, 2009 at 4:54 PM, James P. Kinney III wrote: > The KERNEL is radically different, however. The CentOS version has a > totally different scheduler than the F10 version. Plus a bazillion other > changes that have a huge effect on LTSP kernel usage. Due to sound > changes, it may be challenge to get the CentOS kernel src.rpm and > compile it for the F10 OS environment but it is certainly worth a try. I would probably lose what hair I have left if I tried this right now. I have 'spare' 8 processor and dual processor Dell servers that the state was throwing away that I can learn on for this but that won't be till summer (it is amazing what is sometimes considered obsolete). > Note also that CentOS kernels are by default tuned more for big > installation stuff than is the Fedora kernel. yep, there is a reason that CentOS/RHEL exists. I would prefer to stay with CentOS but I have to resolve sound (talk about beating a dead horse issue) and with LTSP5 I have to admit that sound 'just works' in all the apps I have to have (except for Audacity but I've found some documentation that implies it can be made to work). > > Again, I strongly suspect you are having a serious kernel issue and > recommend compiling your own. It is possible that this is the case. I do know that I have to run mkinitrd to get the kernel to boot properly since I have an Adaptec 2010s controller. (I20 block device). There is an outstanding bug wrt this raid card which will hopefully be resolved 'soon'. I have to manually apply a patch/fix. I'll look at how mkinitrd really works and see if that patch to make the raid card work isn't causing an issue. However, I've tested disk I/O and I don't see any issues with throughput. > The bridge interface is also a serious slowdown hog. It doesn't seem to > support gigabit traffic at all. More like 230-400 Mbps. So the idea of > ditching the bridge and going straight, physical nic is the best, > fastest route to a speed up. Dig on the Brandon elementary pta site for > a write up by William Fragakis on how to turn of the bridge process. I'll look for the notes on how to do this. The teacher's lesson plan for the next few sessions involves using OpenOffice/StarOffice and this works without any issues so I can flip the prior server back into service if I break anything. That gives me a weekend in the worst case to rebuild from scratch. I'll also say that I've tested throughput on the external network-facing NIC and it is what I would expect (it is the same as my systems using CentOS). I have tested the bridge though. > Lastly: you mentioned hyperthreading cpu's. If they are _really_ > hyperthreaded and not full multi-core, turn _off_ the hyperthreading. > The F10 kernel threading will thrash royally doing stupid task swapping > as it seems to not understand that the second cpu is a fake one. There > is a kernel /proc flag to help this but I can't find it right now > (again). They are hyperthreaded Xeons. They have performed very well until recently. Though, how would I detect thrashing? > Also be sure to turn off EVERY process not actively used. Some are more > of a drain on cpu throughput (avahi, a zillion python applets running > desktop applets for notification - system monitor should be removed!! so > students won't load it up - it's a HOG) I've chkconfig'ed to off everything I thought I could but I'll recheck since I know I didn't turn off avahi. Is there a list of essential services somewhere? Things like bluetooth and the like I know I don't need on a production system but a nice list would be wonderful. Sincerely, Dave Hopkins Newark Charter School From dahopkins429 at gmail.com Wed Apr 22 12:57:14 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 22 Apr 2009 08:57:14 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49EE3F9D.6080508@cmosnetworks.com> References: <49EE3F9D.6080508@cmosnetworks.com> Message-ID: On Tue, Apr 21, 2009 at 5:50 PM, "Terrell Prud? Jr." wrote: > I know you're at your wits' end here, and so is that teacher.? This is not > good. > > Remind us what sound card is in these thin clients, would you? > > --TP Actually, the systems are Dell GX150's with either 256Mb or 512Mb of memory.. http://support.dell.com/support/edocs/systems/opgx150/en/ug/specs.htm#audio Though, does that mean I might have the i810 video issue? For whatever reason, I thought the card was detected as an ati. I have to check this. From dahopkins429 at gmail.com Wed Apr 22 13:02:42 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 22 Apr 2009 09:02:42 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49EE59CD.6090503@redhat.com> References: <49EE59CD.6090503@redhat.com> Message-ID: On Tue, Apr 21, 2009 at 7:42 PM, Warren Togami wrote: > On 04/21/2009 01:23 PM, David Hopkins wrote: >> >> As for the Tuxpaint issue. ?That is truly baffling. ?I have the same >> version of Tuxpaint running on an older server and it is very >> responsive. There is a hardware difference for the server ... the one >> that runs very well has CPU's with only 70% the speed of the newer >> server. ?The other difference is again CentOS+LTSP4.2 (using gdm) vs >> FC10+LTSP5 (and ldm). >> > > I totally can't reproduce these problems with Tuxpaint. > > What is the hardware of the clients, especially video card? Dell GX150: http://support.dell.com/support/edocs/systems/opgx150/en/ug/specs.htm > > Is it exactly the same problem with clients of entirely different graphic > chipsets? If I only have 1 or 2 clients connected, there is a slight delay. Add more clients and the delay gets worse. Could be graphics and I'll check but just feels like a network issue? Still, I have a older gateway laptop (same vintage as the Dell GX150's) that I PXE boot from the various servers since it is easy to carry around and also use during the initial build. I don't remember seeing this issue with it. I'll check it again tonight. Sincerely, Dave Hopkins From sruiz at canterburyschool.org Wed Apr 22 13:13:06 2009 From: sruiz at canterburyschool.org (=?ISO-8859-1?Q?=22Sim=F3n_A=2E_Ruiz=22?=) Date: Wed, 22 Apr 2009 09:13:06 -0400 Subject: [K12OSN] FireFox not responding In-Reply-To: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> References: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> Message-ID: <49EF17E2.1090101@canterburyschool.org> Mel Wade wrote: > We continually get users who have messages saying that FireFox is > already running and won't let them login. If I delete the .mozilla > folder in their Home directory everything is fine. How can I keep this > from happening in the future? > > -- > Mel Wade > "A common mistake people make when trying to design something completely > foolproof is to underestimate the ingenuity of complete fools." - > Douglas Adams > http://www.melwade.com Mel, I'm surprised noone has mentioned the "parent.lock" or ".parentlock" file that Firefox keeps in the profile directory to determine whether it's running or not. If Firefox dies or loses access to the profile (e.g. it's on a network drive) before it removes that file (which it does during a normal, orderly shutdown), it'll be there telling Firefox that the profile is "in use". Try finding and deleting that. Sim?n From wtogami at redhat.com Wed Apr 22 13:39:32 2009 From: wtogami at redhat.com (Warren Togami) Date: Wed, 22 Apr 2009 09:39:32 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <49EE59CD.6090503@redhat.com> Message-ID: <49EF1E14.5070505@redhat.com> On 04/22/2009 09:02 AM, David Hopkins wrote: > > If I only have 1 or 2 clients connected, there is a slight delay. Add > more clients and the delay gets worse. Could be graphics and I'll > check but just feels like a network issue? Still, I have a older > gateway laptop (same vintage as the Dell GX150's) that I PXE boot from > the various servers since it is easy to carry around and also use > during the initial build. I don't remember seeing this issue with it. > I'll check it again tonight. > In October we POUNDED the heck out of a single server with a gigabit ethernet port with 8 clients doing full screen video, tuxpaint and other, firefox and other apps. It fell apart quickly with a flaky switch. We replaced the switch and it worked beautifully. In my experience these things failing is almost always a broken switch. My second guess is the graphic driver is having trouble with this particular graphic card. You really need to verify this by testing entirely different client hardware to narrow it down to network vs. graphic driver. Warren From dahopkins429 at gmail.com Wed Apr 22 13:44:55 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 22 Apr 2009 09:44:55 -0400 Subject: [K12OSN] FireFox not responding In-Reply-To: <49EF17E2.1090101@canterburyschool.org> References: <43080f460904191114g41fb7ccenb7a963997619b810@mail.gmail.com> <49EF17E2.1090101@canterburyschool.org> Message-ID: > Mel, > > I'm surprised noone has mentioned the "parent.lock" or ".parentlock" file > that Firefox keeps in the profile directory to determine whether it's > running or not. > > If Firefox dies or loses access to the profile (e.g. it's on a network > drive) before it removes that file (which it does during a normal, orderly > shutdown), it'll be there telling Firefox that the profile is "in use". > > Try finding and deleting that. > Thanks for the reminder on this possibility. It is a quick fix. I actually put an icon on the student's desktops that does just this awhile back and forgot about it. Sincerely, Dave Hopkins From SHarbour at nwresd.k12.or.us Wed Apr 22 14:08:39 2009 From: SHarbour at nwresd.k12.or.us (Sean Harbour) Date: Wed, 22 Apr 2009 07:08:39 -0700 Subject: [K12OSN] Re: More feedback on Fedora 10 + LTSP Message-ID: <6069A203FD01884885C037F81DD75080032AAC0D74@wsc-mail-01.intra.nwresd.k12.or.us> Here is some more feedback on Fedora 10 + LTSP 5. I have 2 types of servers in production: Dual Opteron with 8 GB RAM, dual Gig ethernet. Quad Xeon with 8GB RAM, dual Gig ethernet. The servers are using two nics, one for internet/network access, and each is plugged into a dedicated 24 port private switch over Gigabit that all the thin clients use. We are currently using the ltspbr0 virtual bridge device. Perhaps the only unusual aspect is that the server is joined to a MS AD domain for student authentication and access to shared network and home folders hosted on the MS side. Both systems will only handle up to about 15 thin clients using Fedora 10 + LTSP 5 64 bit. Typically activities are web browsing and working on reports. Removing unneeded services/applications helps minimally. Random students (About 15% at any one time, sometimes more, sometimes less) are experiencing the firefox already running message, slow app performance, and occasionally the keyboard mappings will scramble for no apparent reason. The keyboard scrambling can usually be recreated by pretending to be a frustrated student and pressing keys repeatedly while an app is stalled. Also, some processes seem to hang around forever after a student logs out. After a week, the process list "ps aux | wc -l" balloons from 173 with one user after a reboot to over 1500 with one user logged in. Killing defunct processes manually is not working very well. Fedora 10 LTSP 5 requires constant watching and manual intervention by the administrator. Rebooting the server nightly helps somewhat in reducing the hourly calls to my desk. Warren T., if you would like direct access to one of these servers during a typical classroom day, I think I can arrange it. Let me know. Thanks, Sean Harbour Northwest Regional ESD From Marc.Fromm at wwu.edu Wed Apr 22 15:25:54 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Wed, 22 Apr 2009 08:25:54 -0700 Subject: [K12OSN] client ip VS my ip during boot Message-ID: I have set the client IPs based on mac addresses in the dhcp file. When the client boots it reports CLIENT IP: the ip that I assigned. It then reports my IP seems to be: a different IP than what I assigned. Is the conflict between CLIENT IP and My IP seems to be the problem why this client wont boot? The boot eventually fails with the message that is cannot find kernel image: linux. Thanks Marc -------------- next part -------------- An HTML attachment was scrubbed... URL: From wtogami at redhat.com Wed Apr 22 15:41:04 2009 From: wtogami at redhat.com (Warren Togami) Date: Wed, 22 Apr 2009 11:41:04 -0400 Subject: [K12OSN] Re: More feedback on Fedora 10 + LTSP In-Reply-To: <6069A203FD01884885C037F81DD75080032AAC0D74@wsc-mail-01.intra.nwresd.k12.or.us> References: <6069A203FD01884885C037F81DD75080032AAC0D74@wsc-mail-01.intra.nwresd.k12.or.us> Message-ID: <49EF3A90.7010407@redhat.com> On 04/22/2009 10:08 AM, Sean Harbour wrote: > Also, some processes seem to hang around forever after a student logs > out. After a week, the process list "ps aux | wc -l" balloons from > 173 with one user after a reboot to over 1500 with one user logged > in. What processes specifically are persisting beyond logout? Warren From microman at cmosnetworks.com Wed Apr 22 17:04:44 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 22 Apr 2009 13:04:44 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <49EE3F9D.6080508@cmosnetworks.com> Message-ID: <49EF4E2C.4040105@cmosnetworks.com> David Hopkins wrote: > On Tue, Apr 21, 2009 at 5:50 PM, "Terrell Prud? Jr." > wrote: > >> I know you're at your wits' end here, and so is that teacher. This is not >> good. >> >> Remind us what sound card is in these thin clients, would you? >> >> --TP >> > > Actually, the systems are Dell GX150's with either 256Mb or 512Mb of > memory.. http://support.dell.com/support/edocs/systems/opgx150/en/ug/specs.htm#audio > > Though, does that mean I might have the i810 video issue? For > whatever reason, I thought the card was detected as an ati. I have to > check this. > That sound card (looks like it's built-in) says it has SoundBlaster emulation, so when you were running it w/ K12LTSP 5EL, did you by chance try turning on "SOUND=Y" in /opt/ltsp/i386/etc/lts.conf, and also the setting from here? http://www.k12ltsp.org/mediawiki/index.php/Technical:Sound Something along these lines.... SMODULE_01 = "sb io=0x220 irq=5 dma=1" That sort of thing always worked for me with cards that supported SB emulation, including (especially!) the SoundBlaster 16 and the SoundBlaster AWE32/64 cards themselves. --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From SHarbour at nwresd.k12.or.us Wed Apr 22 17:24:43 2009 From: SHarbour at nwresd.k12.or.us (Sean Harbour) Date: Wed, 22 Apr 2009 10:24:43 -0700 Subject: [K12OSN] RE: K12OSN Digest, Vol 62, Issue 20 In-Reply-To: <20090422160035.AE4658E0603@hormel.redhat.com> References: <20090422160035.AE4658E0603@hormel.redhat.com> Message-ID: <6069A203FD01884885C037F81DD75080032AAC0D76@wsc-mail-01.intra.nwresd.k12.or.us> >> Also, some processes seem to hang around forever after a student logs >> out. After a week, the process list "ps aux | wc -l" balloons from >> 173 with one user after a reboot to over 1500 with one user logged >> in. >What processes specifically are persisting beyond logout? >Warren --------------------------------- The outstanding ones seem to be /usb/bin/pulseaudio --start and /usb/bin/libexec/pulse/gconf-helper After about a week, there were over 500 of these processes after everyone had logged off. But, there is also a lot of random gnome desktop processes that seem to live forever as well. I'll send you capture of the ps aux output from yesterday morning before anybody had logged in. Thanks, Sean ------------------------------ From microman at cmosnetworks.com Wed Apr 22 17:30:59 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 22 Apr 2009 13:30:59 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49EF4E2C.4040105@cmosnetworks.com> References: <49EE3F9D.6080508@cmosnetworks.com> <49EF4E2C.4040105@cmosnetworks.com> Message-ID: <49EF5453.6020809@cmosnetworks.com> Terrell Prud? Jr. wrote: > > > David Hopkins wrote: >> On Tue, Apr 21, 2009 at 5:50 PM, "Terrell Prud? Jr." >> wrote: >> >>> I know you're at your wits' end here, and so is that teacher. This is not >>> good. >>> >>> Remind us what sound card is in these thin clients, would you? >>> >>> --TP >>> >> >> Actually, the systems are Dell GX150's with either 256Mb or 512Mb of >> memory.. http://support.dell.com/support/edocs/systems/opgx150/en/ug/specs.htm#audio >> >> Though, does that mean I might have the i810 video issue? For >> whatever reason, I thought the card was detected as an ati. I have to >> check this. >> > > That sound card (looks like it's built-in) says it has SoundBlaster > emulation, so when you were running it w/ K12LTSP 5EL, did you by > chance try turning on "SOUND=Y" in /opt/ltsp/i386/etc/lts.conf, and > also the setting from here? > > http://www.k12ltsp.org/mediawiki/index.php/Technical:Sound > > Something along these lines.... > > SMODULE_01 = "sb io=0x220 irq=5 dma=1" > > That sort of thing always worked for me with cards that supported SB > emulation, including (especially!) the SoundBlaster 16 and the > SoundBlaster AWE32/64 cards themselves. > > --TP Just did some more Googling. Looks like the "sb" driver won't work, so try changing the "sb" in that SMODULE_01 line to "i810_audio". If that doesn't work (let's hope that it does!), then here's why. You know about "Winmodems" and "Winprinters"? Well, it appears that the AD1885 sound chip is a "WinSound" chip that implements its brains in software like a Winmodem does. I don't know if that's been remedied in recent times, i. e. there's now support in Linux 2.6. Given that K12Linux's Fedora 10 kernel set supports these chips, the answer is "probably", but the RHEL/CentOS kernel is, remember, tried-and-tested v2.6.18. Here's a link explaining the situation with that chip. http://fixunix.com/ubuntu/555161-sound-can-anyone-explain-please.html One possible option would be to copy the newer LTSP kernel tree (it's in /tftpboot) from a K12Linux box over to a K12LTSP 5EL box and adjust your DHCP server's settings accordingly. It's not too hard if you watch your typos. That said, if you want sound to "Just Work," I'd seriously look into getting some Ensoniq AudioPCI cards, since these Dell GX150's appear to have only PCI slots (otherwise I'd suggest SoundBlaster 16 or AWE's). Creative bought Ensoniq a while back and marketed a slightly updated version as "Creative Ensoniq AudioPCI". All such cards work GREAT and should be pretty doggone cheap to pick up. They're what I use, with both K12LTSP 4.2EL and 5EL, and I love 'em. Here's a Google-search "shopping results" link. Note that it's a long link and may word-wrap. My search terms were as follows. "creative ensoniq" audiopci http://www.google.com/products?q=%22creative+ensoniq%22+audiopci&oe=UTF-8&um=1&ie=UTF-8&ei=lFHvSeerHY_EMtyXxfQP&sa=X&oi=product_result_group&resnum=11&ct=title Also check eBay. They're pretty cheap there as well. --TP -------------- next part -------------- An HTML attachment was scrubbed... URL: From dahopkins429 at gmail.com Wed Apr 22 17:41:02 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 22 Apr 2009 13:41:02 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49EF4E2C.4040105@cmosnetworks.com> References: <49EE3F9D.6080508@cmosnetworks.com> <49EF4E2C.4040105@cmosnetworks.com> Message-ID: On Wed, Apr 22, 2009 at 1:04 PM, "Terrell Prud? Jr." wrote: > > That sound card (looks like it's built-in) says it has SoundBlaster > emulation, so when you were running it w/ K12LTSP 5EL, did you by chance try > turning on "SOUND=Y" in /opt/ltsp/i386/etc/lts.conf, and also the setting > from here? > > ? http://www.k12ltsp.org/mediawiki/index.php/Technical:Sound > > Something along these lines.... > > ? SMODULE_01 = "sb io=0x220 irq=5 dma=1" > > That sort of thing always worked for me with cards that supported SB > emulation, including (especially!) the SoundBlaster 16 and the SoundBlaster > AWE32/64 cards themselves. Sound works at the client for some programs with LTSP4.2 but because some packages want esd and some alsa, and some use others, it is hit-or-miss whether a particular package will 'just work' without having to worry about twiddling with settings for the package. With LTSP5 it seems that sound 'just works' for all the apps the teacher wants to use. Also, videos stay in sync better. Sorry if my post implied that sound didn't work at all. It does work but LTSP5 seems to work better. Sincerely, Dave Hopkins From microman at cmosnetworks.com Wed Apr 22 17:44:15 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 22 Apr 2009 13:44:15 -0400 Subject: [K12OSN] client ip VS my ip during boot In-Reply-To: References: Message-ID: <49EF576F.5040406@cmosnetworks.com> Marc Fromm wrote: > > I have set the client IPs based on mac addresses in the dhcp file. > > When the client boots it reports CLIENT IP: the ip that I assigned. > > It then reports my IP seems to be: a different IP than what I assigned. > > > > Is the conflict between CLIENT IP and My IP seems to be the problem > why this client wont boot? > > The boot eventually fails with the message that is cannot find kernel > image: linux. > Hmm...it shouldn't matter what your thin client's IP address is, provided that you're getting an IP address on that subnet. Another possibility is that there's a second DHCP server on that subnet that isn't configured for the LTSP boot stuff. Are you sure you don't have any other DHCP servers on that subnet? Also, any way that you can give us the boot messages, even if you have to hand-type them in yourself, so we can see what's happening before the failure? Thx, --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From Marc.Fromm at wwu.edu Wed Apr 22 18:07:22 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Wed, 22 Apr 2009 11:07:22 -0700 Subject: [K12OSN] adding printer problem Message-ID: I was battleing a client non-booting problem for ages after upgraded Fedora Core 4 to 6. I created a new ltsp server using the K12LTSP-5.0.0EL-32bit-dvd.iso. My clients now boot fine. On the old working ltsp, I added printers using CUPs in the browser (localhost:631). The printers are connected to the terminals by USB. In CUPs I would: --click add printer --select Device AppSocket/HPJetDirect --create a DeviceURI as socket://finaid12:9100. Finaid12 is one of the clients as configured in the hosts file and dhcpd.conf file. I followed these same steps on the new Cent OS server. The clients see the printer as added through CUPs, but the print jobs just hang with the message processing. Is there something new to do to setup printers? Thanks Marc -------------- next part -------------- An HTML attachment was scrubbed... URL: From dahopkins429 at gmail.com Wed Apr 22 18:19:13 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 22 Apr 2009 14:19:13 -0400 Subject: [K12OSN] adding printer problem In-Reply-To: References: Message-ID: On Wed, Apr 22, 2009 at 2:07 PM, Marc Fromm wrote: > I was battleing a client non-booting problem for ages after upgraded Fedora > Core 4 to 6. > > I created a new ltsp server using the K12LTSP-5.0.0EL-32bit-dvd.iso. My > clients now boot fine. > > > > On the old working ltsp, I added printers using CUPs in the browser > (localhost:631). > > The printers are connected to the terminals by USB. > > In CUPs I would: > > --click add printer > > --select Device AppSocket/HPJetDirect > > --create a DeviceURI as socket://finaid12:9100. > > Finaid12 is one of the clients as configured in the hosts file and > dhcpd.conf file. > what happens if you use socket://x.y.w.z:9100 where x.y.w.z is the IP address of finaid12? I would sometimes have problems with name resolution (though not recently) and started using the IP address which has worked flawlessly for me. > > > I followed these same steps on the new Cent OS server. > > The clients see the printer as added through CUPs, but the print jobs just > hang with the message processing. > Can you ping finaid12? or telnet to finaid12 on port 9100? > > Is there something new to do to setup printers? > > > > Thanks > > > > Marc > > > Sincerely, Dave Hopkins From microman at cmosnetworks.com Wed Apr 22 18:55:10 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 22 Apr 2009 14:55:10 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <49EE3F9D.6080508@cmosnetworks.com> <49EF4E2C.4040105@cmosnetworks.com> Message-ID: <49EF680E.5090901@cmosnetworks.com> David Hopkins wrote: > On Wed, Apr 22, 2009 at 1:04 PM, "Terrell Prud? Jr." > wrote: > >> That sound card (looks like it's built-in) says it has SoundBlaster >> emulation, so when you were running it w/ K12LTSP 5EL, did you by chance try >> turning on "SOUND=Y" in /opt/ltsp/i386/etc/lts.conf, and also the setting >> from here? >> >> http://www.k12ltsp.org/mediawiki/index.php/Technical:Sound >> >> Something along these lines.... >> >> SMODULE_01 = "sb io=0x220 irq=5 dma=1" >> >> That sort of thing always worked for me with cards that supported SB >> emulation, including (especially!) the SoundBlaster 16 and the SoundBlaster >> AWE32/64 cards themselves. >> > > Sound works at the client for some programs with LTSP4.2 but because > some packages want esd and some alsa, and some use others, it is > hit-or-miss whether a particular package will 'just work' without > having to worry about twiddling with settings for the package. With > LTSP5 it seems that sound 'just works' for all the apps the teacher > wants to use. Also, videos stay in sync better. > > Sorry if my post implied that sound didn't work at all. It does work > but LTSP5 seems to work better. > Then you've gotta make a choice. Personally, mine'd be for the one that *overall* works better. If you're running into an Adobe Flash issue, then yup, that's Adobe. Apparently they decided not to support sound servers, only kernel modules directly. Another option might be to install K12LTSP 5EL without the stock LTSP 4.2 (but with the edu apps), and then install LTSP 5 from upstream www.ltsp.org. This way you get LTSP 5 with the stability of CentOS. I know that Warren might not like that idea, but you've got to get a system up and running and keep those teachers happy *now*. Otherwise, they'll whine about "it works with Windows!" and our cause will be lost. --TP -------------- next part -------------- An HTML attachment was scrubbed... URL: From microman at cmosnetworks.com Wed Apr 22 18:57:47 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 22 Apr 2009 14:57:47 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <49EE3F9D.6080508@cmosnetworks.com> <49EF4E2C.4040105@cmosnetworks.com> Message-ID: <49EF68AB.80308@cmosnetworks.com> David Hopkins wrote: > On Wed, Apr 22, 2009 at 1:04 PM, "Terrell Prud? Jr." > wrote: > >> That sound card (looks like it's built-in) says it has SoundBlaster >> emulation, so when you were running it w/ K12LTSP 5EL, did you by chance try >> turning on "SOUND=Y" in /opt/ltsp/i386/etc/lts.conf, and also the setting >> from here? >> >> http://www.k12ltsp.org/mediawiki/index.php/Technical:Sound >> >> Something along these lines.... >> >> SMODULE_01 = "sb io=0x220 irq=5 dma=1" >> >> That sort of thing always worked for me with cards that supported SB >> emulation, including (especially!) the SoundBlaster 16 and the SoundBlaster >> AWE32/64 cards themselves. >> > > Sound works at the client for some programs with LTSP4.2 but because > some packages want esd and some alsa, and some use others, it is > hit-or-miss whether a particular package will 'just work' without > having to worry about twiddling with settings for the package. With > LTSP5 it seems that sound 'just works' for all the apps the teacher > wants to use. Also, videos stay in sync better. > > Sorry if my post implied that sound didn't work at all. It does work > but LTSP5 seems to work better. > Hey, I just found something on the LTSP Web site that might help your situation. http://wiki.ltsp.org/twiki/bin/view/Ltsp/WorkInProgress#esd_ALSA_sound_on_LTSP_4_2 --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From Marc.Fromm at wwu.edu Wed Apr 22 20:52:37 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Wed, 22 Apr 2009 13:52:37 -0700 Subject: [K12OSN] adding printer problem In-Reply-To: References: Message-ID: I tried the ip address in place of finaid12 with no success. I can ping the client by both ip and name. -----Original Message----- From: k12osn-bounces at redhat.com [mailto:k12osn-bounces at redhat.com] On Behalf Of David Hopkins Sent: Wednesday, April 22, 2009 11:19 AM To: Support list for open source software in schools. Subject: Re: [K12OSN] adding printer problem On Wed, Apr 22, 2009 at 2:07 PM, Marc Fromm wrote: > I was battleing a client non-booting problem for ages after upgraded Fedora > Core 4 to 6. > > I created a new ltsp server using the K12LTSP-5.0.0EL-32bit-dvd.iso. My > clients now boot fine. > > > > On the old working ltsp, I added printers using CUPs in the browser > (localhost:631). > > The printers are connected to the terminals by USB. > > In CUPs I would: > > --click add printer > > --select Device AppSocket/HPJetDirect > > --create a DeviceURI as socket://finaid12:9100. > > Finaid12 is one of the clients as configured in the hosts file and > dhcpd.conf file. > what happens if you use socket://x.y.w.z:9100 where x.y.w.z is the IP address of finaid12? I would sometimes have problems with name resolution (though not recently) and started using the IP address which has worked flawlessly for me. > > > I followed these same steps on the new Cent OS server. > > The clients see the printer as added through CUPs, but the print jobs just > hang with the message processing. > Can you ping finaid12? or telnet to finaid12 on port 9100? > > Is there something new to do to setup printers? > > > > Thanks > > > > Marc > > > Sincerely, Dave Hopkins _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see From wtogami at redhat.com Wed Apr 22 21:57:20 2009 From: wtogami at redhat.com (Warren Togami) Date: Wed, 22 Apr 2009 17:57:20 -0400 Subject: [K12OSN] Re: More feedback on Fedora 10 + LTSP In-Reply-To: <49EF3A90.7010407@redhat.com> References: <6069A203FD01884885C037F81DD75080032AAC0D74@wsc-mail-01.intra.nwresd.k12.or.us> <49EF3A90.7010407@redhat.com> Message-ID: <49EF92C0.5070109@redhat.com> On 04/22/2009 11:41 AM, Warren Togami wrote: > On 04/22/2009 10:08 AM, Sean Harbour wrote: >> Also, some processes seem to hang around forever after a student logs >> out. After a week, the process list "ps aux | wc -l" balloons from >> 173 with one user after a reboot to over 1500 with one user logged >> in. > > What processes specifically are persisting beyond logout? /etc/xdg/autostart/pulseaudio.desktop pulseaudio daemon run by GNOME autostart from this file is not actually used by the thin client session. This pulseaudio runs /usr/libexec/pulse/gconf-helper, which can keep processes leftover after logout. In my tests however they seem to die within a minute of logout, but your logs indicate they sometimes don't. http://kojipkgs.fedoraproject.org/packages/pulseaudio/0.9.14/3.fc10/ I built a new pulseaudio for Fedora 10 that prevents pulseaudio daemon from running upon desktop login when it wouldn't be used due to PULSE_SERVER being set. This should improve it a little bit. Please upgrade this ON YOUR SERVER and let me know if it improves that part. But it seems there are other bugs of other processes not dying after logout. Please identify those and tell the list. Warren Togami wtogami at redhat.com From rowens at ptd.net Thu Apr 23 01:32:50 2009 From: rowens at ptd.net (Rob Owens) Date: Wed, 22 Apr 2009 21:32:50 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> Message-ID: <20090423013250.GC3774@aurora.owens.net> On Wed, Apr 22, 2009 at 08:48:18AM -0400, David Hopkins wrote: > > yep, there is a reason that CentOS/RHEL exists. I would prefer to > stay with CentOS but I have to resolve sound (talk about beating a > dead horse issue) and with LTSP5 I have to admit that sound 'just > works' in all the apps I have to have (except for Audacity but I've > found some documentation that implies it can be made to work). > > You could try glame. It looks to be similar to Audacity. -Rob From brcisna at eazylivin.net Thu Apr 23 02:48:24 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Wed, 22 Apr 2009 21:48:24 -0500 Subject: [K12OSN] client ip VS my ip during boot Message-ID: <1240454904.701.10.camel@localhost.localdomain> Marc, +1 on Terrell's post Plug a laptop into the switch that is feeding your TC's. If it is a winders laptop at a command prompt do: ipconfig /release .. then .. ipconfig /renew .. then.. ipconfig /all 5 times If it is a linux laptop in a terminal do: service network restart .. then .. ifconfig 5 times This will show you if your dhcp server(ipaddress) is ACTUALLY your k12ltsp server. Also in a terminal on the server do an: tail -f /var/log/messages ..then try to boot one of your TC's. and post the output of this here.This well tell what the bottleneck is. Take Care, Barry Cisna From brcisna at eazylivin.net Thu Apr 23 03:02:13 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Wed, 22 Apr 2009 22:02:13 -0500 Subject: [K12OSN] adding printer problem Message-ID: <1240455733.701.18.camel@localhost.localdomain> Marc, On the TC you have the usb printer attached to you will need to add an entry in your lts.conf file.If the TC is ws100 you will need: [ws100] PRINTER_0_DEVICE = /dev/usb/lp0 PRINTER_0_TYPE = U Restart the TC. AT bootup you should be able to see in the scrolling text an entry scroll by for the attached printer(look quick),,:) Now try re-adding your printer via CUPS as you did earlier. Take Care, Barry Cisna From carl at snarlnet.com Thu Apr 23 19:50:42 2009 From: carl at snarlnet.com (Carl Keil) Date: Thu, 23 Apr 2009 12:50:42 -0700 Subject: [K12OSN] Installing a 1TB Drive on Centos 5 Message-ID: <49F0C692.6000100@snarlnet.com> Hey People, Is there anything special I need to do to install and format a 1TB drive? I installed a 500Gig drive on an older Fedora installation a while ago and it wouldn't see the whole drive until I increased the block size, I'm hoping I don't have to do that on this server. Is a 1TB drive basically plug and play on 32 bit Centos 5. Also, how would you fdisk and format it? Would you do anything special filesystem-wise? I'm pretty new to >300 Gig drives. Thanks so much, ck From microman at cmosnetworks.com Thu Apr 23 20:14:58 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Thu, 23 Apr 2009 16:14:58 -0400 Subject: [K12OSN] Installing a 1TB Drive on Centos 5 In-Reply-To: <49F0C692.6000100@snarlnet.com> References: <49F0C692.6000100@snarlnet.com> Message-ID: <49F0CC42.5000605@cmosnetworks.com> Carl Keil wrote: > Hey People, > > Is there anything special I need to do to install and format a 1TB > drive? I installed a 500Gig drive on an older Fedora installation a > while ago and it wouldn't see the whole drive until I increased the > block size, I'm hoping I don't have to do that on this server. Is a > 1TB drive basically plug and play on 32 bit Centos 5. > > Also, how would you fdisk and format it? Would you do anything > special filesystem-wise? I'm pretty new to >300 Gig drives. > Thanks so much, > > ck Shouldn't be anything special necessary. I routinely do 1TB and 2TB partitions on RAID arrays with CentOS and RHEL. Your 1TB drive should be plug 'n' chug. Now, as to how to partition it, well, that depends on how you're going to use this drive on this system. Can you give us any details on that so we can best advise you? --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From carl at snarlnet.com Thu Apr 23 22:03:37 2009 From: carl at snarlnet.com (Carl Keil) Date: Thu, 23 Apr 2009 15:03:37 -0700 Subject: [K12OSN] Installing a 1TB Drive on Centos 5 Message-ID: <49F0E5B9.9080101@snarlnet.com> > Carl Keil wrote: >> Hey People, >> >> Is there anything special I need to do to install and format a 1TB >> drive? I installed a 500Gig drive on an older Fedora installation a >> while ago and it wouldn't see the whole drive until I increased the >> block size, I'm hoping I don't have to do that on this server. Is a >> 1TB drive basically plug and play on 32 bit Centos 5. >> >> Also, how would you fdisk and format it? Would you do anything >> special filesystem-wise? I'm pretty new to >300 Gig drives. >> Thanks so much, >> >> ck > > Shouldn't be anything special necessary. I routinely do 1TB and 2TB > partitions on RAID arrays with CentOS and RHEL. Your 1TB drive should > be plug 'n' chug. > > Now, as to how to partition it, well, that depends on how you're going > to use this drive on this system. Can you give us any details on that > so we can best advise you? > > --TP Well, I didn't want to get into specifics because this isn't exactly educational. I'm going to be using this drive as storage for MythTV recordings. So, mostly huge files bigger than 5Gigs each. BTW - Here's what I'm talking about. This doesn't look right to me. First I fdisk I typed n for new partition, I selected primary, I numbered it 1, then I took the defaults for first and end block. Then I hit "w" to write it to the partition table. Then I did the following. That looks like an awfully small partition to me. What did I do wrong? Thanks, [root at kitkat ~]# mkfs.ext3 /dev/sdb1 mke2fs 1.39 (29-May-2006) Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) 122109952 inodes, 244190000 blocks 12209500 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=0 7453 block groups 32768 blocks per group, 32768 fragments per group 16384 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968, 102400000, 214990848 Writing inode tables: done Creating journal (32768 blocks): done Writing superblocks and filesystem accounting information: done This filesystem will be automatically checked every 20 mounts or 180 days, whichever comes first. Use tune2fs -c or -i to override. [root at kitkat ~]# df /dev/sdb1 Filesystem 1K-blocks Used Available Use% Mounted on - 513468 124 513344 1% /dev From julius at turtle.com Thu Apr 23 23:17:54 2009 From: julius at turtle.com (Julius Szelagiewicz) Date: Thu, 23 Apr 2009 19:17:54 -0400 (EDT) Subject: [K12OSN] Installing a 1TB Drive on Centos 5 In-Reply-To: <49F0CC42.5000605@cmosnetworks.com> Message-ID: On Thu, 23 Apr 2009, [ISO-8859-1] "Terrell Prud? Jr." wrote: > Carl Keil wrote: > > Hey People, > > > > Is there anything special I need to do to install and format a 1TB > > drive? I installed a 500Gig drive on an older Fedora installation a > > while ago and it wouldn't see the whole drive until I increased the > > block size, I'm hoping I don't have to do that on this server. Is a > > 1TB drive basically plug and play on 32 bit Centos 5. > > > > Also, how would you fdisk and format it? Would you do anything > > special filesystem-wise? I'm pretty new to >300 Gig drives. > > Thanks so much, > > > > ck > > Shouldn't be anything special necessary. I routinely do 1TB and 2TB > partitions on RAID arrays with CentOS and RHEL. Your 1TB drive should > be plug 'n' chug. > > Now, as to how to partition it, well, that depends on how you're going > to use this drive on this system. Can you give us any details on that > so we can best advise you? > > --TP Terrel, The change in block size on older Fedora is not a red herring. The table was sized for 250GB with default block size. I've run into it head on at Fedora 3 and I still have a bruise :-) I am running a 4TB logical volumes with no problems on Centos 5 and it feels good. julius From rowens at ptd.net Fri Apr 24 00:28:18 2009 From: rowens at ptd.net (Rob Owens) Date: Thu, 23 Apr 2009 20:28:18 -0400 Subject: [K12OSN] Installing a 1TB Drive on Centos 5 In-Reply-To: <49F0E5B9.9080101@snarlnet.com> References: <49F0E5B9.9080101@snarlnet.com> Message-ID: <20090424002818.GC9085@aurora.owens.net> On Thu, Apr 23, 2009 at 03:03:37PM -0700, Carl Keil wrote: >> Carl Keil wrote: >>> Hey People, >>> >>> Is there anything special I need to do to install and format a 1TB >>> drive? I installed a 500Gig drive on an older Fedora installation a >>> while ago and it wouldn't see the whole drive until I increased the >>> block size, I'm hoping I don't have to do that on this server. Is a >>> 1TB drive basically plug and play on 32 bit Centos 5. >>> >>> Also, how would you fdisk and format it? Would you do anything >>> special filesystem-wise? I'm pretty new to >300 Gig drives. >>> Thanks so much, >>> >>> ck >> >> Shouldn't be anything special necessary. I routinely do 1TB and 2TB >> partitions on RAID arrays with CentOS and RHEL. Your 1TB drive should >> be plug 'n' chug. >> >> Now, as to how to partition it, well, that depends on how you're going >> to use this drive on this system. Can you give us any details on that >> so we can best advise you? >> >> --TP > > Well, I didn't want to get into specifics because this isn't exactly > educational. I'm going to be using this drive as storage for MythTV > recordings. So, mostly huge files bigger than 5Gigs each. > On my MythTV machine, ext3 takes quite a while to delete a file (due to the size). I use JFS and it deletes files really quick. XFS is supposed to be almost as fast. -Rob > BTW - Here's what I'm talking about. This doesn't look right to me. > First I fdisk I typed n for new partition, I selected primary, I > numbered it 1, then I took the defaults for first and end block. Then I > hit "w" to write it to the partition table. Then I did the following. > That looks like an awfully small partition to me. What did I do wrong? > > Thanks, > > [root at kitkat ~]# mkfs.ext3 /dev/sdb1 > mke2fs 1.39 (29-May-2006) > Filesystem label= > OS type: Linux > Block size=4096 (log=2) > Fragment size=4096 (log=2) > 122109952 inodes, 244190000 blocks > 12209500 blocks (5.00%) reserved for the super user > First data block=0 > Maximum filesystem blocks=0 > 7453 block groups > 32768 blocks per group, 32768 fragments per group > 16384 inodes per group > Superblock backups stored on blocks: > 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, > 2654208, > 4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968, > 102400000, 214990848 > > Writing inode tables: done Creating journal > (32768 blocks): done > Writing superblocks and filesystem accounting information: done > > This filesystem will be automatically checked every 20 mounts or > 180 days, whichever comes first. Use tune2fs -c or -i to override. > [root at kitkat ~]# df /dev/sdb1 > Filesystem 1K-blocks Used Available Use% Mounted on > - 513468 124 513344 1% /dev > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From microman at cmosnetworks.com Fri Apr 24 02:05:52 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Thu, 23 Apr 2009 22:05:52 -0400 Subject: [K12OSN] Installing a 1TB Drive on Centos 5 In-Reply-To: References: Message-ID: <49F11E80.7060603@cmosnetworks.com> _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! Julius Szelagiewicz wrote: > On Thu, 23 Apr 2009, [ISO-8859-1] "Terrell Prud? Jr." wrote: > > >> Carl Keil wrote: >> >>> Hey People, >>> >>> Is there anything special I need to do to install and format a 1TB >>> drive? I installed a 500Gig drive on an older Fedora installation a >>> while ago and it wouldn't see the whole drive until I increased the >>> block size, I'm hoping I don't have to do that on this server. Is a >>> 1TB drive basically plug and play on 32 bit Centos 5. >>> >>> Also, how would you fdisk and format it? Would you do anything >>> special filesystem-wise? I'm pretty new to >300 Gig drives. >>> Thanks so much, >>> >>> ck >>> >> Shouldn't be anything special necessary. I routinely do 1TB and 2TB >> partitions on RAID arrays with CentOS and RHEL. Your 1TB drive should >> be plug 'n' chug. >> >> Now, as to how to partition it, well, that depends on how you're going >> to use this drive on this system. Can you give us any details on that >> so we can best advise you? >> >> --TP >> > > Terrel, > The change in block size on older Fedora is not a red herring. The > table was sized for 250GB with default block size. I've run into it head > on at Fedora 3 and I still have a bruise :-) > I am running a 4TB logical volumes with no problems on Centos 5 > and it feels good. > julius > Oh, really? Thanks for that FC3 tip. I hadn't run into that one, probably because my K12LTSP demos back then were on 160GB disk drives. But indeed, RHEL/CentOS doesn't seem to have any issues. --TP -------------- next part -------------- An HTML attachment was scrubbed... URL: From microman at cmosnetworks.com Fri Apr 24 02:14:29 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Thu, 23 Apr 2009 22:14:29 -0400 Subject: [K12OSN] Installing a 1TB Drive on Centos 5 In-Reply-To: <49F0E5B9.9080101@snarlnet.com> References: <49F0E5B9.9080101@snarlnet.com> Message-ID: <49F12085.60901@cmosnetworks.com> Carl Keil wrote: >> Carl Keil wrote: >>> Hey People, >>> >>> Is there anything special I need to do to install and format a 1TB >>> drive? I installed a 500Gig drive on an older Fedora installation a >>> while ago and it wouldn't see the whole drive until I increased the >>> block size, I'm hoping I don't have to do that on this server. Is a >>> 1TB drive basically plug and play on 32 bit Centos 5. >>> >>> Also, how would you fdisk and format it? Would you do anything >>> special filesystem-wise? I'm pretty new to >300 Gig drives. >>> Thanks so much, >>> >>> ck >> >> Shouldn't be anything special necessary. I routinely do 1TB and 2TB >> partitions on RAID arrays with CentOS and RHEL. Your 1TB drive >> should be plug 'n' chug. >> >> Now, as to how to partition it, well, that depends on how you're >> going to use this drive on this system. Can you give us any details >> on that so we can best advise you? >> >> --TP > > Well, I didn't want to get into specifics because this isn't exactly > educational. I'm going to be using this drive as storage for MythTV > recordings. So, mostly huge files bigger than 5Gigs each. > > BTW - Here's what I'm talking about. This doesn't look right to me. > First I fdisk I typed n for new partition, I selected primary, I > numbered it 1, then I took the defaults for first and end block. Then > I hit "w" to write it to the partition table. Then I did the > following. That looks like an awfully small partition to me. What > did I do wrong? > > Thanks, > > [root at kitkat ~]# mkfs.ext3 /dev/sdb1 > mke2fs 1.39 (29-May-2006) > Filesystem label= > OS type: Linux > Block size=4096 (log=2) > Fragment size=4096 (log=2) > 122109952 inodes, 244190000 blocks > 12209500 blocks (5.00%) reserved for the super user > First data block=0 > Maximum filesystem blocks=0 > 7453 block groups > 32768 blocks per group, 32768 fragments per group > 16384 inodes per group > Superblock backups stored on blocks: > 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, > 2654208, > 4096000, 7962624, 11239424, 20480000, 23887872, 71663616, > 78675968, > 102400000, 214990848 > > Writing inode tables: done Creating journal > (32768 blocks): done > Writing superblocks and filesystem accounting information: done > > This filesystem will be automatically checked every 20 mounts or > 180 days, whichever comes first. Use tune2fs -c or -i to override. > [root at kitkat ~]# df /dev/sdb1 > Filesystem 1K-blocks Used Available Use% Mounted on > - 513468 124 513344 1% /dev Ah, that's becaues you're looking at the (unmounted) device node directly with df. Don't do that. Instead, mount it somewhere (e. g. /mnt/sdb1) and then run df against that mount point. Something like this.... su - root mkdir /MythTVShows mount -t ext3 /dev/sdb1 /MythTVShows df /MythTVShows Note that on all of my GNU/Linux boxes, once you mount the partition somewhere, then you can also run df against the device node itself (/dev/sdb1, in this case) and get correct numbers. --TP _______________________________ Do you GNU ? Microsoft Free since 2003 --the ultimate antivirus protection! -------------- next part -------------- An HTML attachment was scrubbed... URL: From carl at snarlnet.com Fri Apr 24 04:14:48 2009 From: carl at snarlnet.com (Carl Keil) Date: Thu, 23 Apr 2009 21:14:48 -0700 Subject: [K12OSN] Installing a 1TB Drive on Centos 5 Message-ID: <49F13CB8.3060700@snarlnet.com> TP Wrote: > Ah, that's becaues you're looking at the (unmounted) device node > directly with df. Don't do that. Instead, mount it somewhere (e. g. > /mnt/sdb1) and then run df against that mount point. Something like > this.... Oh Jeez. You're right. Thank you so much. From jkinney at localnetsolutions.com Fri Apr 24 04:20:35 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Fri, 24 Apr 2009 00:20:35 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> Message-ID: <1240546835.4535.14.camel@merlin.localnetsolutions.com> On Wed, 2009-04-22 at 08:48 -0400, David Hopkins wrote: > On Tue, Apr 21, 2009 at 4:54 PM, James P. Kinney III > wrote: > > The KERNEL is radically different, however. The CentOS version has a > > totally different scheduler than the F10 version. Plus a bazillion other > > changes that have a huge effect on LTSP kernel usage. Due to sound > > changes, it may be challenge to get the CentOS kernel src.rpm and > > compile it for the F10 OS environment but it is certainly worth a try. > > I would probably lose what hair I have left if I tried this right now. > I have 'spare' 8 processor and dual processor Dell servers that the > state was throwing away that I can learn on for this but that won't be > till summer (it is amazing what is sometimes considered obsolete). Cheat! Get the kernel src.rpm from the CentOS updates collection for 5.3 and just rpmbuild -ba the thing. I'm not 100% but I think it will compile fine on an F10 system. Install and reboot. > > > Note also that CentOS kernels are by default tuned more for big > > installation stuff than is the Fedora kernel. > > yep, there is a reason that CentOS/RHEL exists. I would prefer to > stay with CentOS but I have to resolve sound (talk about beating a > dead horse issue) and with LTSP5 I have to admit that sound 'just > works' in all the apps I have to have (except for Audacity but I've > found some documentation that implies it can be made to work). Load up the EPEL repo and install pulsaudio. That will let flash sound work again with flash > v9 . That change clobbered everybody. > > > > Again, I strongly suspect you are having a serious kernel issue and > > recommend compiling your own. > > It is possible that this is the case. I do know that I have to run > mkinitrd to get the kernel to boot properly since I have an Adaptec > 2010s controller. (I20 block device). There is an outstanding bug wrt > this raid card which will hopefully be resolved 'soon'. I have to > manually apply a patch/fix. I'll look at how mkinitrd really works > and see if that patch to make the raid card work isn't causing an > issue. However, I've tested disk I/O and I don't see any issues with > throughput. > > > The bridge interface is also a serious slowdown hog. It doesn't seem to > > support gigabit traffic at all. More like 230-400 Mbps. So the idea of > > ditching the bridge and going straight, physical nic is the best, > > fastest route to a speed up. Dig on the Brandon elementary pta site for > > a write up by William Fragakis on how to turn of the bridge process. > > I'll look for the notes on how to do this. The teacher's lesson plan > for the next few sessions involves using OpenOffice/StarOffice and > this works without any issues so I can flip the prior server back into > service if I break anything. That gives me a weekend in the worst > case to rebuild from scratch. I'll also say that I've tested > throughput on the external network-facing NIC and it is what I would > expect (it is the same as my systems using CentOS). I have tested the > bridge though. > > > Lastly: you mentioned hyperthreading cpu's. If they are _really_ > > hyperthreaded and not full multi-core, turn _off_ the hyperthreading. > > The F10 kernel threading will thrash royally doing stupid task swapping > > as it seems to not understand that the second cpu is a fake one. There > > is a kernel /proc flag to help this but I can't find it right now > > (again). > > They are hyperthreaded Xeons. They have performed very well until > recently. Though, how would I detect thrashing? A thumbnail check is to run top with details on the processors on (press "1" to see details on each cpu). If you see the load swapping from one to another cpu, you are looking at a poor performer. However to really get the details you need to run sysstat tools. psact will help to trace how things are being used within the system. But it takes some work to grok it all. > > > Also be sure to turn off EVERY process not actively used. Some are more > > of a drain on cpu throughput (avahi, a zillion python applets running > > desktop applets for notification - system monitor should be removed!! so > > students won't load it up - it's a HOG) > > I've chkconfig'ed to off everything I thought I could but I'll recheck > since I know I didn't turn off avahi. Is there a list of essential > services somewhere? Things like bluetooth and the like I know I don't > need on a production system but a nice list would be wonderful. That is very system dependent. Are you running NFSv4? They you need the myriad of rpc* daemons. If only v3, then the *gssd and rpcidmapd can be offed. > > Sincerely, > Dave Hopkins > Newark Charter School > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From Marc.Fromm at wwu.edu Fri Apr 24 21:10:31 2009 From: Marc.Fromm at wwu.edu (Marc Fromm) Date: Fri, 24 Apr 2009 14:10:31 -0700 Subject: [K12OSN] adding printer problem In-Reply-To: <1240455733.701.18.camel@localhost.localdomain> References: <1240455733.701.18.camel@localhost.localdomain> Message-ID: Thanks Barry, that was the step I missed. -----Original Message----- From: k12osn-bounces at redhat.com [mailto:k12osn-bounces at redhat.com] On Behalf Of Barry R Cisna Sent: Wednesday, April 22, 2009 8:02 PM To: K12LTSP Subject: Re: [K12OSN] adding printer problem Marc, On the TC you have the usb printer attached to you will need to add an entry in your lts.conf file.If the TC is ws100 you will need: [ws100] PRINTER_0_DEVICE = /dev/usb/lp0 PRINTER_0_TYPE = U Restart the TC. AT bootup you should be able to see in the scrolling text an entry scroll by for the attached printer(look quick),,:) Now try re-adding your printer via CUPS as you did earlier. Take Care, Barry Cisna _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see From dtrask at vcsvikings.org Sat Apr 25 17:41:18 2009 From: dtrask at vcsvikings.org (David Trask) Date: Sat, 25 Apr 2009 13:41:18 -0400 Subject: [K12OSN] In tough times...knowledge is power....FOSSed 2009 Register today! Message-ID: Good day everyone! By now either your school has or is about to wrap up Spring Break and you're ready to begin the final push toward summer. Sometimes we'd like to think that it's downhill from here, but in my experience, this is the hardest time of the year. Nonetheless, onward we go. As you are all well aware, times are tough. Not just for us personally, but for schools and companies as well. All over the world schools are faced with some tough decisions about budgets, spending, and how that impacts technology in our classrooms. Unfortunately when times are tough it becomes difficult to figure out how we're gonna' get through it all. Many schools are faced with possible layoffs; not buying any new hardware; trying to get yet another year out of that aging computer lab; sticking with that server for another year; weeding out the software budget to save money; and simply going without. Fortunately there are things you can do to help you deal with what lies ahead. LEARN Knowledge is power in times like these. Just as many folks go back to school to learn new skills when the economy deals them a blow...so can we in a manner of speaking. When things are going well it's easy to hire consultants, buy new hardware, or simply maintain the status quo....but when things get tight it's time to learn about new methods and alternatives. This applies not just to the techies, but to the classroom teachers as well. FOSSed is a great way to learn some of those skills, network and discuss options with others, and gather skills and knowledge to carry you further in the year to come. FOSSed is unlike most other conferences in that we offer 3 days of immersive hands-on learning combined with an "out-of-class" atmosphere at meals and in the evenings that is condusive to great conversations and even greater "informal" learning thorugh interaction with your colleagues. Personally, I've gotten some of my greatest ideas from these conversations and I'm happy to share as well. Professional development is a right and a necessity! This is why FOSSed was created in the first place....to help teachers and IT personnel learn more about using Free and Open Source Software and new online technologies in their schools and classrooms. We have lots of great sessions this year taught by folks just like you! Are you a techie? Boy do we have sessions for you! Everything ranging from reusing that old lab and turning it into a fast new thin-client lab to setting up a very powerful router/filter/proxy server....or even learning all about deployment with FOG. Are you a teacher...a tech integrator? Great! We have some fantastic sessions for you as well! High school and middle level folks can learn about teaching basic computer programming, using online tools to create personal learning networks, and so much more! Elementary teachers can learn all about teaching beginner computer skills, using online tools to replace expensive software, using tools to teach and interact with other classrooms around the state, country, and even the world! (Catch Deb White's awesome session on using Twitter in the classroom)....and a whole lot more. Every year teachers learn new and exciting things that enable them to "fire up" their classrooms and enhance learning without spending additional money. Everyone can learn more about Moodle, Mahara, Elgg, and many other technologies to help us out with distance/online learning, digital portfolios, and more. All these items are little or no cost and can greatly change the way you teach and manage IT in your schools. Professional development credits (CEU's) are available from the University of Southern Maine....there's a small fee (around $20) for the official transcript....you can get credit for attending FOSSed! FOSSed is very affordable. EVERYTHING is included! Rooms, meals, and a fantastic learning experience! You come, park the car, and enjoy 3 days of hassle-free learning and comraderie. $495 for on-campus and for those who live nearby...save a little at $455 (off-campus). Everyone who attends is invited to join us the second night for our legendary reception and banquet....all included in the cost! (thanks to generous sponsors) We're flexible! Need to register, but pay at the conference? No problem. Need to do something creative to pay for it from different funding sources? Gotta' wait until July 1st for the new fiscal year budget? Again...let us know and we'll do everything we can to help....we want YOU at FOSSed 2009. ACTEM members can take advantage of a great professional development benefit for up to $400. Don't feel guilty....that's what it's there for...and you're encouraged to use it! In fact, ACTEM is one of the generous sponsors of our banquet! Go here to learn more about the professional development benefit and how to take advantage of it....it's easy! http://www.actem.org/Pages/ACTEM_ProDev/index (also...we've been doing this for year...feel free to email me with any questions and I can help you out) Visit the FOSSed web site to learn more and to register today! Register soon! FOSSed is coming up in just a few weeks! http://fossed.com Check out the flyer....view the sessions....suggest a session....and much more! http://fossed.com Email me with any questions you may have...hope to see you at FOSSed this year! Register soon! (Even if you don't have all the financial details worked out yet...register...so we can get a better count for planning purposes) David Trask FOSSed 2009 copperdoggy at gmail.com (207) 446-2738 David N. Trask Technology Teacher/Director Vassalboro Community School dtrask at vcsvikings.org (207)923-3100 (207)923-4305 (direct) From wtogami at redhat.com Sat Apr 25 22:59:08 2009 From: wtogami at redhat.com (Warren Togami) Date: Sat, 25 Apr 2009 18:59:08 -0400 Subject: [K12OSN] In tough times...knowledge is power....FOSSed 2009 Register today! In-Reply-To: References: Message-ID: <49F395BC.5000602@redhat.com> On 04/25/2009 01:41 PM, David Trask wrote: > > Visit the FOSSed web site to learn more and to register today! Register > soon! FOSSed is coming up in just a few weeks! http://fossed.com > fossed.com goes to godaddy's parking page. From microman at cmosnetworks.com Sat Apr 25 23:11:25 2009 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Sat, 25 Apr 2009 19:11:25 -0400 Subject: [K12OSN] In tough times...knowledge is power....FOSSed 2009 Register today! In-Reply-To: <49F395BC.5000602@redhat.com> References: <49F395BC.5000602@redhat.com> Message-ID: <49F3989D.3040209@cmosnetworks.com> Warren Togami wrote: > On 04/25/2009 01:41 PM, David Trask wrote: >> >> Visit the FOSSed web site to learn more and to register today! Register >> soon! FOSSed is coming up in just a few weeks! http://fossed.com >> > > fossed.com goes to godaddy's parking page. Warren's right. The correct link is http://www.fossed.com. Note the "www" in the hyperlink. --TP From brcisna at eazylivin.net Sun Apr 26 02:25:24 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Sat, 25 Apr 2009 21:25:24 -0500 Subject: [K12OSN] OT: Samba log pw_buffer Message-ID: <1240712724.5987.9.camel@localhost.localdomain> Hello List, Centos5/K12ltsp This is actually Samba related,but thought I would see if anyone here has ever seen the following output in the logwatch nightly. This is the first time I have ever seen anything like this. What is a bit puzzling,is this occured on a Friday night when no one would have been using or accessing anything at school. Google'ing this does not throw any concise answers. --------------------- samba Begin ------------------------ **Unmatched Entries** libsmb/smbencrypt.c:decode_pw_buffer(552) decode_pw_buffer: incorrect password length (-1087115600). : 1 Time(s) libsmb/smbencrypt.c:decode_pw_buffer(552) decode_pw_buffer: incorrect password length (-208434967). : 1 Time(s) libsmb/smbencrypt.c:decode_pw_buffer(552) decode_pw_buffer: incorrect password length (1335557793). : 1 Time(s) libsmb/smbencrypt.c:decode_pw_buffer(552) decode_pw_buffer: incorrect password length (891857579). : 1 Time(s) libsmb/smbencrypt.c:decode_pw_buffer(553) decode_pw_buffer: check that 'encrypt passwords = yes' : 4 Time(s) ---------------------- samba End ------------------------- Thanks, Barry Cisna From william at fragakis.com Sun Apr 26 14:16:16 2009 From: william at fragakis.com (William Fragakis) Date: Sun, 26 Apr 2009 10:16:16 -0400 Subject: [K12OSN] adding printer problem In-Reply-To: <20090425160032.08F3A6196ED@hormel.redhat.com> References: <20090425160032.08F3A6196ED@hormel.redhat.com> Message-ID: <1240755376.24206.46.camel@server.ltsp> A note to others. This works fine in LTSP 4.2. When migrating installations to LTSP 5, I found that I needed to use PRINTER_0_TYPE = U PRINTER_0_DEVICE = /dev/usblp0 (no slash between usb and lp0) Hope that helps prevent some unwanted agony. Regards, William > > Content-Type: text/plain; charset="us-ascii" > > Thanks Barry, that was the step I missed. > > -----Original Message----- > From: k12osn-bounces at redhat.com [mailto:k12osn-bounces at redhat.com] On Behalf Of Barry R Cisna > Sent: Wednesday, April 22, 2009 8:02 PM > To: K12LTSP > Subject: Re: [K12OSN] adding printer problem > > Marc, > > On the TC you have the usb printer attached to you will need to add an > entry in your lts.conf file.If the TC is ws100 you will need: > > [ws100] > PRINTER_0_DEVICE = /dev/usb/lp0 > PRINTER_0_TYPE = U > > Restart the TC. AT bootup you should be able to see in the scrolling > text an entry scroll by for the attached printer(look quick),,:) > Now try re-adding your printer via CUPS as you did earlier. > > Take Care, > Barry Cisna > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > > > ------------------------------ > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > > End of K12OSN Digest, Vol 62, Issue 23 > ************************************** From dahopkins429 at gmail.com Mon Apr 27 15:29:50 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Mon, 27 Apr 2009 11:29:50 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <1240546835.4535.14.camel@merlin.localnetsolutions.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> Message-ID: This is my weekend update on what I did and found. First, I deleted the ltspbr0 bridge (brctl delbr ltspbr0 ). I then reconfigured the formerly slaved eth1 to have the correct IP/subnet info. After a quick reboot (tftp didn't want to work and I couldn't figure out which service to restart, xinetd wasn't it), all the terminals booted. Then, I started a few instances of Tuxmath while monitoring throughput at the switch. Bandwidth seems to max out at around 250Mbit/sec-350Mbit/sec (I have to estimate based on Rx Octets which seems to be around 30M+- octets/sec) Tuxmath crawls at this point and is unusable. If anyone knows of simple command line utility to monitor the throughput at the server, I'd be happy to use it. ethtool reports that the link is up and 1000Mb, full duplex. (mii-tool says it is 100Mbit but that tool is quite old). dmesg also shows the link up at 1000Mb as well. Now, I tested the throughput between the server and the file server using netio. This reports that I have a sustained throughput of 114MB/sec (or about appr 900Mbit/sec) between my servers on the network. Since I am using the same NIC and switches for the thin client side, I would expect about the same performance. I then reverted back to the original server, and noticed the same performance issue. I swapped out cables but that didn't change anything either. The load average on the LTSP server was around 1 with 6 clients running tuxmath. Memory was about 1Gb. So, my best guess is a network bandwidth issue. Tonight I'll connect my newest server to that switch and see if it performs better. Maybe it is just old hardware. Now, as for installing pulseaudio >> yep, there is a reason that CentOS/RHEL exists. ?I would prefer to >> stay with CentOS but I have to resolve sound (talk about beating a >> dead horse issue) and with LTSP5 I have to admit that sound 'just >> works' in all the apps I have to have (except for Audacity but I've >> found some documentation that implies it can be made to work). > > Load up the EPEL repo and install pulsaudio. That will let flash sound > work again with flash > v9 . That change clobbered everybody. I started to do this and noticed that to do so means uninstalling esound to avoid comflicts with the pulseaudio-esound-compat rpm. There are a few posts (found via google) which suggest there might be issues going this route. Since it was almost midnight, I decided to wait until later to see if this will work. I'll find out tonight. But, it does get back to the whole sound issue that I was hoping FC10 resolved: pulseaudio at the client. Still have to sort that out it seems. Hopefully esd at the client will be me far enough for the moment. :) During the summer I'll test other distributions and approaches. Sincerely, Dave Hopkins Newark Charter School Newark Delaware From rowens at ptd.net Mon Apr 27 17:03:53 2009 From: rowens at ptd.net (Rob Owens) Date: Mon, 27 Apr 2009 13:03:53 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> Message-ID: <20090427170353.GA9506@aurora.owens.net> On Mon, Apr 27, 2009 at 11:29:50AM -0400, David Hopkins wrote: > This is my weekend update on what I did and found. > > First, I deleted the ltspbr0 bridge (brctl delbr ltspbr0 ). I then > reconfigured the formerly slaved eth1 to have the correct IP/subnet > info. After a quick reboot (tftp didn't want to work and I couldn't > figure out which service to restart, xinetd wasn't it), all the > terminals booted. > > Then, I started a few instances of Tuxmath while monitoring throughput > at the switch. Bandwidth seems to max out at around > 250Mbit/sec-350Mbit/sec (I have to estimate based on Rx Octets which > seems to be around 30M+- octets/sec) Tuxmath crawls at this point and > is unusable. If anyone knows of simple command line utility to > monitor the throughput at the server, I'd be happy to use it. > Try running 'iftop' on the server. -Rob From lesmikesell at gmail.com Mon Apr 27 17:35:35 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Mon, 27 Apr 2009 12:35:35 -0500 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> Message-ID: <49F5ECE7.6060303@gmail.com> David Hopkins wrote: > > ethtool reports that the link is up and 1000Mb, full duplex. (mii-tool > says it is 100Mbit but that tool is quite old). dmesg also shows the > link up at 1000Mb as well. Is the switch managed so you can check there as well? You should see no errors in the server 'ifconfig' output or the switch interface statistics (or at most a couple if you have replugged the connections). If the switch isn't managed, it might have a link light indicator that shows full/half duplex. > Now, I tested the throughput between the server and the file server > using netio. This reports that I have a sustained throughput of > 114MB/sec (or about appr 900Mbit/sec) between my servers on the > network. Since I am using the same NIC and switches for the thin > client side, I would expect about the same performance. Are you testing this with an equal number of concurrent connections as where you see tuxtype/math performance fall - and across the same switch connections? Often a duplex mismatch or similar error degrades badly with concurrent operations but might not show up with a single connection test. > I then reverted back to the original server, and noticed the same > performance issue. I swapped out cables but that didn't change > anything either. > > The load average on the LTSP server was around 1 with 6 clients > running tuxmath. Memory was about 1Gb. So, my best guess is a network > bandwidth issue. Tonight I'll connect my newest server to that switch > and see if it performs better. Maybe it is just old hardware. "Old" Gb switch uplinks should work as well as new ones if the cabling and configuration is OK. -- Les Mikesell lesmikesell at gmail.com From dahopkins429 at gmail.com Mon Apr 27 17:51:13 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Mon, 27 Apr 2009 13:51:13 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F5ECE7.6060303@gmail.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F5ECE7.6060303@gmail.com> Message-ID: On Mon, Apr 27, 2009 at 1:35 PM, Les Mikesell wrote: > David Hopkins wrote: >> >> ethtool reports that the link is up and 1000Mb, full duplex. (mii-tool >> says it is 100Mbit but that tool is quite old). dmesg also shows the >> link up at 1000Mb as well. > > Is the switch managed so you can check there as well? ?You should see no > errors in the server 'ifconfig' output or the switch interface statistics > (or at most a couple if you have replugged the connections). If the switch > isn't managed, it might have a link light indicator that shows full/half > duplex. I tried both SGR24W (24 port Gigabit) and SR24G2W (24 port 10/100, 2 gigabit ports) from amer.com. They are unmanaged websmart switches. Looking at the systems via the web interface, neither show any network errors at all. In the case of the SR24G2W, it had been up 46 days without a single error. Also, ifconfig also shows no network errors. They report that all connections are full duplex with flow control enabled. >> Now, I tested the throughput between the server and the file server >> using netio. ?This reports that I have a sustained throughput of >> 114MB/sec (or about appr 900Mbit/sec) between my servers on the >> network. ?Since I am using the same NIC and switches for the thin >> client side, I would expect about the same performance. > > Are you testing this with an equal number of concurrent connections as where > you see tuxtype/math performance fall - and across the same switch > connections? ?Often a duplex mismatch or similar error degrades badly with > concurrent operations but might not show up with a single connection test. Yep ... same clients connected, same programs launched. I'll try the iftop (other post) and also ntop (completely forgot about it) this evening to see what that tells me. Results seem to be the same for both the CentOS K12LTSP and the FC10+LTSP5 systems. Hopefully, I'll get hard numbers tonight. >> I then reverted back to the original server, and noticed the same >> performance issue. I swapped out cables but that didn't change >> anything either. >> >> The load average on the LTSP server was around 1 with 6 clients >> running tuxmath. ?Memory was about 1Gb. So, my best guess is a network >> bandwidth issue. Tonight I'll connect my newest server to that switch >> and see if it performs better. Maybe it is just old hardware. > > "Old" Gb switch uplinks should work as well as new ones if the cabling and > configuration is OK. Very true, but since it is just moving a cable between two switches and then rebooting the clients to test, I might as well just to be sure. From wtogami at redhat.com Mon Apr 27 23:45:11 2009 From: wtogami at redhat.com (Warren Togami) Date: Mon, 27 Apr 2009 19:45:11 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> Message-ID: <49F64387.9080606@redhat.com> On 04/27/2009 11:29 AM, David Hopkins wrote: > This is my weekend update on what I did and found. > > First, I deleted the ltspbr0 bridge (brctl delbr ltspbr0 ). I then > reconfigured the formerly slaved eth1 to have the correct IP/subnet > info. After a quick reboot (tftp didn't want to work and I couldn't > figure out which service to restart, xinetd wasn't it), all the > terminals booted. Have you confirmed that changing from ltspbr0 to direct ethX makes a measurable difference in throughput? Warren Togami wtogami at redhat.com From wtogami at redhat.com Tue Apr 28 00:15:03 2009 From: wtogami at redhat.com (Warren Togami) Date: Mon, 27 Apr 2009 20:15:03 -0400 Subject: [K12OSN] Important: Regarding disabling ltspbr0 In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> Message-ID: <49F64A87.3060500@redhat.com> On 04/27/2009 11:29 AM, David Hopkins wrote: > This is my weekend update on what I did and found. > > First, I deleted the ltspbr0 bridge (brctl delbr ltspbr0 ). I then > reconfigured the formerly slaved eth1 to have the correct IP/subnet > info. After a quick reboot (tftp didn't want to work and I couldn't > figure out which service to restart, xinetd wasn't it), all the > terminals booted. https://fedorahosted.org/k12linux/wiki/AdvancedNetworkSetup I have wrote down the official procedure for disabling ltspbr0 here. Be warned that deleting ifcfg-ltspbr0 alone will cause you problems, because it will come back the next time you upgrade the ltsp-server package. At the moment I have not witnessed or seen hard numbers supporting the claim that getting rid of the bridge actually helps throughput. Could somebody please support this with real numbers? Warren Togami wtogami at redhat.com From dahopkins429 at gmail.com Tue Apr 28 01:23:29 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Mon, 27 Apr 2009 21:23:29 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F64387.9080606@redhat.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> Message-ID: On Mon, Apr 27, 2009 at 7:45 PM, Warren Togami wrote: > On 04/27/2009 11:29 AM, David Hopkins wrote: >> >> This is my weekend update on what I did and found. >> >> First, I deleted the ltspbr0 bridge (brctl delbr ltspbr0 ). ?I then >> reconfigured the formerly slaved eth1 to have the correct IP/subnet >> info. ?After a quick reboot (tftp didn't want to work and I couldn't >> figure out which service to restart, xinetd wasn't it), all the >> terminals booted. > > Have you confirmed that changing from ltspbr0 to direct ethX makes a > measurable difference in throughput? Actually I can't make a claim either way since right now I'm confused about just how much throughput I am getting without ltspbr0 even present. I install iftop as well as ntop and have tried launching tuxmath on several thin clients simultaneously. For iftop: The first thin client used appr 38Mb/s just to bring up the menu. When running, it was using appr 70Mb/s for the game. The second client likewise used appr 38Mb/s for the menu. However, the moment I selected an option from the menu, both instances dropped to around 56Mb/s. When I added a third client, this dropped again, into the mid 30Mb/s range. The total throughput stayed constant at around 110Mb/s. However, even at 3 clients, the game was not playable (it was very very slow). I continued adding clients and iftop continued to report that the max stayed around 110Mb/s but the bandwidth for each client dropped with each addition. At six clients, it was about 18Mb/s per client. However, the switch still showed that the link between it and the server was 1000Mb/s. So .. why is only about 1/10 of the bandwidth being used as far as iftop is concerned? ntop was a little different. The peak for Global Statistics, Traffic report stated 130Mb/s. However, looking at the Network Throughput, All Hosts page, the Peak value for a 10 second interval for the server stated 739Mb/s. if this is really 739Mb/(10sec interval) = 73.9, then it seems to agree with iftop (different runs), similarly for the hosts, they report values about a factor of 10 above what iftop reported for peak instant values. The load average on the server with 6 clients running tuxmath was 1.12. This is an 8 processor system (32bit Xeon MP cpu 2.8Ghz) with 4Gb of memory. top showed that there wasn't any swap being used. I ran a similar test on my newer server, and observed similar behavior. That server is completely different hardware (Supermicro MB, 64bit Intel Xeon 5410, dual dual-core system) So, why does it appear that bandwidth is an issue with the switch somehow (Amer.com switches). Or is there something else I can look at? Sincerely, Dave Hopkins From dahopkins429 at gmail.com Tue Apr 28 01:29:37 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Mon, 27 Apr 2009 21:29:37 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F64387.9080606@redhat.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> Message-ID: >> First, I deleted the ltspbr0 bridge (brctl delbr ltspbr0 ). ?I then >> reconfigured the formerly slaved eth1 to have the correct IP/subnet >> info. ?After a quick reboot (tftp didn't want to work and I couldn't >> figure out which service to restart, xinetd wasn't it), all the >> terminals booted. > > Have you confirmed that changing from ltspbr0 to direct ethX makes a > measurable difference in throughput? Warren, A quick note on the bridge. I will run some tests with and without ltspbr0 present once I can troubleshoot my current issue but I did not do so this time as I have to 'just try it' or else I'll end up without LTSP at all the way things are going at present. Of course, the way I removed the bridge means putting it back with brctrl addbr ltspbr0 (I think) and then configuring eth1 to be the slave again. And now that I have iftop/ntop, I'll hopefully get real numbers to look at. Sincerely, Dave Hopkins From wtogami at redhat.com Tue Apr 28 05:22:09 2009 From: wtogami at redhat.com (Warren Togami) Date: Tue, 28 Apr 2009 01:22:09 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> Message-ID: <49F69281.9000206@redhat.com> On 04/27/2009 09:23 PM, David Hopkins wrote: > Actually I can't make a claim either way since right now I'm confused > about just how much throughput I am getting without ltspbr0 even > present. I install iftop as well as ntop and have tried launching > tuxmath on several thin clients simultaneously. Where do you find tuxmath? It seems it isn't in Fedora 11 at least. > > For iftop: The first thin client used appr 38Mb/s just to bring up the > menu. When running, it was using appr 70Mb/s for the game. The second > client likewise used appr 38Mb/s for the menu. However, the moment I > selected an option from the menu, both instances dropped to around > 56Mb/s. When I added a third client, this dropped again, into the mid > 30Mb/s range. The total throughput stayed constant at around 110Mb/s. > However, even at 3 clients, the game was not playable (it was very > very slow). I continued adding clients and iftop continued to report > that the max stayed around 110Mb/s but the bandwidth for each client > dropped with each addition. At six clients, it was about 18Mb/s per > client. Does tuxmath normally use that insane amount of bandwidth? 70Mb/sec is comparable to 320x240 30fps Youtube video. Absolutely nothing other than video in my experience using anything near that amount of bandwidth. > > However, the switch still showed that the link between it and the > server was 1000Mb/s. So .. why is only about 1/10 of the bandwidth > being used as far as iftop is concerned? I have never used iftop before this, so I don't know how good it is. Warren From gianugo.altieri at tiscali.it Tue Apr 28 06:09:59 2009 From: gianugo.altieri at tiscali.it (Gianugo Altieri) Date: Tue, 28 Apr 2009 08:09:59 +0200 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> Message-ID: <49F69DB7.60003@tiscali.it> David Hopkins ha scritto: > ... > ntop was a little different. The peak for Global Statistics, Traffic > report stated 130Mb/s. However, looking at the Network Throughput, > All Hosts page, the Peak value for a 10 second interval for the > server stated 739Mb/s. if this is really 739Mb/(10sec interval) = > 73.9, then it seems to agree with iftop (different runs), similarly > for the hosts, they report values about a factor of 10 above what > iftop reported for peak instant values. > > The load average on the server with 6 clients running tuxmath was > 1.12. This is an 8 processor system (32bit Xeon MP cpu 2.8Ghz) with > 4Gb of memory. top showed that there wasn't any swap being used. > > David, since I'm not so expert, please clarify the BIT vs BYTE issue. Of course there is an 8 factor between them, which is not far from a 10 factor, and can change numbers completely. When speaking of bandwidth the numbers are usually expressed in mega bits per second (Mbps) and lower-case 'b' means bits. Upper-case 'B' means 'bytes'. I get confused when you speak of "4Gb" of memory. You meant "GB", didn't you? I wonder if in other places you meant 'B' for bytes but wrote 'b'. Please clarify. Thanks for your tests, which are interesting for us as well Gianugo Altieri From dahopkins429 at gmail.com Tue Apr 28 12:18:19 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 28 Apr 2009 08:18:19 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F69DB7.60003@tiscali.it> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69DB7.60003@tiscali.it> Message-ID: >> ntop was a little different. The peak for Global Statistics, Traffic >> report stated 130Mb/s. ?However, looking at the Network Throughput, >> All Hosts page, ?the Peak value for a 10 second interval for the >> server stated 739Mb/s. ?if this is really 739Mb/(10sec interval) = >> 73.9, then it seems to agree with iftop (different runs), similarly >> for the hosts, they report values about a factor of 10 above what >> iftop reported for peak instant values. >> >> The load average on the server with 6 clients running tuxmath was >> 1.12. ?This is an 8 processor system (32bit Xeon MP cpu 2.8Ghz) with >> 4Gb of memory. ?top showed that there wasn't any swap being used. >> >> > > ? David, > > ? since I'm not so expert, please clarify the BIT vs BYTE issue. Of course > there is an 8 factor between them, which is not far from a 10 factor, and > can change numbers completely. When speaking of bandwidth the numbers are > usually expressed in mega bits per second (Mbps) and lower-case 'b' means > bits. Upper-case 'B' means 'bytes'. I get confused when you speak of ?"4Gb" > of memory. You meant "GB", didn't you? I wonder if in other places you meant > 'B' for bytes but wrote 'b'. Please clarify. > My mistake ... should have typed 4GB. "b" means bits and I mistyped that one. The rest are correct as far as the reports are concerned. iftop and ntop both report Mb which I take to mean Mbps (Megabits/sec). Sincerely, Dave Hopkins From dahopkins429 at gmail.com Tue Apr 28 12:38:29 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 28 Apr 2009 08:38:29 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F69281.9000206@redhat.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> Message-ID: >On Tue, Apr 28, 2009 at 1:22 AM, Warren Togami wrote: > On 04/27/2009 09:23 PM, David Hopkins wrote: >> >> Actually I can't make a claim either way since right now I'm confused >> about just how much throughput I am getting without ltspbr0 even >> present. ?I install iftop as well as ntop and have tried launching >> tuxmath on several thin clients simultaneously. > > Where do you find tuxmath? ?It seems it isn't in Fedora 11 at least. > One location is here: http://sourceforge.net/project/showfiles.php?group_id=34443 There seems to be a newer version than what I am using (tuxmath-1.5.4-1.src.rpm) so I'll try upgrading to the latest version. It requires compilation but looks fairly straightforward. I'll try that tonight. >> For iftop: The first thin client used appr 38Mb/s just to bring up the >> menu. When running, it was using appr 70Mb/s for the game. ?The second >> client likewise used appr 38Mb/s for the menu. However, the moment I >> selected an option from the menu, both instances dropped to around >> 56Mb/s. ?When I added a third client, this dropped again, into the mid >> 30Mb/s range. ?The total throughput stayed constant at around 110Mb/s. >> ?However, even at 3 clients, the game was not playable (it was very >> very slow). ?I continued adding clients and iftop continued to report >> that the max stayed around 110Mb/s but the bandwidth for each client >> dropped with each addition. ?At six clients, it was about 18Mb/s per >> client. > > Does tuxmath normally use that insane amount of bandwidth? ?70Mb/sec is > comparable to 320x240 30fps Youtube video. ?Absolutely nothing other than > video in my experience using anything near that amount of bandwidth. In my experience, it has always been a huge bandwidth hog. >> However, the switch still showed that the link between it and the >> server was 1000Mb/s. ?So .. why is only about 1/10 of the bandwidth >> being used as far as iftop is concerned? > > I have never used iftop before this, so I don't know how good it is. I have to admit that I like iftop for spot checking. Now, if I can actually trust the numbers it would be excellent :) From lesmikesell at gmail.com Tue Apr 28 12:54:13 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 28 Apr 2009 07:54:13 -0500 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> Message-ID: <49F6FC75.2020602@gmail.com> David Hopkins wrote: > >> Does tuxmath normally use that insane amount of bandwidth? 70Mb/sec is >> comparable to 320x240 30fps Youtube video. Absolutely nothing other than >> video in my experience using anything near that amount of bandwidth. > > In my experience, it has always been a huge bandwidth hog. Yes, I think it just redraws the X frames as fast as it can to do the animation. >>> However, the switch still showed that the link between it and the >>> server was 1000Mb/s. So .. why is only about 1/10 of the bandwidth >>> being used as far as iftop is concerned? >> I have never used iftop before this, so I don't know how good it is. > > I have to admit that I like iftop for spot checking. Now, if I can > actually trust the numbers it would be excellent :) Didn't this start as a comparison with a better-working version of k12ltsp (except for sound)? While this sounds like a switch issue, if that same hardware worked better with different software revs, maybe you should focus on the specific differences. What about the client side of the network connections? Any errors showing on ifconfig or the switch ports there? -- Les Mikesell lesmikesell at gmail.com From wtogami at redhat.com Tue Apr 28 13:13:12 2009 From: wtogami at redhat.com (Warren Togami) Date: Tue, 28 Apr 2009 09:13:12 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F6FC75.2020602@gmail.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> Message-ID: <49F700E8.5010808@redhat.com> On 04/28/2009 08:54 AM, Les Mikesell wrote: > David Hopkins wrote: >> >>> Does tuxmath normally use that insane amount of bandwidth? 70Mb/sec is >>> comparable to 320x240 30fps Youtube video. Absolutely nothing other than >>> video in my experience using anything near that amount of bandwidth. >> >> In my experience, it has always been a huge bandwidth hog. > > Yes, I think it just redraws the X frames as fast as it can to do the > animation. I recall tuxmath several years ago, tuxmath did not draw frames this rapidly. This sounds like a new bug. Even with tuxmath causing terrible bandwidth issues, this shouldn't be causing problems like you are describing. Our test of 10 clients with several users using Youtube on a gigabit switch wasn't this bad. There sounds to be something else going on like bad switches. Did you also test the cabling with an ethernet tester? http://fedoraproject.org/wiki/PackageMaintainers/Join Also, could someone please follow this procedure to become a package maintainer and add tuxmath to Fedora? Warren From microman at cmosnetworks.com Tue Apr 28 13:35:55 2009 From: microman at cmosnetworks.com (Terrell Prude' Jr.) Date: Tue, 28 Apr 2009 09:35:55 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> Message-ID: <49F7063B.1080302@cmosnetworks.com> David Hopkins wrote: > >> Does tuxmath normally use that insane amount of bandwidth? 70Mb/sec is >> comparable to 320x240 30fps Youtube video. Absolutely nothing other than >> video in my experience using anything near that amount of bandwidth. >> > > In my experience, it has always been a huge bandwidth hog. > Yup, it does! I've measured 73Mb/sec on my Cisco Catalyst switch with TuxType and similar numbers with TuxMath. Got similar numbers from the ifconfig and a stopwatch. That's why you need 100Mbps/FullDuplex on your clients, Gig-E on your server, and a clean, error-free network. X11 is a wonderful protocol that allows LTSP to exist in the first place, but heavy screen updates do eat up bandwidth. --TP From lesmikesell at gmail.com Tue Apr 28 13:49:13 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 28 Apr 2009 08:49:13 -0500 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F7063B.1080302@cmosnetworks.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F7063B.1080302@cmosnetworks.com> Message-ID: <49F70959.8070906@gmail.com> Terrell Prude' Jr. wrote: >>> Does tuxmath normally use that insane amount of bandwidth? 70Mb/sec is >>> comparable to 320x240 30fps Youtube video. Absolutely nothing other >>> than >>> video in my experience using anything near that amount of bandwidth. >>> >> >> In my experience, it has always been a huge bandwidth hog. >> > > Yup, it does! I've measured 73Mb/sec on my Cisco Catalyst switch with > TuxType and similar numbers with TuxMath. Got similar numbers from the > ifconfig and a stopwatch. That's why you need 100Mbps/FullDuplex on > your clients, Gig-E on your server, and a clean, error-free network. > > X11 is a wonderful protocol that allows LTSP to exist in the first > place, but heavy screen updates do eat up bandwidth. I think X has more efficient ways to make things move on the screen, but tuxmath doesn't use them. -- Les Mikesell lesmikesell at gmail.com From dahopkins429 at gmail.com Tue Apr 28 14:03:52 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 28 Apr 2009 10:03:52 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F6FC75.2020602@gmail.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> Message-ID: > Didn't this start as a comparison with a better-working version of k12ltsp > (except for sound)? ?While this sounds like a switch issue, if that same > hardware worked better with different software revs, maybe you should focus > on the specific differences. ?What about the client side of the network > connections? ?Any errors showing on ifconfig or the switch ports there? > The clients aren't showing any errors at all (for LTSP5, not sure how to check at LTSP4.2 except to monitor what the switch is saying and the switch shows no errors. I've tried two different switches at this point though both are Amer.com. I've also swapped cables (which are rated at Cat5e). The difference in the systems is CentOS5.2+LTSP4.2 vs FC10+LTSP5. From dahopkins429 at gmail.com Tue Apr 28 14:17:54 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 28 Apr 2009 10:17:54 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F7063B.1080302@cmosnetworks.com> References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F7063B.1080302@cmosnetworks.com> Message-ID: >On Tue, Apr 28, 2009 at 9:35 AM, Terrell Prude' Jr. wrote: >> David Hopkins wrote: >> >>> Does tuxmath normally use that insane amount of bandwidth? ?70Mb/sec is >>> comparable to 320x240 30fps Youtube video. ?Absolutely nothing other than >>> video in my experience using anything near that amount of bandwidth. >>> >> >> In my experience, it has always been a huge bandwidth hog. >> > > Yup, it does! ?I've measured 73Mb/sec on my Cisco Catalyst switch with > TuxType and similar numbers with TuxMath. ?Got similar numbers from the > ifconfig and a stopwatch. ?That's why you need 100Mbps/FullDuplex on your > clients, Gig-E on your server, and a clean, error-free network. Which I have (or at least think that I have based on not finding errors reported). It just seems that the system isn't using the Gig-E properly for some reason. I'll run NetIO (http://www.ars.de/ars/ars.nsf/docs/netio) tonight as well and push some data around the network while I let iftop and ntop monitor the NICs. I know that NetIO reports 114MB/s transfer speeds (912Mbps) sustained between the file servers and my LTSP servers. Since that is through the same types of switches with the same NICs, it at least gives me a baseline to see what iftop/ntop think is happening. And, this topic is drifting from the feedback on LTSP5, but I'd rather keep the chain together since if there is a resolution, it would be nice to have it all together. Sincerely, Dave Hopkins From lesmikesell at gmail.com Tue Apr 28 15:16:22 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 28 Apr 2009 10:16:22 -0500 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240347249.13306.16.camel@merlin.localnetsolutions.com> <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> Message-ID: <49F71DC6.20901@gmail.com> David Hopkins wrote: >> Didn't this start as a comparison with a better-working version of k12ltsp >> (except for sound)? While this sounds like a switch issue, if that same >> hardware worked better with different software revs, maybe you should focus >> on the specific differences. What about the client side of the network >> connections? Any errors showing on ifconfig or the switch ports there? >> > > The clients aren't showing any errors at all (for LTSP5, not sure how > to check at LTSP4.2 except to monitor what the switch is saying and > the switch shows no errors. I've tried two different switches at this > point though both are Amer.com. I've also swapped cables (which are > rated at Cat5e). The difference in the systems is CentOS5.2+LTSP4.2 > vs FC10+LTSP5. With k12ltsp you should be able to enable ssh on the clients so you could connect and run ifconfig. Did you measure the server NIC throughput under CentOS5.2+LTSP4.2 to see if there is a large difference? I think you mentioned the switch having flow control enabled. Can you turn it off? http://www.smallnetbuilder.com/index2.php?option=com_content&task=view&id=30212&pop=1&page=0&Itemid=54 -- Les Mikesell lesmikesell at gmail.com From fyrteach at hotmail.com Tue Apr 28 18:35:20 2009 From: fyrteach at hotmail.com (Jeff Bonton) Date: Tue, 28 Apr 2009 13:35:20 -0500 Subject: [K12OSN] Strange edubuntu Behavior Message-ID: I have a strange issue with an Edubuntu server that is running 9.04 in a school. I have been reading this list for some time and the level of knowledge is amazing. I am hoping that someone has experienced this problem and found a fix, or has an idea of what to look for. I am a noob compared to the skills I see on this list. I am thinking it has something to do with permissions, but that is only a guess When I go to log into the server it asks for my username once, but my password twice. If I want to launch an application that requires root access, ie: synaptic, it prompts for my password then does nothing. I can launch the application from the CLI but it also asks for my password twice. About the same time, I cannot remember if it appeared at exactly the same time, the clients started refusing to login. I enter the username and password, the screen says verifying password for about a minute and then it says that the server is not responding and goes to a black screen. Any help or ideas will be appreciated. Jeff Bonton _________________________________________________________________ Rediscover Hotmail?: Get quick friend updates right in your inbox. http://windowslive.com/RediscoverHotmail?ocid=TXT_TAGLM_WL_HM_Rediscover_Updates2_042009 -------------- next part -------------- An HTML attachment was scrubbed... URL: From dtrask at vcsvikings.org Tue Apr 28 20:05:11 2009 From: dtrask at vcsvikings.org (David Trask) Date: Tue, 28 Apr 2009 16:05:11 -0400 Subject: [K12OSN] (fixed link) In tough times knowledge is power....FOSSed 2009...register today! Message-ID: Oops! fossed.com simply goes to a GoDaddy parking page...the link SHOULD have been [Link]http://www.fossed.com Sorry! ;-) [Link]http://www.fossed.com Good day everyone! By now either your school has or is about to wrap up Spring Break and you're ready to begin the final push toward summer. Sometimes we'd like to think that it's downhill from here, but in my experience, this is the hardest time of the year. Nonetheless, onward we go. As you are all well aware, times are tough. Not just for us personally, but for schools and companies as well. All over the world schools are faced with some tough decisions about budgets, spending, and how that impacts technology in our classrooms. Unfortunately when times are tough it becomes difficult to figure out how we're gonna' get through it all. Many schools are faced with possible layoffs; not buying any new hardware; trying to get yet another year out of that aging computer lab; sticking with that server for another year; weeding out the software budget to save money; and simply going without. Fortunately there are things you can do to help you deal with what lies ahead. LEARN Knowledge is power in times like these. Just as many folks go back to school to learn new skills when the economy deals them a blow...so can we in a manner of speaking. When things are going well it's easy to hire consultants, buy new hardware, or simply maintain the status quo....but when things get tight it's time to learn about new methods and alternatives. This applies not just to the techies, but to the classroom teachers as well. FOSSed is a great way to learn some of those skills, network and discuss options with others, and gather skills and knowledge to carry you further in the year to come. FOSSed is unlike most other conferences in that we offer 3 days of immersive hands-on learning combined with an "out-of-class" atmosphere at meals and in the evenings that is condusive to great conversations and even greater "informal" learning thorugh interaction with your colleagues. Personally, I've gotten some of my greatest ideas from these conversations and I'm happy to share as well. Professional development is a right and a necessity! This is why FOSSed was created in the first place....to help teachers and IT personnel learn more about using Free and Open Source Software and new online technologies in their schools and classrooms. We have lots of great sessions this year taught by folks just like you! Are you a techie? Boy do we have sessions for you! Everything ranging from reusing that old lab and turning it into a fast new thin-client lab to setting up a very powerful router/filter/proxy server....or even learning all about deployment with FOG. Are you a teacher...a tech integrator? Great! We have some fantastic sessions for you as well! High school and middle level folks can learn about teaching basic computer programming, using online tools to create personal learning networks, and so much more! Elementary teachers can learn all about teaching beginner computer skills, using online tools to replace expensive software, using tools to teach and interact with other classrooms around the state, country, and even the world! (Catch Deb White's awesome session on using Twitter in the classroom)....and a whole lot more. Every year teachers learn new and exciting things that enable them to "fire up" their classrooms and enhance learning without spending additional money. Everyone can learn more about Moodle, Mahara, Elgg, and many other technologies to help us out with distance/online learning, digital portfolios, and more. All these items are little or no cost and can greatly change the way you teach and manage IT in your schools. Professional development credits (CEU's) are available from the University of Southern Maine....there's a small fee (around $20) for the official transcript....you can get credit for attending FOSSed! FOSSed is very affordable. EVERYTHING is included! Rooms, meals, and a fantastic learning experience! You come, park the car, and enjoy 3 days of hassle-free learning and comraderie. $495 for on-campus and for those who live nearby...save a little at $455 (off-campus). Everyone who attends is invited to join us the second night for our legendary reception and banquet....all included in the cost! (thanks to generous sponsors) We're flexible! Need to register, but pay at the conference? No problem. Need to do something creative to pay for it from different funding sources? Gotta' wait until July 1st for the new fiscal year budget? Again...let us know and we'll do everything we can to help....we want YOU at FOSSed 2009. ACTEM members can take advantage of a great professional development benefit for up to $400. Don't feel guilty....that's what it's there for...and you're encouraged to use it! In fact, ACTEM is one of the generous sponsors of our banquet! Go here to learn more about the professional development benefit and how to take advantage of it....it's easy! [ http://www.actem.org/Pages/ACTEM_ProDev/index ]http://www.actem.org/Pages/ACTEM_ProDev/index (also...we've been doing this for year...feel free to email me with any questions and I can help you out) Visit the FOSSed web site to learn more and to register today! Register soon! FOSSed is coming up in just a few weeks! [ http://www.fossed.com ]http://www.fossed.com Check out the flyer....view the sessions....suggest a session....and much more! [ http://www.fossed.com ]http://www.fossed.com Email me with any questions you may have...hope to see you at FOSSed this year! Register soon! (Even if you don't have all the financial details worked out yet...register...so we can get a better count for planning purposes) David Trask FOSSed 2009 copperdoggy at gmail.com (207) 446-2738 David N. Trask Technology Teacher/Director Vassalboro Community School dtrask at vcsvikings.org (207)923-3100 (207)923-4305 (direct) _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see David N. Trask Technology Teacher/Director Vassalboro Community School dtrask at vcsvikings.org (207)923-3100 (207)923-4305 (direct) From news at siddall.name Wed Apr 29 00:24:11 2009 From: news at siddall.name (Jeff Siddall) Date: Tue, 28 Apr 2009 20:24:11 -0400 Subject: [K12OSN] Strange edubuntu Behavior In-Reply-To: References: Message-ID: <49F79E2B.6040707@siddall.name> Jeff Bonton wrote: > I have a strange issue with an Edubuntu server that is running 9.04 in a > school. I have been reading this list for some time and the level of > knowledge is amazing. I am hoping that someone has experienced this > problem and found a fix, or has an idea of what to look for. I am a noob > compared to the skills I see on this list. > > I am thinking it has something to do with permissions, but that is only > a guess > > When I go to log into the server it asks for my username once, but my > password twice. If I want to launch an application that requires root > access, ie: synaptic, it prompts for my password then does nothing. I > can launch the application from the CLI but it also asks for my password > twice. > > About the same time, I cannot remember if it appeared at exactly the > same time, the clients started refusing to login. I enter the username > and password, the screen says verifying password for about a minute and > then it says that the server is not responding and goes to a black screen. > > Any help or ideas will be appreciated. > > Jeff Bonton I had some strange problems like "server not responding" until I ran ltsp-update-sshkeys. In my case the cause was a server IP change so this may not help you. Can't hurt to try though. Jeff From dahopkins429 at gmail.com Wed Apr 29 01:59:04 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 28 Apr 2009 21:59:04 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F71DC6.20901@gmail.com> References: <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F71DC6.20901@gmail.com> Message-ID: > With k12ltsp you should be able to enable ssh on the clients so you could > connect and run ifconfig. ? Did you measure the server NIC throughput under > CentOS5.2+LTSP4.2 to see if there is a large difference? ?I think you > mentioned the switch having flow control enabled. ?Can you turn it off? > http://www.smallnetbuilder.com/index2.php?option=com_content&task=view&id=30212&pop=1&page=0&Itemid=54 > Thanks!! The switches had flow control turned on which apparently was limiting the Gig-E link to appr 100Mbps. I turned it off and launched 21 sessions of Tuxmath. It wasn't playable but once I backed it down to around 8-10 sessions, it was tolerable. However, tracking the throughput on the switch as I backed it down, the max I ever saw was 400Mbps and it stayed closer to 300Mbps. Load on the server never exceeded 8 (on an 8cpu system). So, while a lot better, something is still limiting the throughput. I thought it might be packet size related, so I ran NetIO from this system to another on a Gig-E link with the folllowing typical results: TCP connection established. Packet size 1k bytes: 82750 KByte/s Tx, 112086 KByte/s Rx. Packet size 2k bytes: 87499 KByte/s Tx, 113424 KByte/s Rx. Packet size 4k bytes: 95495 KByte/s Tx, 112055 KByte/s Rx. Packet size 8k bytes: 94037 KByte/s Tx, 114064 KByte/s Rx. Packet size 16k bytes: 102218 KByte/s Tx, 114743 KByte/s Rx. Packet size 32k bytes: 104406 KByte/s Tx, 114689 KByte/s Rx. These are representative (I ran the test 10 times). So, I'm not totally sure what is up with the Tx numbers. Still, thanks to everyone ... I at least have a good idea of where to look now. And, worst case, I can get a card and just use channel bonding to get the throughput up if I have to with some hope that it will actually work. Sincerely, Dave Hopkins Newark Charter School From dahopkins429 at gmail.com Wed Apr 29 02:00:41 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 28 Apr 2009 22:00:41 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F700E8.5010808@redhat.com> References: <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F700E8.5010808@redhat.com> Message-ID: > Did you also test the cabling with an ethernet tester? I have not tested them but they supposedly are Cat5e, that is what the label on the package said ;) > > http://fedoraproject.org/wiki/PackageMaintainers/Join > Also, could someone please follow this procedure to become a package > maintainer and add tuxmath to Fedora? Search for Tuxmath and Fedora 10 at this site: http://software.opensuse.org/search It comes back with 4 hits, with DBruce as the submitter? So, does this imply it already is being maintained? Sincerely, Dave Hopkins From jkinney at localnetsolutions.com Wed Apr 29 02:11:38 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Tue, 28 Apr 2009 22:11:38 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F71DC6.20901@gmail.com> Message-ID: <1240971098.10761.8.camel@merlin.localnetsolutions.com> David, As wonderful as tuxmath is (it's a great game/tool) is has a design problem that makes it unsuitable for LTSP. The screen draw rate is clocked by wait cycles as dummy loops. The system literally polls "is it time to send a screen update now". While the load is minimal, it pushes the next cycle onto a timer. So the system get dog slow and the load is not high yet the client end is uselessly slow. The loop timer is calculated once at startup and can't change as the load the goes up. The _entire_ game loop depends on this timing cycle. The heavier the server load the more out of sync this clocking becomes. Short of a rewrite I was never able to do more than a few tweaks to it. Because of the way multicore systems handle thread swaps, tuxmath actually makes a killer test on cpu efficiency. On Tue, 2009-04-28 at 21:59 -0400, David Hopkins wrote: > > With k12ltsp you should be able to enable ssh on the clients so you could > > connect and run ifconfig. Did you measure the server NIC throughput under > > CentOS5.2+LTSP4.2 to see if there is a large difference? I think you > > mentioned the switch having flow control enabled. Can you turn it off? > > http://www.smallnetbuilder.com/index2.php?option=com_content&task=view&id=30212&pop=1&page=0&Itemid=54 > > > > Thanks!! > > The switches had flow control turned on which apparently was limiting > the Gig-E link to appr 100Mbps. I turned it off and launched 21 > sessions of Tuxmath. It wasn't playable but once I backed it down to > around 8-10 sessions, it was tolerable. However, tracking the > throughput on the switch as I backed it down, the max I ever saw was > 400Mbps and it stayed closer to 300Mbps. Load on the server never > exceeded 8 (on an 8cpu system). So, while a lot better, something is > still limiting the throughput. I thought it might be packet size > related, so I ran NetIO from this system to another on a Gig-E link > with the folllowing typical results: > > TCP connection established. > Packet size 1k bytes: 82750 KByte/s Tx, 112086 KByte/s Rx. > Packet size 2k bytes: 87499 KByte/s Tx, 113424 KByte/s Rx. > Packet size 4k bytes: 95495 KByte/s Tx, 112055 KByte/s Rx. > Packet size 8k bytes: 94037 KByte/s Tx, 114064 KByte/s Rx. > Packet size 16k bytes: 102218 KByte/s Tx, 114743 KByte/s Rx. > Packet size 32k bytes: 104406 KByte/s Tx, 114689 KByte/s Rx. > > These are representative (I ran the test 10 times). So, I'm not > totally sure what is up with the Tx numbers. > > Still, thanks to everyone ... I at least have a good idea of where to > look now. And, worst case, I can get a card and just use channel > bonding to get the throughput up if I have to with some hope that it > will actually work. > > Sincerely, > Dave Hopkins > Newark Charter School > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From wtogami at redhat.com Wed Apr 29 02:28:29 2009 From: wtogami at redhat.com (Warren Togami) Date: Tue, 28 Apr 2009 22:28:29 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F700E8.5010808@redhat.com> Message-ID: <49F7BB4D.6090205@redhat.com> On 04/28/2009 10:00 PM, David Hopkins wrote: > Search for Tuxmath and Fedora 10 at this site: > http://software.opensuse.org/search > > It comes back with 4 hits, with DBruce as the submitter? So, does > this imply it already is being maintained? No, builds at other sites are 3rd party software. Nothing wrong with that, but it is inconvenient for users and you don't automatically get updates. It should be maintained directly in Fedora. Warren From wtogami at redhat.com Wed Apr 29 02:30:04 2009 From: wtogami at redhat.com (Warren Togami) Date: Tue, 28 Apr 2009 22:30:04 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <1240971098.10761.8.camel@merlin.localnetsolutions.com> References: <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F71DC6.20901@gmail.com> <1240971098.10761.8.camel@merlin.localnetsolutions.com> Message-ID: <49F7BBAC.3010600@redhat.com> On 04/28/2009 10:11 PM, James P. Kinney III wrote: > David, > > As wonderful as tuxmath is (it's a great game/tool) is has a design > problem that makes it unsuitable for LTSP. The screen draw rate is > clocked by wait cycles as dummy loops. The system literally polls "is it > time to send a screen update now". While the load is minimal, it pushes > the next cycle onto a timer. So the system get dog slow and the load is > not high yet the client end is uselessly slow. The loop timer is > calculated once at startup and can't change as the load the goes up. > > The _entire_ game loop depends on this timing cycle. The heavier the > server load the more out of sync this clocking becomes. Short of a > rewrite I was never able to do more than a few tweaks to it. Because of > the way multicore systems handle thread swaps, tuxmath actually makes a > killer test on cpu efficiency. This is an accurate assessment. Tuxmath and tuxtype are unusable for LTSP. It is just poorly designed. It could possibly be fixed by a programmer who understands this stuff. Warren From jkinney at localnetsolutions.com Wed Apr 29 02:41:59 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Tue, 28 Apr 2009 22:41:59 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F7BBAC.3010600@redhat.com> References: <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F71DC6.20901@gmail.com> <1240971098.10761.8.camel@merlin.localnetsolutions.com> <49F7BBAC.3010600@redhat.com> Message-ID: <1240972919.12045.4.camel@merlin.localnetsolutions.com> On Tue, 2009-04-28 at 22:30 -0400, Warren Togami wrote: > On 04/28/2009 10:11 PM, James P. Kinney III wrote: > > David, > > > > As wonderful as tuxmath is (it's a great game/tool) is has a design > > problem that makes it unsuitable for LTSP. The screen draw rate is > > clocked by wait cycles as dummy loops. The system literally polls "is it > > time to send a screen update now". While the load is minimal, it pushes > > the next cycle onto a timer. So the system get dog slow and the load is > > not high yet the client end is uselessly slow. The loop timer is > > calculated once at startup and can't change as the load the goes up. > > > > The _entire_ game loop depends on this timing cycle. The heavier the > > server load the more out of sync this clocking becomes. Short of a > > rewrite I was never able to do more than a few tweaks to it. Because of > > the way multicore systems handle thread swaps, tuxmath actually makes a > > killer test on cpu efficiency. > > This is an accurate assessment. Tuxmath and tuxtype are unusable for > LTSP. It is just poorly designed. It could possibly be fixed by a > programmer who understands this stuff. I tried digging into it about 2 years back but realized it's an architectural issue. At that point it was more than I could tackle to start from scratch. Yes. Tuxtype has the same issues as it uses the same engine.The SDL stuff is cross-platform nice but it has to make many, many compromises to work. > > Warren > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From robark at gmail.com Wed Apr 29 04:19:59 2009 From: robark at gmail.com (Robert Arkiletian) Date: Tue, 28 Apr 2009 21:19:59 -0700 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: <49F7BBAC.3010600@redhat.com> References: <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F71DC6.20901@gmail.com> <1240971098.10761.8.camel@merlin.localnetsolutions.com> <49F7BBAC.3010600@redhat.com> Message-ID: On Tue, Apr 28, 2009 at 7:30 PM, Warren Togami wrote: > On 04/28/2009 10:11 PM, James P. Kinney III wrote: >> >> David, >> >> As wonderful as tuxmath is (it's a great game/tool) is has a design >> problem that makes it unsuitable for LTSP. The screen draw rate is >> clocked by wait cycles as dummy loops. The system literally polls "is it >> time to send a screen update now". While the load is minimal, it pushes >> the next cycle onto a timer. So the system get dog slow and the load is >> not high yet the client end is uselessly slow. The loop timer is >> calculated once at startup and can't change as the load the goes up. >> >> The _entire_ game loop depends on this timing cycle. The heavier the >> server load the more out of sync this clocking becomes. Short of a >> rewrite I was never able to do more than a few tweaks to it. Because of >> the way multicore systems handle thread swaps, tuxmath actually makes a >> killer test on cpu efficiency. > > This is an accurate assessment. Tuxmath and tuxtype are unusable for LTSP. > It is just poorly designed. It could possibly be fixed by a programmer who > understands this stuff. What about just running it as a local app? /usr/bin/ltsp-localapps /usr/bin/tuxmath -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From lesmikesell at gmail.com Wed Apr 29 05:42:55 2009 From: lesmikesell at gmail.com (Les Mikesell) Date: Wed, 29 Apr 2009 00:42:55 -0500 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <1240546835.4535.14.camel@merlin.localnetsolutions.com> <49F64387.9080606@redhat.com> <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F71DC6.20901@gmail.com> Message-ID: <49F7E8DF.5010908@gmail.com> David Hopkins wrote: >> With k12ltsp you should be able to enable ssh on the clients so you could >> connect and run ifconfig. Did you measure the server NIC throughput under >> CentOS5.2+LTSP4.2 to see if there is a large difference? I think you >> mentioned the switch having flow control enabled. Can you turn it off? >> http://www.smallnetbuilder.com/index2.php?option=com_content&task=view&id=30212&pop=1&page=0&Itemid=54 >> > > Thanks!! > > The switches had flow control turned on which apparently was limiting > the Gig-E link to appr 100Mbps. I turned it off and launched 21 > sessions of Tuxmath. It wasn't playable but once I backed it down to > around 8-10 sessions, it was tolerable. However, tracking the > throughput on the switch as I backed it down, the max I ever saw was > 400Mbps and it stayed closer to 300Mbps. Load on the server never > exceeded 8 (on an 8cpu system). So, while a lot better, something is > still limiting the throughput. I thought it might be packet size > related, so I ran NetIO from this system to another on a Gig-E link > with the folllowing typical results: > > TCP connection established. > Packet size 1k bytes: 82750 KByte/s Tx, 112086 KByte/s Rx. > Packet size 2k bytes: 87499 KByte/s Tx, 113424 KByte/s Rx. > Packet size 4k bytes: 95495 KByte/s Tx, 112055 KByte/s Rx. > Packet size 8k bytes: 94037 KByte/s Tx, 114064 KByte/s Rx. > Packet size 16k bytes: 102218 KByte/s Tx, 114743 KByte/s Rx. > Packet size 32k bytes: 104406 KByte/s Tx, 114689 KByte/s Rx. > > These are representative (I ran the test 10 times). So, I'm not > totally sure what is up with the Tx numbers. > > Still, thanks to everyone ... I at least have a good idea of where to > look now. And, worst case, I can get a card and just use channel > bonding to get the throughput up if I have to with some hope that it > will actually work. Interesting thread on LTSP and flow control here: http://www.mail-archive.com/ltsp-discuss at lists.sourceforge.net/msg36260.html -- Les Mikesell lesmikesell at gmail.com From cisna-barry at wc235.k12.il.us Wed Apr 29 12:31:58 2009 From: cisna-barry at wc235.k12.il.us (Barry Cisna) Date: Wed, 29 Apr 2009 07:31:58 -0500 Subject: [K12OSN] Strange edubuntu Behavior Message-ID: <1241008318.10792.2.camel@hi2.wc235.k12.il.us> Jeff, Check your /etc/hosts file to verify you have any entry at the top of it : 127.0.0.1 localhost.localdomain localhost Reboot server then try the login. Barry Cisna From fyrteach at hotmail.com Wed Apr 29 12:59:40 2009 From: fyrteach at hotmail.com (Jeff Bonton) Date: Wed, 29 Apr 2009 07:59:40 -0500 Subject: [K12OSN] Strange edubuntu Behavior In-Reply-To: <49F79E2B.6040707@siddall.name> References: <49F79E2B.6040707@siddall.name> Message-ID: Jeff I did try the ltsp-update-sshkeys command with no success. The server has a static IP. Jeff > Date: Tue, 28 Apr 2009 20:24:11 -0400 > From: news at siddall.name > To: k12osn at redhat.com > Subject: Re: [K12OSN] Strange edubuntu Behavior > > Jeff Bonton wrote: > > I have a strange issue with an Edubuntu server that is running 9.04 in a > > school. I have been reading this list for some time and the level of > > knowledge is amazing. I am hoping that someone has experienced this > > problem and found a fix, or has an idea of what to look for. I am a noob > > compared to the skills I see on this list. > > > > I am thinking it has something to do with permissions, but that is only > > a guess > > > > When I go to log into the server it asks for my username once, but my > > password twice. If I want to launch an application that requires root > > access, ie: synaptic, it prompts for my password then does nothing. I > > can launch the application from the CLI but it also asks for my password > > twice. > > > > About the same time, I cannot remember if it appeared at exactly the > > same time, the clients started refusing to login. I enter the username > > and password, the screen says verifying password for about a minute and > > then it says that the server is not responding and goes to a black screen. > > > > Any help or ideas will be appreciated. > > > > Jeff Bonton > > I had some strange problems like "server not responding" until I ran > ltsp-update-sshkeys. In my case the cause was a server IP change so > this may not help you. Can't hurt to try though. > > Jeff > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see _________________________________________________________________ Rediscover Hotmail?: Get e-mail storage that grows with you. http://windowslive.com/RediscoverHotmail?ocid=TXT_TAGLM_WL_HM_Rediscover_Storage2_042009 -------------- next part -------------- An HTML attachment was scrubbed... URL: From fyrteach at hotmail.com Wed Apr 29 14:31:35 2009 From: fyrteach at hotmail.com (Jeff Bonton) Date: Wed, 29 Apr 2009 09:31:35 -0500 Subject: [K12OSN] Strange edubuntu Behavior In-Reply-To: <1241008318.10792.2.camel@hi2.wc235.k12.il.us> References: <1241008318.10792.2.camel@hi2.wc235.k12.il.us> Message-ID: Barry, The top 2 lines of my hosts file are in the format shown below with the server Ip and localhost and localdomain names specified for the domain. 127.0.0.1 localhost xxx.xxx.xxx.xxx localhost.localdomain localhost Jeff > Subject: Re: [K12OSN] Strange edubuntu Behavior > From: cisna-barry at wc235.k12.il.us > To: k12osn at redhat.com > Date: Wed, 29 Apr 2009 07:31:58 -0500 > > Jeff, > > Check your /etc/hosts file to verify you have any entry at the top of > it : > 127.0.0.1 localhost.localdomain localhost > > > Reboot server then try the login. > > > Barry Cisna > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see _________________________________________________________________ Rediscover Hotmail?: Get quick friend updates right in your inbox. http://windowslive.com/RediscoverHotmail?ocid=TXT_TAGLM_WL_HM_Rediscover_Updates2_042009 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brian at portsmouth-college.ac.uk Wed Apr 29 14:32:28 2009 From: brian at portsmouth-college.ac.uk (Brian Chivers) Date: Wed, 29 Apr 2009 15:32:28 +0100 Subject: [K12OSN] Help with script Message-ID: <49F864FC.9070508@portsmouth-college.ac.uk> I'm trying to use the script on the k12ltsp wiki(1) to remove firefox cache but the problem is that our students directories are in sub directories under /home_student e.g /home_student tutorgroup 2S-EG k123456 How can I alter the script to descend into each if the sub directories and wipe as it goes ?? Thanks Brian Chivers Portsmouth College (1)http://k12ltsp.org/mediawiki/index.php/Cache_Removal ------------------------------------------------------------------------------------------------ The views expressed here are my own and not necessarily the views of Portsmouth College From klaus at skolelinux.no Wed Apr 29 15:30:27 2009 From: klaus at skolelinux.no (Klaus Ade Johnstad) Date: Wed, 29 Apr 2009 17:30:27 +0200 Subject: [K12OSN] Help with script In-Reply-To: <49F864FC.9070508@portsmouth-college.ac.uk> References: <49F864FC.9070508@portsmouth-college.ac.uk> Message-ID: <200904291730.31805.klaus@skolelinux.no> Onsdag 29. april 2009 16:32, skrev Brian Chivers: > I'm trying to use the script on the k12ltsp wiki(1) to remove firefox > cache but the problem is that our students directories are in sub > directories under /home_student e.g > > /home_student > tutorgroup > 2S-EG > k123456 > > How can I alter the script to descend into each if the sub > directories and wipe as it goes ?? > > Thanks > Brian Chivers > Portsmouth College > > (1)http://k12ltsp.org/mediawiki/index.php/Cache_Removal > I'd use something like this, for x in `getent passwd|grep home_student|cut -d":" -f6`; do echo rm $x/.mozilla/firefox/*.default/Cache/*;done Remove the echo once you are sure that the rm command only will remove what you intend. Oh, and there is a millions ways to do this, wait for a few more answers, and choose the one you are most comfortable with. -- Klaus Ade 67E61D18B2C44F8A3DA35C6D849F9F5F 26FA477D -------------- 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 burke at thealmquists.net Wed Apr 29 16:14:15 2009 From: burke at thealmquists.net (Almquist Burke) Date: Wed, 29 Apr 2009 11:14:15 -0500 Subject: [K12OSN] Strange edubuntu Behavior In-Reply-To: References: <1241008318.10792.2.camel@hi2.wc235.k12.il.us> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Apr 29, 2009, at 9:31 AM, Jeff Bonton wrote: > Barry, > > The top 2 lines of my hosts file are in the format shown below with > the server Ip and localhost and localdomain names specified for the > domain. > > 127.0.0.1 localhost > xxx.xxx.xxx.xxx localhost.localdomain localhost Did you add that second line? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (Darwin) iEYEARECAAYFAkn4fNcACgkQxWV7OPa/g5HDKACdHac1DL8q9EVoYdaaIt/RWd3U HEYAnRYXZmgAvCpULNAeJm7aARcMl00Z =P/E6 -----END PGP SIGNATURE----- From fyrteach at hotmail.com Wed Apr 29 16:38:45 2009 From: fyrteach at hotmail.com (Jeff Bonton) Date: Wed, 29 Apr 2009 11:38:45 -0500 Subject: [K12OSN] Strange edubuntu Behavior In-Reply-To: References: <1241008318.10792.2.camel@hi2.wc235.k12.il.us> Message-ID: Almquist, I do not remember adding it. Should I save a copy with a different name, edit this one and try it to see what happens? Jeff > From: burke at thealmquists.net > Subject: Re: [K12OSN] Strange edubuntu Behavior > Date: Wed, 29 Apr 2009 11:14:15 -0500 > To: k12osn at redhat.com > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > On Apr 29, 2009, at 9:31 AM, Jeff Bonton wrote: > > > Barry, > > > > The top 2 lines of my hosts file are in the format shown below with > > the server Ip and localhost and localdomain names specified for the > > domain. > > > > 127.0.0.1 localhost > > xxx.xxx.xxx.xxx localhost.localdomain localhost > > > Did you add that second line? > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (Darwin) > > iEYEARECAAYFAkn4fNcACgkQxWV7OPa/g5HDKACdHac1DL8q9EVoYdaaIt/RWd3U > HEYAnRYXZmgAvCpULNAeJm7aARcMl00Z > =P/E6 > -----END PGP SIGNATURE----- > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see _________________________________________________________________ Rediscover Hotmail?: Get quick friend updates right in your inbox. http://windowslive.com/RediscoverHotmail?ocid=TXT_TAGLM_WL_HM_Rediscover_Updates2_042009 -------------- next part -------------- An HTML attachment was scrubbed... URL: From dahopkins429 at gmail.com Wed Apr 29 16:46:03 2009 From: dahopkins429 at gmail.com (David Hopkins) Date: Wed, 29 Apr 2009 12:46:03 -0400 Subject: [K12OSN] In tough times...knowledge is power....FOSSed 2009 Register today! In-Reply-To: <49F3989D.3040209@cmosnetworks.com> References: <49F395BC.5000602@redhat.com> <49F3989D.3040209@cmosnetworks.com> Message-ID: And that domain is blocked by the state firewall as a "Parked Domain" ... Delaware uses Lightspeed and I guess I need to send in a request to have the site cleared from the block list. On Sat, Apr 25, 2009 at 7:11 PM, "Terrell Prud? Jr." wrote: > Warren Togami wrote: >> >> On 04/25/2009 01:41 PM, David Trask wrote: >>> >>> Visit the FOSSed web site to learn more and to register today! ?Register >>> soon! ?FOSSed is coming up in just a few weeks! ?http://fossed.com >>> >> >> fossed.com goes to godaddy's parking page. > > Warren's right. ?The correct link is http://www.fossed.com. ?Note the "www" > in the hyperlink. > > --TP > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From monteslu at cox.net Wed Apr 29 17:34:38 2009 From: monteslu at cox.net (monteslu at cox.net) Date: Wed, 29 Apr 2009 13:34:38 -0400 (EDT) Subject: [K12OSN] public school technology budgets Message-ID: <21205595.4436.1241026478193.JavaMail.monteslu@127.0.0.1> Does anyone on the list have any idea what a typical yearly technology budget is at a public K-6 or K-8? I know this will vary wildy depending on districts, but I'm just looking for a ballpark. Over severaly years, I've volunteered at a private school building a large network of about 70 thin clients for 200 or so students on very little available for technology (about 10K per year) using ltsp and other various other open source software. Lately I haven't had much time to spend on fixing some things and of course a few MS fanboy parents have siezed the opportunity to say how much better off we'd be using windows. Any idea what kind of cash it takes to run a network that size on windows? Not just inital hardware and software purchases, but ongooing maintenance. I'm not trying to make a case for either side, I just want everyone to be aware of what it will cost them to buy machines and have someone to maintain them while kids are using them? I have no interest in volunteering hours in cleaning up spyware and viruses from a bunch of windows computers with hard drives, so they'll need to pay someone to do it. Thanks, Luis From dhuckaby at paasda.org Wed Apr 29 18:05:11 2009 From: dhuckaby at paasda.org (Huck) Date: Wed, 29 Apr 2009 11:05:11 -0700 Subject: [K12OSN] public school technology budgets In-Reply-To: <21205595.4436.1241026478193.JavaMail.monteslu@127.0.0.1> References: <21205595.4436.1241026478193.JavaMail.monteslu@127.0.0.1> Message-ID: <49F896D7.50804@paasda.org> for a 250 student private high school the budget is 25-30k/yr to cover everything hardware & software based...not support time/staff with about 150 computers in total..MAC/Linux/XP a K-8 private school with about 100 students and 60 machines, including tech time(20hrs/month) is about 15k/yr --Huck monteslu at cox.net wrote: > Does anyone on the list have any idea what a typical yearly technology > budget is at a public K-6 or K-8? > > I know this will vary wildy depending on districts, but I'm just looking > for a ballpark. > > Over severaly years, I've volunteered at a private school building a > large network of about 70 thin clients for 200 or so students on very > little available for technology (about 10K per year) using ltsp and > other various other open source software. Lately I haven't had much > time to spend on fixing some things and of course a few MS fanboy > parents have siezed the opportunity to say how much better off we'd be > using windows. > > Any idea what kind of cash it takes to run a network that size on > windows? Not just inital hardware and software purchases, but ongooing > maintenance. > I'm not trying to make a case for either side, I just want everyone to > be aware of what it will cost them to buy machines and have someone to > maintain them while kids are using them? > > I have no interest in volunteering hours in cleaning up spyware and > viruses from a bunch of windows computers with hard drives, so they'll > need to pay someone to do it. > > Thanks, > > Luis > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > From brcisna at eazylivin.net Wed Apr 29 18:24:52 2009 From: brcisna at eazylivin.net (Barry R Cisna) Date: Wed, 29 Apr 2009 13:24:52 -0500 (CDT) Subject: [K12OSN] Strange edubuntu Behavior Message-ID: <36364.216.24.126.66.1241029492.squirrel@www.eazylivin.net> Jeff, Comment # out the second line you posted( or delete it): the xxx.xxx. localhost edit the 127.0.0.1 line to reflect the first post i posted here subistitute < servername> to YOUR k12ltsp servername. reboot, i'm guessing you may be golden now. let us know. Barry From Steven at SimplyCircus.com Wed Apr 29 19:00:56 2009 From: Steven at SimplyCircus.com (Steven Santos) Date: Wed, 29 Apr 2009 15:00:56 -0400 Subject: [K12OSN] public school technology budgets In-Reply-To: <21205595.4436.1241026478193.JavaMail.monteslu@127.0.0.1> References: <21205595.4436.1241026478193.JavaMail.monteslu@127.0.0.1> Message-ID: Since I am looking at opening a charter school... State says that the complete tech budget should be between $135 and $250 per year per pupil, plus a start-up of about 1.5 years use. Out of this should come: - Tech Support - Hardware replacement (5 year cycle) - Software licensing - Database licensing - Internet access - Web Hosting - Printing In our budget we are putting in $207 per pupil for the 50 students and 8 staff we want to start with, or $12k per year, and $22k initial investment. Here is our breakdown: Initial Costs: 2 LTSP Servers $8,000 1 MS2k7 Server+lic $4,000 2 24 Port Gb switches $1,000 1 Wiring of building $1,000 (parents doing labor) 35 terminals $7,245 Total: $21,245 - Tech Support $0 (parent is doing it) - Hardware replacement $2,000 (5 year cycle) - Software licensing $1,000 - Database licensing $2,000 - Internet access $1,200 - Web Hosting $ 800 That still leaves us with 6K/year to work with. --- Steven Santos Director, Simply Circus, Inc. Email: Steven at SimplyCircus.com Mail: 14 Pierrepont Road Newton, MA 02462 Phone: 617-527-0667 Web: www.SimplyCircus.com > -----Original Message----- > From: k12osn-bounces at redhat.com [mailto:k12osn-bounces at redhat.com] On > Behalf Of monteslu at cox.net > Sent: Wednesday, April 29, 2009 1:35 PM > To: k12osn at redhat.com > Subject: [K12OSN] public school technology budgets > > Does anyone on the list have any idea what a typical yearly technology > budget is at a public K-6 or K-8? > > I know this will vary wildy depending on districts, but I'm just > looking > for a ballpark. > > Over severaly years, I've volunteered at a private school building a > large network of about 70 thin clients for 200 or so students on very > little available for technology (about 10K per year) using ltsp and > other various other open source software. Lately I haven't had much > time to spend on fixing some things and of course a few MS fanboy > parents have siezed the opportunity to say how much better off we'd be > using windows. > > Any idea what kind of cash it takes to run a network that size on > windows? Not just inital hardware and software purchases, but ongooing > maintenance. > I'm not trying to make a case for either side, I just want everyone to > be aware of what it will cost them to buy machines and have someone to > maintain them while kids are using them? > > I have no interest in volunteering hours in cleaning up spyware and > viruses from a bunch of windows computers with hard drives, so they'll > need to pay someone to do it. > > Thanks, > > Luis > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From ascensiontech at gmail.com Wed Apr 29 20:31:16 2009 From: ascensiontech at gmail.com (Peter Hartmann) Date: Wed, 29 Apr 2009 16:31:16 -0400 Subject: [K12OSN] dhcpd windows clients Message-ID: <9bd317560904291331j27c3a6efq82ee336ae98d46dc@mail.gmail.com> For some reason Windows dhcp clients of the ltsp server are getting an ip address of 0.0.0.0. Weird because I don't remember messing around with the conf file lately. ipconfig /all shows the dhcp server as 192.168.0.254. Does this sound familiar to anyone? Thanks, Peter From jkinney at localnetsolutions.com Wed Apr 29 22:51:47 2009 From: jkinney at localnetsolutions.com (James P. Kinney III) Date: Wed, 29 Apr 2009 18:51:47 -0400 Subject: [K12OSN] More feedback on Fedora 10 + LTSP In-Reply-To: References: <49F69281.9000206@redhat.com> <49F6FC75.2020602@gmail.com> <49F71DC6.20901@gmail.com> <1240971098.10761.8.camel@merlin.localnetsolutions.com> <49F7BBAC.3010600@redhat.com> Message-ID: <1241045507.12045.7.camel@merlin.localnetsolutions.com> On Tue, 2009-04-28 at 21:19 -0700, Robert Arkiletian wrote: > What about just running it as a local app? > > /usr/bin/ltsp-localapps /usr/bin/tuxmath > That's the only way to run tuxmath and tuxtype. In fact most SDL applications should be local_app only. > -- > Robert Arkiletian > Eric Hamber Secondary, Vancouver, Canada > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- James P. Kinney III CEO & Director of Engineering Local Net Solutions,LLC http://www.localnetsolutions.com GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics) Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From wtogami at redhat.com Thu Apr 30 22:46:15 2009 From: wtogami at redhat.com (Warren Togami) Date: Thu, 30 Apr 2009 18:46:15 -0400 Subject: [K12OSN] K12Linux Live Server F11 Beta1 Message-ID: <49FA2A37.8050702@redhat.com> http://alt.fedoraproject.org/pub/alt/ltsp/k12linux/f11/beta1/ x86_64 only for now. The next beta or final release will have i386 ISO's as well. The final release will be sometime after Fedora 11 is released. This is based on Fedora 11 rawhide as of today. The drivers are newer than Fedora 10. I suspect random things will work better, while other random things wont. The harmless but annoying selinux error messages were reported in Bugzilla. Warren Togami wtogami at redhat.com