From member at linkedin.com Thu Aug 4 15:19:27 2011 From: member at linkedin.com (Daniel Newman via LinkedIn) Date: Thu, 4 Aug 2011 15:19:27 +0000 (UTC) Subject: [K12OSN] Invitation to connect on LinkedIn Message-ID: <1847094268.8742146.1312471167878.JavaMail.app@ela4-app0128.prod> LinkedIn ------------ Daniel Newman requested to add you as a connection on LinkedIn: ------------------------------------------ Sergio, I'd like to add you to my professional network on LinkedIn. - Daniel Accept invitation from Daniel Newman http://www.linkedin.com/e/4uridb-gqxvg2gz-56/pALzX4kTsIfRQ1ozG0pFkv3pCtGs/blk/I207722693_20/1BpC5vrmRLoRZcjkkZt5YCpnlOt3RApnhMpmdzgmhxrSNBszYMcBYPejoOczsTc399bSheijdLtkh9bPsMcz0RejwRdPcLrCBxbOYWrSlI/EML_comm_afe/ View invitation from Daniel Newman http://www.linkedin.com/e/4uridb-gqxvg2gz-56/pALzX4kTsIfRQ1ozG0pFkv3pCtGs/blk/I207722693_20/30OnPcVdz8OdPsMcAALqnpPbOYWrSlI/svi/ ------------------------------------------ DID YOU KNOW you can be the first to know when a trusted member of your network changes jobs? With Network Updates on your LinkedIn home page, you'll be notified as members of your network change their current position. Be the first to know and reach out! http://www.linkedin.com/ -- (c) 2011, LinkedIn Corporation -------------- next part -------------- An HTML attachment was scrubbed... URL: From warren at togami.com Tue Aug 9 11:11:08 2011 From: warren at togami.com (Warren Togami) Date: Tue, 09 Aug 2011 01:11:08 -1000 Subject: [K12OSN] Progress: Fedora 14 Now Supported, Fixes, Public Beta Soon Message-ID: <4E4115CC.2030206@togami.com> Public Beta Soon ================ https://fedorahosted.org/k12linux/wiki/EL6Status 3 remaining easy bugs before the release of the Public Beta. I am attempting to get that done and to revamp the k12linux.org documentation this week. Does ANYONE have a non-US keyboard? If so, please contact me, we need some non-US keyboard testing done before the Public Beta. Today I thought of a new way to do the replacement EL6 kernel (PAE disabled, nbd.ko enabled, extraneous drivers removed). This would improve the Public Beta much in terms of client hardware compatibility, so I will attempt to do it too. Major Changes since Private Beta ================================ * Fedora 14 is now supported. If you have the URL from the Private Beta, you should find k12linux-release for Fedora 14 and you can proceed to install like normal. Please report back your experiences. * Many little fixes, faster install, less noisy, smaller. * K12Linux EL6 is now based on EL6. Everything works great, except they added an annoying regression to livecd-creator that I am trying to fix. * LDM login screen is now less broken, language and desktop selection now works. Detailed Changelog ================== - Run ltsp-chroot-setup only once during ltsp-build-client. - Speed up ltsp-build-client - Remove many error and warning messages during ltsp-build-client NOTE: Remaining warnings are from a new bug introduced by SL-6.1 livecd-creator. They are harmless. - Added wraplinux and mkelfimage to EL6 chroot Needed to boot on linuxbios, coreboot and some old client hardware. - ltsp-update-kernels prunes old symvers-*.gz - Shrink initramfs image by 1.5MB by removing dracut modules unneeded by netboot thin clients. - Allow ltsp-build-client to work on Fedora 14 as a server. Theoretically this should make Fedora 14 fully supported. - SL-6.1 - k12linux-release package contains yum definitions for K12Linux repo Warren Togami warren at togami.com From microman at cmosnetworks.com Tue Aug 9 14:43:37 2011 From: microman at cmosnetworks.com (Terrell Prude' Jr.) Date: Tue, 09 Aug 2011 10:43:37 -0400 Subject: [K12OSN] Progress: Fedora 14 Now Supported, Fixes, Public Beta Soon In-Reply-To: <4E4115CC.2030206@togami.com> References: <4E4115CC.2030206@togami.com> Message-ID: <4E414799.9000808@cmosnetworks.com> Warren, You are the man for doing all this. Talk about a bargain.... --TP Warren Togami wrote: > Public Beta Soon > ================ > https://fedorahosted.org/k12linux/wiki/EL6Status > 3 remaining easy bugs before the release of the Public Beta. I am > attempting to get that done and to revamp the k12linux.org > documentation this week. > > Does ANYONE have a non-US keyboard? If so, please contact me, we need > some non-US keyboard testing done before the Public Beta. > > Today I thought of a new way to do the replacement EL6 kernel (PAE > disabled, nbd.ko enabled, extraneous drivers removed). This would > improve the Public Beta much in terms of client hardware > compatibility, so I will attempt to do it too. > > Major Changes since Private Beta > ================================ > * Fedora 14 is now supported. If you have the URL from the Private > Beta, you should find k12linux-release for Fedora 14 and you can > proceed to install like normal. Please report back your experiences. > * Many little fixes, faster install, less noisy, smaller. > * K12Linux EL6 is now based on EL6. Everything works great, except > they added an annoying regression to livecd-creator that I am trying > to fix. > * LDM login screen is now less broken, language and desktop selection > now works. > > Detailed Changelog > ================== > - Run ltsp-chroot-setup only once during ltsp-build-client. > - Speed up ltsp-build-client > - Remove many error and warning messages during ltsp-build-client > NOTE: Remaining warnings are from a new bug introduced by > SL-6.1 livecd-creator. They are harmless. > - Added wraplinux and mkelfimage to EL6 chroot > Needed to boot on linuxbios, coreboot and some old client hardware. > - ltsp-update-kernels prunes old symvers-*.gz > - Shrink initramfs image by 1.5MB by removing dracut modules > unneeded by netboot thin clients. > - Allow ltsp-build-client to work on Fedora 14 as a server. > Theoretically this should make Fedora 14 fully supported. > - SL-6.1 > - k12linux-release package contains yum definitions for K12Linux repo > > Warren Togami > warren at togami.com > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From warren at togami.com Wed Aug 10 03:21:59 2011 From: warren at togami.com (Warren Togami) Date: Tue, 09 Aug 2011 17:21:59 -1000 Subject: [K12OSN] EL6 32bit kernel-thinclient config Message-ID: <4E41F957.4010404@togami.com> Hi folks, I managed to get this to build with the RHEL6 kernel sources plus one patch to workaround Red Hat Bugzilla #729488. I am now working on figuring out how to package it in a supportable RPM. http://togami.com/~warren/temp/config-thin-client.txt This config file disables PAE, enables nbd.ko, and disables many components that are irrelevant to thin clients. So far it shrank the initramfs image from 16.7MB to 10.7MB. It still could be reduced further. I plan on disabling IPv6 and maybe some individual RAID controller drivers. Any recommendations of more options to disable? Warren From microman at cmosnetworks.com Wed Aug 10 14:17:07 2011 From: microman at cmosnetworks.com (=?ISO-8859-1?Q?=22Terrell_Prud=E9_Jr=2E=22?=) Date: Wed, 10 Aug 2011 07:17:07 -0700 Subject: [K12OSN] EL6 32bit kernel-thinclient config In-Reply-To: <4E41F957.4010404@togami.com> References: <4E41F957.4010404@togami.com> Message-ID: <4E4292E3.6000404@cmosnetworks.com> The only possible concern I might have is if LTSP5 supports IPv6, and if that's a foreseeable deployment in the next 6 years (remaining life of CentOS 6). The reason for this is that some folks, like me, have done K12LTSP deployments with the thin clients on the school's main LAN for both technical and political reasons. I know of at least one school which has LAN segments that are IPv6. Perhaps this isn't a problem in the real world, since I personally don't see even that school turning off IPv4 support for at least 10 years. Thoughts, anyone? --TP Warren Togami wrote: > Hi folks, > > I managed to get this to build with the RHEL6 kernel sources plus one > patch to workaround Red Hat Bugzilla #729488. I am now working on > figuring out how to package it in a supportable RPM. > > http://togami.com/~warren/temp/config-thin-client.txt > This config file disables PAE, enables nbd.ko, and disables many > components that are irrelevant to thin clients. So far it shrank the > initramfs image from 16.7MB to 10.7MB. > > It still could be reduced further. I plan on disabling IPv6 and maybe > some individual RAID controller drivers. Any recommendations of more > options to disable? > > Warren > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From news at siddall.name Wed Aug 10 15:49:20 2011 From: news at siddall.name (Jeff Siddall) Date: Wed, 10 Aug 2011 11:49:20 -0400 Subject: [K12OSN] Flash sound gone? Message-ID: <4E42A880.7040307@siddall.name> My 64 bit SL6 LTSP system had working Flash for a while. I think it was 32 bit Flash 10 with a wrapper running in 64 bit Firefox. Anyway, for whatever reason sound suddenly stopped working. So I tried upgrading to the version 11 64 bit native flash player from Adobe. Again, video works fine but still no audio -- either on the server console or the thin client. PulseAudio seems fine, as paplay works on both the server and the client. However, aplay returns: ALSA lib pcm_dmix.c:1018:(snd_pcm_dmix_open) unable to open slave aplay: main:660: audio open error: Device or resource busy so if flash is trying to use alsa that might be the problem. I have the following alsa stuff installed: rpm -qa | grep alsa alsa-utils-1.0.24.2-24.el6.x86_64 alsa-oss-1.0.17-4.el6.x86_64 alsa-lib-1.0.24.1-35.el6.x86_64 alsa-oss-libs-1.0.17-4.el6.x86_64 alsa-plugins-pulseaudio-1.0.21-3.el6.x86_64 and I don't have any asoundrc files anywhere. Any ideas how to get sound going again? Thanks, Jeff From jim.kinney at gmail.com Wed Aug 10 15:48:28 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Wed, 10 Aug 2011 11:48:28 -0400 Subject: [K12OSN] EL6 32bit kernel-thinclient config In-Reply-To: <4E4292E3.6000404@cmosnetworks.com> References: <4E41F957.4010404@togami.com> <4E4292E3.6000404@cmosnetworks.com> Message-ID: I don't see a compelling reason to disable a future capability like IPv6 for space reasons. Most schools will be IPv4 until their upstream ISP requires them to be otherwise. The ones that do implement IPv6 will likely do a dual stack first anyway so a sudden failure of the thin clients will be bad. Things that can come out of the thin client kernel are the oddball networking protocols for novell netware and windows shares as that is handled at the server end anyway and all the ham network protocols. On Wed, Aug 10, 2011 at 10:17 AM, "Terrell Prud? Jr." < microman at cmosnetworks.com> wrote: > The only possible concern I might have is if LTSP5 supports IPv6, and if > that's a foreseeable deployment in the next 6 years (remaining life of > CentOS 6). The reason for this is that some folks, like me, have done > K12LTSP deployments with the thin clients on the school's main LAN for both > technical and political reasons. I know of at least one school which has > LAN segments that are IPv6. > > Perhaps this isn't a problem in the real world, since I personally don't > see even that school turning off IPv4 support for at least 10 years. > > Thoughts, anyone? > > --TP > > > Warren Togami wrote: > >> Hi folks, >> >> I managed to get this to build with the RHEL6 kernel sources plus one >> patch to workaround Red Hat Bugzilla #729488. I am now working on figuring >> out how to package it in a supportable RPM. >> >> http://togami.com/~warren/**temp/config-thin-client.txt >> This config file disables PAE, enables nbd.ko, and disables many >> components that are irrelevant to thin clients. So far it shrank the >> initramfs image from 16.7MB to 10.7MB. >> >> It still could be reduced further. I plan on disabling IPv6 and maybe >> some individual RAID controller drivers. Any recommendations of more >> options to disable? >> >> Warren >> >> ______________________________**_________________ >> 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 > -- -- James P. Kinney III As long as the general population is passive, apathetic, diverted to consumerism or hatred of the vulnerable, then the powerful can do as they please, and those who survive will be left to contemplate the outcome. - *2011 Noam Chomsky http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From ltsp at symbio-technologies.com Wed Aug 10 16:00:28 2011 From: ltsp at symbio-technologies.com (Gideon Romm) Date: Wed, 10 Aug 2011 12:00:28 -0400 Subject: [K12OSN] EL6 32bit kernel-thinclient config In-Reply-To: <4E4292E3.6000404@cmosnetworks.com> References: <4E41F957.4010404@togami.com> <4E4292E3.6000404@cmosnetworks.com> Message-ID: The PXE code in the BIOS is only IPv4 aware. An IPv6 LTSP solution will necessarily rely on something other than PXE for booting. I am sure gPXE is most likely working on IPv6 support (if they don't have it already), but if you are to rely on the PXE code that ships with most machines by default, you will be limited to IPv4. That said, having the possibility in the OS might still be welcome, both for alternative boot methods and in case you want to boot from an IPv4 segment but have the OS use IPv6 once booted. -Gadi On Wed, Aug 10, 2011 at 10:17 AM, "Terrell Prud? Jr." wrote: > The only possible concern I might have is if LTSP5 supports IPv6, and if > that's a foreseeable deployment in the next 6 years (remaining life of > CentOS 6). ?The reason for this is that some folks, like me, have done > K12LTSP deployments with the thin clients on the school's main LAN for both > technical and political reasons. ?I know of at least one school which has > LAN segments that are IPv6. > > Perhaps this isn't a problem in the real world, since I personally don't see > even that school turning off IPv4 support for at least 10 years. > > Thoughts, anyone? > > --TP > > Warren Togami wrote: >> >> Hi folks, >> >> I managed to get this to build with the RHEL6 kernel sources plus one >> patch to workaround Red Hat Bugzilla #729488. ?I am now working on figuring >> out how to package it in a supportable RPM. >> >> http://togami.com/~warren/temp/config-thin-client.txt >> This config file disables PAE, enables nbd.ko, and disables many >> components that are irrelevant to thin clients. ?So far it shrank the >> initramfs image from 16.7MB to 10.7MB. >> >> It still could be reduced further. ?I plan on disabling IPv6 and maybe >> some individual RAID controller drivers. ?Any recommendations of more >> options to disable? >> >> Warren >> >> _______________________________________________ >> 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 dean.jones at oregonstate.edu Wed Aug 10 17:14:43 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Wed, 10 Aug 2011 10:14:43 -0700 Subject: [K12OSN] EL6 repo? In-Reply-To: References: <4E41F957.4010404@togami.com> <4E4292E3.6000404@cmosnetworks.com> Message-ID: <4E42BC83.7060108@oregonstate.edu> Hi, Is there an EL6 repo setup for testing? I can't find anything on the wiki but perhaps i'm looking in the wrong spot. I have a test setup running on CentOS 6 using the k12linux-temporary RHEL 5 repo. It is working fine for basic functions but I would be interested in testing RHEL 6. thanks From warren at togami.com Wed Aug 10 20:31:24 2011 From: warren at togami.com (Warren Togami Jr.) Date: Wed, 10 Aug 2011 10:31:24 -1000 Subject: [K12OSN] EL6 32bit kernel-thinclient config In-Reply-To: References: <4E41F957.4010404@togami.com> <4E4292E3.6000404@cmosnetworks.com> Message-ID: <4E42EA9C.1060308@togami.com> On 8/10/2011 6:00 AM, Gideon Romm wrote: > The PXE code in the BIOS is only IPv4 aware. An IPv6 LTSP solution > will necessarily rely on something other than PXE for booting. > > I am sure gPXE is most likely working on IPv6 support (if they don't > have it already), but if you are to rely on the PXE code that ships > with most machines by default, you will be limited to IPv4. > > That said, having the possibility in the OS might still be welcome, > both for alternative boot methods and in case you want to boot from an > IPv4 segment but have the OS use IPv6 once booted. > > -Gadi Indeed you have it right. In any case, I figured out that I cannot easily disable IPv6 because it causes the heavily hacked-up EL6 kernel to fail to build. kernel-thinclient.i686 is now in the Private Beta repo. It enables NBD, disables PAE, and shrinks the initramfs size from 16.7MB to 13.8MB. I think I can shrink the initramfs size more not by excluding more modules from the kernel build, but rather tweaking dracut's configuration that is including so many rarely needed kernel modules just-in-case they might be needed to boot something. Warren From warren at togami.com Thu Aug 11 00:31:55 2011 From: warren at togami.com (Warren Togami) Date: Wed, 10 Aug 2011 14:31:55 -1000 Subject: [K12OSN] Flash sound gone? In-Reply-To: <4E42A880.7040307@siddall.name> References: <4E42A880.7040307@siddall.name> Message-ID: <4E4322FB.4000509@togami.com> On 8/10/2011 5:49 AM, Jeff Siddall wrote: > My 64 bit SL6 LTSP system had working Flash for a while. I think it was > 32 bit Flash 10 with a wrapper running in 64 bit Firefox. Anyway, for > whatever reason sound suddenly stopped working. So I tried upgrading to > the version 11 64 bit native flash player from Adobe. Again, video works > fine but still no audio -- either on the server console or the thin client. > > PulseAudio seems fine, as paplay works on both the server and the > client. However, aplay returns: > > ALSA lib pcm_dmix.c:1018:(snd_pcm_dmix_open) unable to open slave > aplay: main:660: audio open error: Device or resource busy > > so if flash is trying to use alsa that might be the problem. I have the > following alsa stuff installed: > > rpm -qa | grep alsa > alsa-utils-1.0.24.2-24.el6.x86_64 > alsa-oss-1.0.17-4.el6.x86_64 > alsa-lib-1.0.24.1-35.el6.x86_64 > alsa-oss-libs-1.0.17-4.el6.x86_64 > alsa-plugins-pulseaudio-1.0.21-3.el6.x86_64 > > and I don't have any asoundrc files anywhere. Never experienced any problems here. I recommend NOT using 32bit flash-plugin on a 64bit server because it pulls in a huge 32bit library stack. I now recommend uninstalling nspluginwrapper completely because Firefox now has its own out-of-process plugin-container that protects the browser from plugin crashes. Flash itself seems to crash less often when run outside of nspluginwrapper. [warren at rush ~]$ rpm -qa 'alsa*' alsa-plugins-pulseaudio-1.0.21-3.el6.x86_64 alsa-utils-1.0.21-3.el6.x86_64 alsa-lib-1.0.21-3.el6.x86_64 [warren at rush ~]$ rpm -q nspluginwrapper package nspluginwrapper is not installed [warren at rush ~]$ rpm -q firefox firefox-3.6.18-1.el6_1.x86_64 [warren at rush ~]$ ls -l /usr/lib64/mozilla/plugins/libflashplayer.so -rwxr-xr-x. 1 root root 18708232 Aug 10 13:46 /usr/lib64/mozilla/plugins/libflashplayer.so [warren at rush ~]$ strings /usr/lib64/mozilla/plugins/libflashplayer.so |grep "Shockwave Flash 11" Shockwave Flash 11.0 d1 http://labs.adobe.com/technologies/flashplatformruntimes/flashplayer11/ Download the 64bit version here (updated a few days ago), put libflashplayer.so in /usr/lib64/mozilla/plugins. I suspect you need to make it chmod 755. Warren From sbarar at gmail.com Thu Aug 11 01:30:35 2011 From: sbarar at gmail.com (Sudev Barar) Date: Thu, 11 Aug 2011 07:00:35 +0530 Subject: [K12OSN] EL6 repo? In-Reply-To: <4E42BC83.7060108@oregonstate.edu> References: <4E41F957.4010404@togami.com> <4E4292E3.6000404@cmosnetworks.com> <4E42BC83.7060108@oregonstate.edu> Message-ID: On 10-Aug-2011 10:49 PM, "Dean Jones" wrote: Is there an EL6 repo setup for testing? I can't find anything on the wiki but perhaps i'm looking in the wrong spot. > Warren has been working on this for the community. See his emails in archives. He is in beta testing add of now. -- Sudev -------------- next part -------------- An HTML attachment was scrubbed... URL: From warren at togami.com Thu Aug 11 01:39:53 2011 From: warren at togami.com (Warren Togami Jr.) Date: Wed, 10 Aug 2011 15:39:53 -1000 Subject: [K12OSN] EL6 repo? In-Reply-To: References: <4E41F957.4010404@togami.com> <4E4292E3.6000404@cmosnetworks.com> <4E42BC83.7060108@oregonstate.edu> Message-ID: <4E4332E9.2030602@togami.com> On 8/10/2011 3:30 PM, Sudev Barar wrote: > On 10-Aug-2011 10:49 PM, "Dean Jones" > wrote: > Is there an EL6 repo setup for testing? I can't find anything on the > wiki but perhaps i'm looking in the wrong spot. > > > > Warren has been working on this for the community. See his emails in > archives. He is in beta testing add of now. > > -- > Sudev Hint: You can find the download link if you can find the source code repository. I am not announcing it yet because I haven't had time to redo the documentation on k12linux.org. But I consider the current code there to be almost ready for production. Warren From news at siddall.name Thu Aug 11 02:20:41 2011 From: news at siddall.name (Jeff Siddall) Date: Wed, 10 Aug 2011 22:20:41 -0400 Subject: [K12OSN] Flash sound gone? In-Reply-To: <4E4322FB.4000509@togami.com> References: <4E42A880.7040307@siddall.name> <4E4322FB.4000509@togami.com> Message-ID: <4E433C79.4040203@siddall.name> On 08/10/2011 08:31 PM, Warren Togami wrote: > Never experienced any problems here. I recommend NOT using 32bit > flash-plugin on a 64bit server because it pulls in a huge 32bit library > stack. I now recommend uninstalling nspluginwrapper completely because > Firefox now has its own out-of-process plugin-container that protects > the browser from plugin crashes. Flash itself seems to crash less often > when run outside of nspluginwrapper. Thanks for update. Does aplay work on your server/client? That might point me in the right direction. Jeff From warren at togami.com Thu Aug 11 05:14:22 2011 From: warren at togami.com (Warren Togami Jr.) Date: Wed, 10 Aug 2011 19:14:22 -1000 Subject: [K12OSN] Flash sound gone? In-Reply-To: <4E433C79.4040203@siddall.name> References: <4E42A880.7040307@siddall.name> <4E4322FB.4000509@togami.com> <4E433C79.4040203@siddall.name> Message-ID: <4E43652E.8070602@togami.com> On 8/10/2011 4:20 PM, Jeff Siddall wrote: > On 08/10/2011 08:31 PM, Warren Togami wrote: >> Never experienced any problems here. I recommend NOT using 32bit >> flash-plugin on a 64bit server because it pulls in a huge 32bit library >> stack. I now recommend uninstalling nspluginwrapper completely because >> Firefox now has its own out-of-process plugin-container that protects >> the browser from plugin crashes. Flash itself seems to crash less often >> when run outside of nspluginwrapper. > > Thanks for update. > > Does aplay work on your server/client? That might point me in the right > direction. > > Jeff Yes, aplay works fine on two of my thin clients I tested just now. Are you sure this specific client hardware had working sound with SL-6 before? Is the server SL-6.1 with all updates? You may also try updating to the latest ltsp-server then rebuild /opt/ltsp/i386? Warren From warren at togami.com Thu Aug 11 09:05:01 2011 From: warren at togami.com (Warren Togami Jr.) Date: Wed, 10 Aug 2011 23:05:01 -1000 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback Message-ID: <4E439B3D.3040208@togami.com> Hey folks, http://k12linux.org I just spent a few hours reorganizing the K12Linux site, hiding most of the old and obsolete pages and putting the important stuff up front. I still plan on doing more reorganizing and writing. I have strong confidence in the code itself, but the documentation is probably too confusing and requires improvement. For now I really need people who are already familiar with LTSP to test the instructions AS WRITTEN with a fresh install and to provide feedback here on K12OSN list. After some verification and tweaking of the documentation, we will be ready for the formal announcement where we invite the wider community to K12Linux EL6. https://www.facebook.com/k12linux.org Follow K12Linux on Facebook! Warren From news at siddall.name Thu Aug 11 15:29:35 2011 From: news at siddall.name (Jeff Siddall) Date: Thu, 11 Aug 2011 11:29:35 -0400 Subject: [K12OSN] Flash sound gone? In-Reply-To: <4E43652E.8070602@togami.com> References: <4E42A880.7040307@siddall.name> <4E4322FB.4000509@togami.com> <4E433C79.4040203@siddall.name> <4E43652E.8070602@togami.com> Message-ID: <4E43F55F.5080601@siddall.name> On 08/11/2011 01:14 AM, Warren Togami Jr. wrote: > Yes, aplay works fine on two of my thin clients I tested just now. Interesting. I guess alsa is mucked up somehow. > Are you sure this specific client hardware had working sound with SL-6 > before? Yup, both the server and client had working flash. PA aware sound applications still work fine on both so there _is_ sound, just not for alsa apps it appears. > Is the server SL-6.1 with all updates? I have all the yum updates but /etc/redhat-release still shows SL 6.0. Interesting. In CentOS if you installed updates you automatically got the latest version. I guess I need to look into how to upgrade point releases in SL. > You may also try updating to the latest ltsp-server then rebuild > /opt/ltsp/i386? OK, I will do that also. Thanks, Jeff From news at siddall.name Thu Aug 11 17:08:13 2011 From: news at siddall.name (Jeff Siddall) Date: Thu, 11 Aug 2011 13:08:13 -0400 Subject: [K12OSN] SL 6.1 upgrade Message-ID: <4E440C7D.7030305@siddall.name> For anyone who got lazy with how CentOS automatically updates to the latest dot release, here's the command to get a specific dot release in SL (6.1 in this case): yum --releasever=6.1 update Jeff From jim.kinney at gmail.com Thu Aug 11 17:37:30 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Thu, 11 Aug 2011 13:37:30 -0400 Subject: [K12OSN] SL 6.1 upgrade In-Reply-To: <4E440C7D.7030305@siddall.name> References: <4E440C7D.7030305@siddall.name> Message-ID: Very good to know! On Thu, Aug 11, 2011 at 1:08 PM, Jeff Siddall wrote: > For anyone who got lazy with how CentOS automatically updates to the latest > dot release, here's the command to get a specific dot release in SL (6.1 in > this case): > > yum --releasever=6.1 update > > Jeff > > ______________________________**_________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/**mailman/listinfo/k12osn > For more info see > -- -- James P. Kinney III As long as the general population is passive, apathetic, diverted to consumerism or hatred of the vulnerable, then the powerful can do as they please, and those who survive will be left to contemplate the outcome. - *2011 Noam Chomsky http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From dean.jones at oregonstate.edu Thu Aug 11 17:44:06 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Thu, 11 Aug 2011 10:44:06 -0700 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E439B3D.3040208@togami.com> References: <4E439B3D.3040208@togami.com> Message-ID: <4E4414E6.50404@oregonstate.edu> Hi, I am working on a fresh install for feedback following the instructions here: https://fedorahosted.org/k12linux/wiki/InstallGuide But I can't connect to the mplug.org repo from any of my centos hosts (old or new)? I just get a time out trying to hit port 80. Browser, wget, telnet port 80 all behave the same. Works fine from a mac?! On 8/11/11 2:05 AM, Warren Togami Jr. wrote: > Hey folks, > > http://k12linux.org > I just spent a few hours reorganizing the K12Linux site, hiding most of > the old and obsolete pages and putting the important stuff up front. I > still plan on doing more reorganizing and writing. > > I have strong confidence in the code itself, but the documentation is > probably too confusing and requires improvement. > > For now I really need people who are already familiar with LTSP to test > the instructions AS WRITTEN with a fresh install and to provide feedback > here on K12OSN list. After some verification and tweaking of the > documentation, we will be ready for the formal announcement where we > invite the wider community to K12Linux EL6. > > https://www.facebook.com/k12linux.org > Follow K12Linux on Facebook! > > Warren > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From jim.kinney at gmail.com Thu Aug 11 18:33:03 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Thu, 11 Aug 2011 14:33:03 -0400 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E4414E6.50404@oregonstate.edu> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> Message-ID: Must be a temporary thing. I can hit it just fine from a firefox browser on F15. On Thu, Aug 11, 2011 at 1:44 PM, Dean Jones wrote: > Hi, > > I am working on a fresh install for feedback following the instructions > here: > https://fedorahosted.org/**k12linux/wiki/InstallGuide > > But I can't connect to the mplug.org repo from any of my centos hosts (old > or new)? I just get a time out trying to hit port 80. Browser, wget, > telnet port 80 all behave the same. > > Works fine from a mac?! > > > On 8/11/11 2:05 AM, Warren Togami Jr. wrote: > >> Hey folks, >> >> http://k12linux.org >> I just spent a few hours reorganizing the K12Linux site, hiding most of >> the old and obsolete pages and putting the important stuff up front. I >> still plan on doing more reorganizing and writing. >> >> I have strong confidence in the code itself, but the documentation is >> probably too confusing and requires improvement. >> >> For now I really need people who are already familiar with LTSP to test >> the instructions AS WRITTEN with a fresh install and to provide feedback >> here on K12OSN list. After some verification and tweaking of the >> documentation, we will be ready for the formal announcement where we >> invite the wider community to K12Linux EL6. >> >> https://www.facebook.com/**k12linux.org >> Follow K12Linux on Facebook! >> >> Warren >> >> ______________________________**_________________ >> 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 > -- -- James P. Kinney III As long as the general population is passive, apathetic, diverted to consumerism or hatred of the vulnerable, then the powerful can do as they please, and those who survive will be left to contemplate the outcome. - *2011 Noam Chomsky http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From dean.jones at oregonstate.edu Thu Aug 11 19:42:34 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Thu, 11 Aug 2011 12:42:34 -0700 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> Message-ID: <4E4430AA.1020303@oregonstate.edu> Hah yeah, it started working 5 minutes after I sent this. No idea what the issue was. Here are my test notes: From these instructions: https://fedorahosted.org/k12linux/wiki/InstallGuide CentOS 6 minimal install. x86_64, netinstall. Target clients are Thin clients in kiosk mode (firefox only). (old hardware, slow video, no fuse etc) Not really related to LTSP steps before starting: yum update yum install libssh2.x86_64 openssh openssh-askpass openssh-clients openssh-server wget bind-utils disable selinux reboot setup ssh keys Using instructions now: yum groupinstall Desktop (gnome) rpm -Uhv http://download.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-5.noarch.rpm yum update -- nothing to update rpm -Uhv http://mplug.org/~k12linux/rpm/el6/x86_64/k12linux-release-5.2.16-1.el6.x86_64.rpm yum install ltsp-server echo "/opt/ltsp *(ro,async,no_root_squash)" >> /etc/exports chkconfig network on for service in xinetd ltsp-dhcpd rpcbind nfs sshd; do chkconfig $service on; service $service restart; done for server in ldminfod nbdrootd nbdswapd tftp; do chkconfig $server on; done ltsp-build-client vi /etc/sysconfig/network-scripts/ifcfg-eth1 (ltsp interface) add BRIDGE=ltspbr0, change onboot to yes edited /etc/sysconfig/iptables and allowed all from eth1 interface ltsp-server-tweaks mkdir /opt/ltsp/kiosk-home useradd -d /opt/ltsp/kiosk-home/ws504 ws504 groupadd kiosk usermod -a -G kiosk ws504 usermod -a -G fuse ws504 edit /etc/ltsp/dhcpd.conf add in group for thin clients, mac addy and static ip /etc/init.d/ltsp-dhcpd restart Client will PXE boot and hit TFTP properly but my old thin clients don't like the kernel so I need to work on building a custom one for the chroot. I have a mini-lab setup and can also test with a PXE booted PC if anyone needs something specific tested etc. On 8/11/11 11:33 AM, Jim Kinney wrote: > Must be a temporary thing. I can hit it just fine from a firefox browser on F15. > > On Thu, Aug 11, 2011 at 1:44 PM, Dean Jones > wrote: > > Hi, > > I am working on a fresh install for feedback following the instructions here: > https://fedorahosted.org/__k12linux/wiki/InstallGuide > > > But I can't connect to the mplug.org repo from any of my > centos hosts (old or new)? I just get a time out trying to hit port 80. > Browser, wget, telnet port 80 all behave the same. > > Works fine from a mac?! > > > On 8/11/11 2:05 AM, Warren Togami Jr. wrote: > > Hey folks, > > http://k12linux.org > I just spent a few hours reorganizing the K12Linux site, hiding most of > the old and obsolete pages and putting the important stuff up front. I > still plan on doing more reorganizing and writing. > > I have strong confidence in the code itself, but the documentation is > probably too confusing and requires improvement. > > For now I really need people who are already familiar with LTSP to test > the instructions AS WRITTEN with a fresh install and to provide feedback > here on K12OSN list. After some verification and tweaking of the > documentation, we will be ready for the formal announcement where we > invite the wider community to K12Linux EL6. > > https://www.facebook.com/__k12linux.org > > Follow K12Linux on Facebook! > > Warren > > _________________________________________________ > 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 > > > > > -- > -- > James P. Kinney III > > As long as the general population is passive, apathetic, diverted to consumerism > or hatred of the vulnerable, then the powerful can do as they please, and those > who survive will be left to contemplate the outcome. > - ////2011 Noam Chomsky > > http://heretothereideas.blogspot.com/ > //// From warren at togami.com Thu Aug 11 19:57:06 2011 From: warren at togami.com (Warren Togami Jr.) Date: Thu, 11 Aug 2011 09:57:06 -1000 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E4430AA.1020303@oregonstate.edu> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> Message-ID: <4E443412.2030200@togami.com> On 8/11/2011 9:42 AM, Dean Jones wrote: > Client will PXE boot and hit TFTP properly but my old thin clients don't > like the kernel so I need to work on building a custom one for the chroot. How old are the clients? What hardware specifications? Since the /opt/ltsp/i386 is based on EL6 which is i686, it will be impossible to make anything i586 work with only a kernel rebuild. Warren From news at siddall.name Thu Aug 11 20:20:24 2011 From: news at siddall.name (Jeff Siddall) Date: Thu, 11 Aug 2011 16:20:24 -0400 Subject: [K12OSN] Flash sound gone? SOLVED In-Reply-To: <4E43F55F.5080601@siddall.name> References: <4E42A880.7040307@siddall.name> <4E4322FB.4000509@togami.com> <4E433C79.4040203@siddall.name> <4E43652E.8070602@togami.com> <4E43F55F.5080601@siddall.name> Message-ID: <4E443988.40109@siddall.name> On 08/11/2011 11:29 AM, Jeff Siddall wrote: > Interesting. I guess alsa is mucked up somehow. Alsa defaults do _not_ route alsa clients through PA even if PA has grabbed PCM device. Would be nice if it checked for that. Anyway, the system installed with an /etc/asound.conf file (which is not owned by any package BTW) which somehow got removed. I restored it from a backup and that got alsa clients working again. In case anyone cares, here is what asound.conf contains: # # Place your global alsa-lib configuration here... # @hooks [ { func load files [ "/etc/alsa/pulse-default.conf" ] errors false } ] and /etc/alsa/pulse-default.conf contains: # PulseAudio plugin configuration # $Id: pulse-default.conf,v 1.3 2008/03/09 15:50:49 lkundrak Exp $ # Let's create a virtual device "pulse" for mixer and PCM pcm.pulse { type pulse hint { description "PulseAudio Sound Server" } } ctl.pulse { type pulse hint { description "PulseAudio Sound Server" } } # Let's make it the default! pcm.!default { type pulse hint { description "Default" } } ctl.!default { type pulse hint { description "Default" } } From warren at togami.com Thu Aug 11 21:27:09 2011 From: warren at togami.com (Warren Togami Jr.) Date: Thu, 11 Aug 2011 11:27:09 -1000 Subject: [K12OSN] Flash sound gone? SOLVED In-Reply-To: <4E443988.40109@siddall.name> References: <4E42A880.7040307@siddall.name> <4E4322FB.4000509@togami.com> <4E433C79.4040203@siddall.name> <4E43652E.8070602@togami.com> <4E43F55F.5080601@siddall.name> <4E443988.40109@siddall.name> Message-ID: <4E44492D.9040801@togami.com> On 8/11/2011 10:20 AM, Jeff Siddall wrote: > On 08/11/2011 11:29 AM, Jeff Siddall wrote: >> Interesting. I guess alsa is mucked up somehow. > > Alsa defaults do _not_ route alsa clients through PA even if PA has > grabbed PCM device. Would be nice if it checked for that. > Are you sure you didn't change some defaults? This has *just worked* for years. The alsa-plugins-pulseaudio package makes a fake ALSA device as "default", which routes ALSA output through pulseaudio before directing it out the real ALSA. You must have alsa-plugins-pulseaudio installed for all architectures, so if you are running both 32bit and 64bit applications you need both .i686 and .x86_64 packages installed in parallel. Warren From news at siddall.name Thu Aug 11 22:19:29 2011 From: news at siddall.name (Jeff Siddall) Date: Thu, 11 Aug 2011 18:19:29 -0400 Subject: [K12OSN] SL 6.1 upgrade In-Reply-To: References: <4E440C7D.7030305@siddall.name> Message-ID: <4E445571.901@siddall.name> On 08/11/2011 01:37 PM, Jim Kinney wrote: > Very good to know! > > On Thu, Aug 11, 2011 at 1:08 PM, Jeff Siddall > wrote: > > For anyone who got lazy with how CentOS automatically updates to the > latest dot release, here's the command to get a specific dot release > in SL (6.1 in this case): > > yum --releasever=6.1 update I also stumbled across: yum --releasever=6rolling update which may result in always getting the most recent dot update (ie: CentOS like). Jeff From news at siddall.name Thu Aug 11 22:24:14 2011 From: news at siddall.name (Jeff Siddall) Date: Thu, 11 Aug 2011 18:24:14 -0400 Subject: [K12OSN] Flash sound gone? SOLVED In-Reply-To: <4E44492D.9040801@togami.com> References: <4E42A880.7040307@siddall.name> <4E4322FB.4000509@togami.com> <4E433C79.4040203@siddall.name> <4E43652E.8070602@togami.com> <4E43F55F.5080601@siddall.name> <4E443988.40109@siddall.name> <4E44492D.9040801@togami.com> Message-ID: <4E44568E.3090905@siddall.name> On 08/11/2011 05:27 PM, Warren Togami Jr. wrote: > On 8/11/2011 10:20 AM, Jeff Siddall wrote: >> On 08/11/2011 11:29 AM, Jeff Siddall wrote: >>> Interesting. I guess alsa is mucked up somehow. >> >> Alsa defaults do _not_ route alsa clients through PA even if PA has >> grabbed PCM device. Would be nice if it checked for that. >> > > Are you sure you didn't change some defaults? This has *just worked* for > years. By "defaults" I meant with no alsa config files. With no config files alsa is entirely unaware of pulse and doesn't do the right thing (as I found out), which is a bit disappointing. If you have an /etc/asound.conf then alsa is not running with "defaults". Jeff From dean.jones at oregonstate.edu Thu Aug 11 23:14:09 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Thu, 11 Aug 2011 16:14:09 -0700 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E4430AA.1020303@oregonstate.edu> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> Message-ID: <4E446241.30604@oregonstate.edu> On 8/11/11 12:42 PM, Dean Jones wrote: > Hah yeah, it started working 5 minutes after I sent this. No idea what > the issue was. > > Here are my test notes: > > From these instructions: > https://fedorahosted.org/k12linux/wiki/InstallGuide > > CentOS 6 minimal install. x86_64, netinstall. > Target clients are Thin clients in kiosk mode (firefox only). (old > hardware, slow video, no fuse etc) > > Not really related to LTSP steps before starting: > yum update > yum install libssh2.x86_64 openssh openssh-askpass openssh-clients > openssh-server wget bind-utils > disable selinux > reboot > setup ssh keys > > Using instructions now: > yum groupinstall Desktop (gnome) > rpm -Uhv > http://download.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-5.noarch.rpm > yum update -- nothing to update > rpm -Uhv > http://mplug.org/~k12linux/rpm/el6/x86_64/k12linux-release-5.2.16-1.el6.x86_64.rpm > > yum install ltsp-server > echo "/opt/ltsp *(ro,async,no_root_squash)">> /etc/exports > chkconfig network on > for service in xinetd ltsp-dhcpd rpcbind nfs sshd; do chkconfig $service > on; service $service restart; done > for server in ldminfod nbdrootd nbdswapd tftp; do chkconfig $server on; done > ltsp-build-client > vi /etc/sysconfig/network-scripts/ifcfg-eth1 (ltsp interface) > add BRIDGE=ltspbr0, change onboot to yes > edited /etc/sysconfig/iptables and allowed all from eth1 interface > ltsp-server-tweaks > mkdir /opt/ltsp/kiosk-home > useradd -d /opt/ltsp/kiosk-home/ws504 ws504 > groupadd kiosk > usermod -a -G kiosk ws504 > usermod -a -G fuse ws504 > edit /etc/ltsp/dhcpd.conf > add in group for thin clients, mac addy and static ip > /etc/init.d/ltsp-dhcpd restart > > Client will PXE boot and hit TFTP properly but my old thin clients don't > like the kernel so I need to work on building a custom one for the chroot. > > I have a mini-lab setup and can also test with a PXE booted PC if anyone > needs something specific tested etc. > Replying to myself to update. Using Dell Optiplex 745 as PXE client PXE Boots just fine into Scientific Linux 6 Can log in, gnome desktop, nice and fast. USB stick automounts, read/write ok can not unmount: error: Unable to unmount usbdisk-sb1 is not in the fstab (and you are not root) root can umount of course user is in fuse group does package ltspfsd need to be installed? Does not seem to change the behavior if it is flash works with adobes new 64 bit plugin. sound works fine and volume changer in panel works. flash video very fast and perfectly usable. slow boot ata2.00 link is slow to respond, please be paitent. time out. tries 3x same message limit SATA link to 1.5g speed - then boots. NFS statd service fails to start mounting local file systems registers as failed? all those errors hiddne by Scientific linux splash screen though works fine both if specified or not in the /etc/ltsp/dhcpd.conf file If anyone has specific test needs let me know as i'll keep the test systems around for a few days. From warren at togami.com Fri Aug 12 03:21:37 2011 From: warren at togami.com (Warren Togami Jr.) Date: Thu, 11 Aug 2011 17:21:37 -1000 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E446241.30604@oregonstate.edu> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> <4E446241.30604@oregonstate.edu> Message-ID: <4E449C41.1060603@togami.com> On 8/11/2011 1:14 PM, Dean Jones wrote: > PXE Boots just fine into Scientific Linux 6 > Can log in, gnome desktop, nice and fast. > USB stick automounts, read/write ok > can not unmount: > error: Unable to unmount usbdisk-sb1 > is not in the fstab (and you are not root) > root can umount of course > user is in fuse group > does package ltspfsd need to be installed? > Does not seem to change the behavior if it is https://fedorahosted.org/k12linux/wiki/EL6Status#LOWPRIORITYTODO Nautilus in general is completely unaware of user unmountable fuse mounts. I tried several approaches to fix this but failed. I may need to just do an ugly hack patching nautilus itself to dumbly run "fusermount -u /path/to/mountpoint" and it will just magically work. I didn't want to replace any upstream packages though, as this creates a long-term maintenance burden. =( > slow boot > ata2.00 link is slow to respond, please be paitent. time out. tries 3x > same message > limit SATA link to 1.5g speed - then boots. Likely it is trying to workaround a bad controller behavior and there is nothing we can do about this. Can you disable the SATA controller in the BIOS? Does that help? > NFS statd service fails to start > mounting local file systems registers as failed? > all those errors hiddne by Scientific linux splash screen though Only cosmetic, I want to fix this eventually. Warren From warren at togami.com Fri Aug 12 03:30:07 2011 From: warren at togami.com (Warren Togami Jr.) Date: Thu, 11 Aug 2011 17:30:07 -1000 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E449C41.1060603@togami.com> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> <4E446241.30604@oregonstate.edu> <4E449C41.1060603@togami.com> Message-ID: <4E449E3F.7030109@togami.com> On 8/11/2011 5:21 PM, Warren Togami Jr. wrote: > https://fedorahosted.org/k12linux/wiki/EL6Status#LOWPRIORITYTODO > Nautilus in general is completely unaware of user unmountable fuse > mounts. I tried several approaches to fix this but failed. I may need to > just do an ugly hack patching nautilus itself to dumbly run "fusermount > -u /path/to/mountpoint" and it will just magically work. I didn't want > to replace any upstream packages though, as this creates a long-term > maintenance burden. =( NOTE: It is perfectly safe to unplug the USB stick at any time it isn't writing. Yes, this is unusual, and unique to ltspfs. Warren From warren at togami.com Fri Aug 12 06:52:57 2011 From: warren at togami.com (Warren Togami Jr.) Date: Thu, 11 Aug 2011 20:52:57 -1000 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E4430AA.1020303@oregonstate.edu> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> Message-ID: <4E44CDC9.3030508@togami.com> On 8/11/2011 9:42 AM, Dean Jones wrote: > Client will PXE boot and hit TFTP properly but my old thin clients don't > like the kernel so I need to work on building a custom one for the chroot. I created a solution for you. Upgrade to ltsp-server-5.2.17 then run: ltsp-build-client --release 11 You will have a Fedora 11 i586 chroot that is a bit old, but much newer than RHEL5 which was based on Fedora 6. Does this work for your old thin clients? Warren From warren at togami.com Fri Aug 12 07:29:58 2011 From: warren at togami.com (Warren Togami Jr.) Date: Thu, 11 Aug 2011 21:29:58 -1000 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E44CDC9.3030508@togami.com> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> <4E44CDC9.3030508@togami.com> Message-ID: <4E44D676.1010308@togami.com> On 8/11/2011 8:52 PM, Warren Togami Jr. wrote: > On 8/11/2011 9:42 AM, Dean Jones wrote: >> Client will PXE boot and hit TFTP properly but my old thin clients don't >> like the kernel so I need to work on building a custom one for the >> chroot. > > I created a solution for you. > > Upgrade to ltsp-server-5.2.17 then run: > ltsp-build-client --release 11 > > You will have a Fedora 11 i586 chroot that is a bit old, but much newer > than RHEL5 which was based on Fedora 6. > > Does this work for your old thin clients? > > Warren https://fedorahosted.org/k12linux/wiki/InstallGuide I updated this page to explain in detail the three /opt/ltsp/i386 chroot install options. Dean, you have some kind of deployment with multiple parallel /opt/ltsp chroots? Did you make a custom dhcpd.conf for this? It would help the community if you could provide examples for this page for those who want parallel chroots for different clients. Warren From warren at togami.com Fri Aug 12 07:53:07 2011 From: warren at togami.com (Warren Togami Jr.) Date: Thu, 11 Aug 2011 21:53:07 -1000 Subject: [K12OSN] Test NBD Root? Message-ID: <4E44DBE3.2010204@togami.com> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration Could someone please test if the NBD Root boot documentation still works? Warren From dean.jones at oregonstate.edu Fri Aug 12 15:35:40 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Fri, 12 Aug 2011 08:35:40 -0700 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E449C41.1060603@togami.com> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> <4E446241.30604@oregonstate.edu> <4E449C41.1060603@togami.com> Message-ID: <4E45484C.7030004@oregonstate.edu> On 8/11/11 8:21 PM, Warren Togami Jr. wrote: >> slow boot >> ata2.00 link is slow to respond, please be paitent. time out. tries 3x >> same message >> limit SATA link to 1.5g speed - then boots. > > Likely it is trying to workaround a bad controller behavior and there is > nothing we can do about this. Can you disable the SATA controller in > the BIOS? Does that help? Still errors when disabled (slightly different error but same result) and I have tried all of the BIOS settings in various combinations. Just this particular controller probably. It is hidden by the splash screen and only noticeable by a delay. No biggie, i'll test some other stuff. From news at siddall.name Fri Aug 12 16:51:53 2011 From: news at siddall.name (Jeff Siddall) Date: Fri, 12 Aug 2011 12:51:53 -0400 Subject: [K12OSN] SL 6.1 upgrade In-Reply-To: <4E445571.901@siddall.name> References: <4E440C7D.7030305@siddall.name> <4E445571.901@siddall.name> Message-ID: <4E455A29.1060601@siddall.name> On 08/11/2011 06:19 PM, Jeff Siddall wrote: > I also stumbled across: > > yum --releasever=6rolling update > > which may result in always getting the most recent dot update (ie: > CentOS like). Correction: 6rolling is actually the equivalent of Fedora rawhide -- the latest of everything even if it is not stable. To always get the latest dot releases instead run: yum install yum-conf-sl6x Jeff From news at siddall.name Fri Aug 12 18:12:56 2011 From: news at siddall.name (Jeff Siddall) Date: Fri, 12 Aug 2011 14:12:56 -0400 Subject: [K12OSN] Why PA system mode? Message-ID: <4E456D28.1020904@siddall.name> Anyone know why the clients run pulseaudio in system mode? Here's an excerpt from the client system log: Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: OK, so you are running PA in system mode. Please note that you most likely shouldn't be doing that. Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: If you do it nonetheless then it's your own fault if things don't work as expected. Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: Please read http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode for an explanation why system mode is usually a bad idea I did read http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode and sure enough they say it is there for stuff like thin clients. However, the reasons for it are really not that good: we _do_ have a real user, and things like home directories typically _aren't_ flushed on each session termination. Further, since PA is really designed to be run as a user, could we remove some headache by not having it tun in system mode (ie: not have to fixup permissions etc.)? The reason I am interested is that it appears system mode causes problems with some pulse stuff like pulseaudio-equalizer. Thanks, Jeff From burke at thealmquists.net Fri Aug 12 18:15:28 2011 From: burke at thealmquists.net (Burke Almquist) Date: Fri, 12 Aug 2011 13:15:28 -0500 Subject: [K12OSN] Test NBD Root? In-Reply-To: <4E44DBE3.2010204@togami.com> References: <4E44DBE3.2010204@togami.com> Message-ID: <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: > https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration > > Could someone please test if the NBD Root boot documentation still works? > > Warren > My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on a fresh install of SL6.1. Even the network setup seemed to work fine for me. When I enable NBD root instead of NFS, it gets to the login screen fine. But when I try to log in, it keeps going back to the login screen again. Does NBD require more memory to work? > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From dahopkins429 at gmail.com Fri Aug 12 19:25:29 2011 From: dahopkins429 at gmail.com (David Hopkins) Date: Fri, 12 Aug 2011 15:25:29 -0400 Subject: [K12OSN] OT: Authentication server options Message-ID: A new charter school is opening this fall and has asked us (Newark Charter) to help set up their IT infrastructure. At NCS we used the samba+ldap roll our own approach for authentication. It has worked quite well. However, I'm considering using Active Directory + winbind instead but I'm not sure of all the pros/cons of this approach. Will it be transparent for logging onto thin clients? There will be a dedicated linux-based (CentOS) file server for home directories. Will home directories work as expected for the linux users? I appreciate any insights and recommendations on this. Sincerely, Dave Hopkins Newark Charter School From dean.jones at oregonstate.edu Fri Aug 12 19:36:10 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Fri, 12 Aug 2011 12:36:10 -0700 Subject: [K12OSN] Test NBD Root? In-Reply-To: <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> Message-ID: <4E4580AA.9060901@oregonstate.edu> On 8/12/11 11:15 AM, Burke Almquist wrote: > > On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: > >> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >> >> Could someone please test if the NBD Root boot documentation still >> works? >> >> Warren >> > > My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on > a fresh install of SL6.1. Even the network setup seemed to work fine > for me. When I enable NBD root instead of NFS, it gets to the login > screen fine. But when I try to log in, it keeps going back to the > login screen again. Does NBD require more memory to work? It is working fine for me with a box with 2gigs ram. I just followed the steps I posted yesterday for a base thin client on NFS, then followed the NBD page and it is working. Maybe it is a mem issue? But that sounds like it can't mount the user home dir or something to me. > > > >> _______________________________________________ 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 dean.jones at oregonstate.edu Fri Aug 12 20:24:07 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Fri, 12 Aug 2011 13:24:07 -0700 Subject: [K12OSN] Test NBD Root? In-Reply-To: <4E4580AA.9060901@oregonstate.edu> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4580AA.9060901@oregonstate.edu> Message-ID: <4E458BE7.90508@oregonstate.edu> On 8/12/11 12:36 PM, Dean Jones wrote: > On 8/12/11 11:15 AM, Burke Almquist wrote: >> >> On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: >> >>> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >>> >>> Could someone please test if the NBD Root boot documentation still >>> works? >>> >>> Warren >>> >> >> My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on >> a fresh install of SL6.1. Even the network setup seemed to work fine >> for me. When I enable NBD root instead of NFS, it gets to the login >> screen fine. But when I try to log in, it keeps going back to the >> login screen again. Does NBD require more memory to work? > > It is working fine for me with a box with 2gigs ram. I just followed > the steps I posted yesterday for a base thin client on NFS, then > followed the NBD page and it is working. > > Maybe it is a mem issue? But that sounds like it can't mount the user > home dir or something to me. > I retract my statement. I did a clean install (to clear out all the random things i did but didn't take notes about), basic thin client NFS is working fine. Add in the NBD via the instructions on the official page and the results are just as you describe. Can't login with any valid user, right back to the login screen. ldminfod is the only log output i can see from an attempted login: Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5569 from=::ffff:172.31.100.74 Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5569 duration=0(sec) Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5572 from=::ffff:172.31.100.74 Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5572 duration=0(sec) It tries twice? It also doesn't boot right to the GUI login, but instead to the message console, and ALT-F7 jumps to the GUI. The rootfs looks to be mounted: Aug 12 13:16:33 hostname123 xinetd[5801]: START: nbdrootd pid=5822 from=::ffff:172.31.100.74 Aug 12 13:16:33 hostname123 nbd-server: connect from 172.31.100.74, assigned file is /opt/ltsp/images/i386.img Aug 12 13:16:33 hostname123 nbd-server: Size of exported file/device is 232845312 From dean.jones at oregonstate.edu Fri Aug 12 21:01:36 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Fri, 12 Aug 2011 14:01:36 -0700 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E44CDC9.3030508@togami.com> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> <4E44CDC9.3030508@togami.com> Message-ID: <4E4594B0.5060206@oregonstate.edu> On 8/11/11 11:52 PM, Warren Togami Jr. wrote: > On 8/11/2011 9:42 AM, Dean Jones wrote: >> Client will PXE boot and hit TFTP properly but my old thin clients don't >> like the kernel so I need to work on building a custom one for the chroot. > > I created a solution for you. > > Upgrade to ltsp-server-5.2.17 then run: > ltsp-build-client --release 11 > > You will have a Fedora 11 i586 chroot that is a bit old, but much newer > than RHEL5 which was based on Fedora 6. > > Does this work for your old thin clients? > Hi, Wow thanks! This works great for the old thin client hardware. Works with NFS or NBD root FS, and everything in Gnome seems happy. Even flash and sound works but the systems are so old that they drop lots of frames. But that is to be expected! From jim.kinney at gmail.com Fri Aug 12 21:47:19 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Fri, 12 Aug 2011 17:47:19 -0400 Subject: [K12OSN] OT: Authentication server options In-Reply-To: References: Message-ID: I would use Fedora 389 server for LDAP (supports multi-master for automatic updates and failover by having ldap.conf list dual servers!) What is used by the system for user auth should have no impact on home directories unless the source of user auth has different data for the home dir. On Fri, Aug 12, 2011 at 3:25 PM, David Hopkins wrote: > A new charter school is opening this fall and has asked us (Newark > Charter) to help set up their IT infrastructure. At NCS we used the > samba+ldap roll our own approach for authentication. It has worked > quite well. However, I'm considering using Active Directory + winbind > instead but I'm not sure of all the pros/cons of this approach. Will > it be transparent for logging onto thin clients? There will be a > dedicated linux-based (CentOS) file server for home directories. Will > home directories work as expected for the linux users? > > I appreciate any insights and recommendations on this. > > 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 As long as the general population is passive, apathetic, diverted to consumerism or hatred of the vulnerable, then the powerful can do as they please, and those who survive will be left to contemplate the outcome. - *2011 Noam Chomsky http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From dahopkins429 at gmail.com Fri Aug 12 22:41:30 2011 From: dahopkins429 at gmail.com (David Hopkins) Date: Fri, 12 Aug 2011 18:41:30 -0400 Subject: [K12OSN] OT: Authentication server options In-Reply-To: References: Message-ID: Thanks! I've looked at and wanted to try Fedora 389 but I am concerned about the support for Samba/Windows Domain. The two links I've found describing how to do this are: http://directory.fedoraproject.org/wiki/Howto:Samba and http://wiki.neddix.com/How_to_setup_the_CentOS_Directory_Server_%28389_Directory_Server%29 I guess I'll just have to try a few times and see if I can get it working. Thanks! Dave On Fri, Aug 12, 2011 at 5:47 PM, Jim Kinney wrote: > I would use Fedora 389 server for LDAP (supports multi-master for automatic > updates and failover by having ldap.conf list dual servers!) > > What is used by the system for user auth should have no impact on home > directories unless the source of user auth has different data for the home > dir. > > On Fri, Aug 12, 2011 at 3:25 PM, David Hopkins > wrote: >> >> A new charter school is opening this fall and has asked us (Newark >> Charter) to help set up their IT infrastructure. At NCS we used the >> samba+ldap roll our own approach for authentication. It has worked >> quite well. However, I'm considering using Active Directory + winbind >> instead but I'm not sure of all the pros/cons of this approach. Will >> it be transparent for logging onto thin clients? There will be a >> dedicated linux-based (CentOS) file server for home directories. Will >> home directories work as expected for the linux users? >> >> I appreciate any insights and recommendations on this. >> >> 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 > > As long as the general population is passive, apathetic, diverted to > consumerism or hatred of the vulnerable, then the powerful can do as they > please, and those who survive will be left to contemplate the outcome. > - 2011 Noam Chomsky > > http://heretothereideas.blogspot.com/ > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From pxeboot at gmail.com Fri Aug 12 22:52:20 2011 From: pxeboot at gmail.com (Conrad Lawes) Date: Fri, 12 Aug 2011 22:52:20 +0000 Subject: [K12OSN] OT: Authentication server options In-Reply-To: References: Message-ID: <194220993-1313189542-cardhu_decombobulator_blackberry.rim.net-523768446-@b11.c8.bise6.blackberry> If you are seeking AD integration then I strongly recommend Likewise-Open or Centrify DirectControl Sent on the TELUS Mobility network with BlackBerry -----Original Message----- From: David Hopkins Sender: k12osn-bounces at redhat.com Date: Fri, 12 Aug 2011 18:41:30 To: Support list for open source software in schools. Reply-To: "Support list for open source software in schools." Subject: Re: [K12OSN] OT: Authentication server options Thanks! I've looked at and wanted to try Fedora 389 but I am concerned about the support for Samba/Windows Domain. The two links I've found describing how to do this are: http://directory.fedoraproject.org/wiki/Howto:Samba and http://wiki.neddix.com/How_to_setup_the_CentOS_Directory_Server_%28389_Directory_Server%29 I guess I'll just have to try a few times and see if I can get it working. Thanks! Dave On Fri, Aug 12, 2011 at 5:47 PM, Jim Kinney wrote: > I would use Fedora 389 server for LDAP (supports multi-master for automatic > updates and failover by having ldap.conf list dual servers!) > > What is used by the system for user auth should have no impact on home > directories unless the source of user auth has different data for the home > dir. > > On Fri, Aug 12, 2011 at 3:25 PM, David Hopkins > wrote: >> >> A new charter school is opening this fall and has asked us (Newark >> Charter) to help set up their IT infrastructure. At NCS we used the >> samba+ldap roll our own approach for authentication. It has worked >> quite well. However, I'm considering using Active Directory + winbind >> instead but I'm not sure of all the pros/cons of this approach. Will >> it be transparent for logging onto thin clients? There will be a >> dedicated linux-based (CentOS) file server for home directories. Will >> home directories work as expected for the linux users? >> >> I appreciate any insights and recommendations on this. >> >> 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 > > As long as the general population is passive, apathetic, diverted to > consumerism or hatred of the vulnerable, then the powerful can do as they > please, and those who survive will be left to contemplate the outcome. > - 2011 Noam Chomsky > > http://heretothereideas.blogspot.com/ > > > _______________________________________________ > 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 dean.jones at oregonstate.edu Fri Aug 12 23:13:42 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Fri, 12 Aug 2011 16:13:42 -0700 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E44D676.1010308@togami.com> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> <4E44CDC9.3030508@togami.com> <4E44D676.1010308@togami.com> Message-ID: <4E45B3A6.7020306@oregonstate.edu> On 8/12/11 12:29 AM, Warren Togami Jr. wrote: > On 8/11/2011 8:52 PM, Warren Togami Jr. wrote: >> On 8/11/2011 9:42 AM, Dean Jones wrote: >>> Client will PXE boot and hit TFTP properly but my old thin clients don't >>> like the kernel so I need to work on building a custom one for the >>> chroot. >> >> I created a solution for you. >> >> Upgrade to ltsp-server-5.2.17 then run: >> ltsp-build-client --release 11 >> >> You will have a Fedora 11 i586 chroot that is a bit old, but much newer >> than RHEL5 which was based on Fedora 6. >> >> Does this work for your old thin clients? >> >> Warren > > https://fedorahosted.org/k12linux/wiki/InstallGuide > I updated this page to explain in detail the three /opt/ltsp/i386 chroot > install options. > > Dean, you have some kind of deployment with multiple parallel /opt/ltsp > chroots? Did you make a custom dhcpd.conf for this? It would help the > community if you could provide examples for this page for those who want > parallel chroots for different clients. > Sure I have a couple of different chroots going now. The standard EL6 release plus the release 11 for the old hardware, and some old scary legacy ones. For each chroot, the first step is to do the client build command but specify a chroot name other than the default. Or in my case i left the EL6 release in the standard i386 location and release 11 in a specifically named one. ltsp-build-client (build the standard chroot) ltsp-build-client --chroot i386-thin --release 11 (non-standard) In the /etc/ltsp/dhcpd.conf file I have the standard release as the default. So i uncomment or create a group entry at the bottom of the config and specify non-standard hosts by mac address. For each host set "option root-path" to point to the proper nfs or nbd server and "filename" to the proper tftpboot tree. host ws504 { hardware ethernet 00:50:41:03:9F:6B; fixed-address 172.31.100.74; # for NBD option root-path "nbd:172.31.100.254:2001:squashfs:ro"; filename "/ltsp/i386-thin/pxelinux.0"; # for NFS # option root-path "172.31.100.254:/opt/ltsp/i386-thin"; } for additional NBD image serving: cp /etc/xinetd.d/nbdroot /etc/xinetd.d/nbdroot2 and edit nbdroot2 to set it to a different port (2001?) and point to the path of the second chroot image, in my case server_args = /opt/ltsp/images/i386-thin.img /etc/init.d/ltsp-dhcpd restart /etc/init.d/xinetd restart ltsp-update-sshkeys ltsp-update-kernels ltsp-update-image then I also had to: rsync -avu /var/lib/tftpboot/ltsp/i386/pxe* /var/lib/tftpboot/ltsp/i386-thin/ Hopefully this is what you are looking for. From rowens at ptd.net Sat Aug 13 13:04:15 2011 From: rowens at ptd.net (Rob Owens) Date: Sat, 13 Aug 2011 09:04:15 -0400 Subject: [K12OSN] HELP NEEDED NOW: Need Documentation Testing and Feedback In-Reply-To: <4E45B3A6.7020306@oregonstate.edu> References: <4E439B3D.3040208@togami.com> <4E4414E6.50404@oregonstate.edu> <4E4430AA.1020303@oregonstate.edu> <4E44CDC9.3030508@togami.com> <4E44D676.1010308@togami.com> <4E45B3A6.7020306@oregonstate.edu> Message-ID: <20110813130415.GA9164@aurora.owens.net> > On 8/12/11 12:29 AM, Warren Togami Jr. wrote: > >Dean, you have some kind of deployment with multiple parallel /opt/ltsp > >chroots? Did you make a custom dhcpd.conf for this? It would help the > >community if you could provide examples for this page for those who want > >parallel chroots for different clients. > > I did something similar a while back and documented it on the LTSP wiki. I needed to run LTSP 5 and LTSP 4.2 on the same server. You can check out my dhcpd.conf modifications and ignore the LTSP 4.2 stuff if that's what you need: http://sourceforge.net/apps/mediawiki/ltsp/index.php?title=Ltsp5SameServerLTSP42 -Rob From burke at thealmquists.net Sat Aug 13 15:58:27 2011 From: burke at thealmquists.net (Burke Almquist) Date: Sat, 13 Aug 2011 10:58:27 -0500 Subject: [K12OSN] Test NBD Root? In-Reply-To: <4E458BE7.90508@oregonstate.edu> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4580AA.9060901@oregonstate.edu> <4E458BE7.90508@oregonstate.edu> Message-ID: <0BEAD13A-6A29-4EF1-BCC4-128EF9306906@thealmquists.net> On Aug 12, 2011, at 3:24 PM, Dean Jones wrote: > On 8/12/11 12:36 PM, Dean Jones wrote: >> On 8/12/11 11:15 AM, Burke Almquist wrote: >>> >>> On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: >>> >>>> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >>>> >>>> Could someone please test if the NBD Root boot documentation still >>>> works? >>>> >>>> Warren >>>> >>> >>> My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on >>> a fresh install of SL6.1. Even the network setup seemed to work fine >>> for me. When I enable NBD root instead of NFS, it gets to the login >>> screen fine. But when I try to log in, it keeps going back to the >>> login screen again. Does NBD require more memory to work? >> >> It is working fine for me with a box with 2gigs ram. I just followed >> the steps I posted yesterday for a base thin client on NFS, then >> followed the NBD page and it is working. >> >> Maybe it is a mem issue? But that sounds like it can't mount the user >> home dir or something to me. >> > > I retract my statement. > > I did a clean install (to clear out all the random things i did but didn't take notes about), basic thin client NFS is working fine. > > Add in the NBD via the instructions on the official page and the results are just as you describe. Can't login with any valid user, right back to the login screen. > > ldminfod is the only log output i can see from an attempted login: > > Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5569 from=::ffff:172.31.100.74 > Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5569 duration=0(sec) > Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5572 from=::ffff:172.31.100.74 > Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5572 duration=0(sec) > > It tries twice? > > It also doesn't boot right to the GUI login, but instead to the message console, and ALT-F7 jumps to the GUI. > > The rootfs looks to be mounted: > Aug 12 13:16:33 hostname123 xinetd[5801]: START: nbdrootd pid=5822 from=::ffff:172.31.100.74 > Aug 12 13:16:33 hostname123 nbd-server: connect from 172.31.100.74, assigned file is /opt/ltsp/images/i386.img > Aug 12 13:16:33 hostname123 nbd-server: Size of exported file/device is 232845312 If your client with 2GB RAM isn't working, then it's probably not because it needs more memory. From dahopkins429 at gmail.com Sat Aug 13 22:29:31 2011 From: dahopkins429 at gmail.com (David Hopkins) Date: Sat, 13 Aug 2011 18:29:31 -0400 Subject: [K12OSN] Test NBD Root? In-Reply-To: <0BEAD13A-6A29-4EF1-BCC4-128EF9306906@thealmquists.net> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4580AA.9060901@oregonstate.edu> <4E458BE7.90508@oregonstate.edu> <0BEAD13A-6A29-4EF1-BCC4-128EF9306906@thealmquists.net> Message-ID: On Sat, Aug 13, 2011 at 11:58 AM, Burke Almquist wrote: > > On Aug 12, 2011, at 3:24 PM, Dean Jones wrote: > >> On 8/12/11 12:36 PM, Dean Jones wrote: >>> On 8/12/11 11:15 AM, Burke Almquist wrote: >>>> >>>> On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: >>>> >>>>> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >>>>> >>>>> Could someone please test if the NBD Root boot documentation still >>>>> works? >>>>> >>>>> Warren >>>>> >>>> >>>> My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on >>>> a fresh install of SL6.1. Even the network setup seemed to work fine >>>> for me. When I enable NBD root instead of NFS, it gets to the login >>>> screen fine. But when I try to log in, it keeps going back to the >>>> login screen again. Does NBD require more memory to work? >>> >>> It is working fine for me with a box with 2gigs ram. ?I just followed >>> the steps I posted yesterday for a base thin client on NFS, then >>> followed the NBD page and it is working. >>> >>> Maybe it is a mem issue? ?But that sounds like it can't mount the user >>> home dir or something to me. >>> >> >> I retract my statement. >> >> I did a clean install (to clear out all the random things i did but didn't take notes about), basic thin client NFS is working fine. >> >> Add in the NBD via the instructions on the official page and the results are just as you describe. ?Can't login with any valid user, ?right back to the login screen. >> >> ldminfod is the only log output i can see from an attempted login: >> >> Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5569 from=::ffff:172.31.100.74 >> Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5569 duration=0(sec) >> Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5572 from=::ffff:172.31.100.74 >> Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5572 duration=0(sec) >> >> It tries twice? >> >> It also doesn't boot right to the GUI login, but instead to the message console, and ALT-F7 jumps to the GUI. >> >> The rootfs looks to be mounted: >> Aug 12 13:16:33 hostname123 xinetd[5801]: START: nbdrootd pid=5822 from=::ffff:172.31.100.74 >> Aug 12 13:16:33 hostname123 nbd-server: connect from 172.31.100.74, assigned file is /opt/ltsp/images/i386.img >> Aug 12 13:16:33 hostname123 nbd-server: Size of exported file/device is 232845312 > > > If your client with 2GB RAM isn't working, then it's probably not because it needs more memory. > Does the client boot if you set metacity as the default/mandatory window manager instead of compiz? I think the command is: gconftool-2 --direct --configure-source xml:readwrite:/etc/gconf/gconf.xml.mandatory --type string --set /desktop/gnome/session/required_components/windowmanager metacity Sincerely, Dave Hopkins From burke at thealmquists.net Sun Aug 14 18:57:47 2011 From: burke at thealmquists.net (Burke Almquist) Date: Sun, 14 Aug 2011 13:57:47 -0500 Subject: [K12OSN] Test NBD Root? In-Reply-To: References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4580AA.9060901@oregonstate.edu> <4E458BE7.90508@oregonstate.edu> <0BEAD13A-6A29-4EF1-BCC4-128EF9306906@thealmquists.net> Message-ID: <7605C981-40BC-4CDE-9E2D-68236200D7F1@thealmquists.net> On Aug 13, 2011, at 5:29 PM, David Hopkins wrote: > On Sat, Aug 13, 2011 at 11:58 AM, Burke Almquist wrote: >> >> On Aug 12, 2011, at 3:24 PM, Dean Jones wrote: >> >>> On 8/12/11 12:36 PM, Dean Jones wrote: >>>> On 8/12/11 11:15 AM, Burke Almquist wrote: >>>>> >>>>> On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: >>>>> >>>>>> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >>>>>> >>>>>> Could someone please test if the NBD Root boot documentation still >>>>>> works? >>>>>> >>>>>> Warren >>>>>> >>>>> >>>>> My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on >>>>> a fresh install of SL6.1. Even the network setup seemed to work fine >>>>> for me. When I enable NBD root instead of NFS, it gets to the login >>>>> screen fine. But when I try to log in, it keeps going back to the >>>>> login screen again. Does NBD require more memory to work? >>>> >>>> It is working fine for me with a box with 2gigs ram. I just followed >>>> the steps I posted yesterday for a base thin client on NFS, then >>>> followed the NBD page and it is working. >>>> >>>> Maybe it is a mem issue? But that sounds like it can't mount the user >>>> home dir or something to me. >>>> >>> >>> I retract my statement. >>> >>> I did a clean install (to clear out all the random things i did but didn't take notes about), basic thin client NFS is working fine. >>> >>> Add in the NBD via the instructions on the official page and the results are just as you describe. Can't login with any valid user, right back to the login screen. >>> >>> ldminfod is the only log output i can see from an attempted login: >>> >>> Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5569 from=::ffff:172.31.100.74 >>> Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5569 duration=0(sec) >>> Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5572 from=::ffff:172.31.100.74 >>> Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5572 duration=0(sec) >>> >>> It tries twice? >>> >>> It also doesn't boot right to the GUI login, but instead to the message console, and ALT-F7 jumps to the GUI. >>> >>> The rootfs looks to be mounted: >>> Aug 12 13:16:33 hostname123 xinetd[5801]: START: nbdrootd pid=5822 from=::ffff:172.31.100.74 >>> Aug 12 13:16:33 hostname123 nbd-server: connect from 172.31.100.74, assigned file is /opt/ltsp/images/i386.img >>> Aug 12 13:16:33 hostname123 nbd-server: Size of exported file/device is 232845312 >> >> >> If your client with 2GB RAM isn't working, then it's probably not because it needs more memory. >> > Does the client boot if you set metacity as the default/mandatory > window manager instead of compiz? > > I think the command is: gconftool-2 --direct --configure-source > xml:readwrite:/etc/gconf/gconf.xml.mandatory --type string --set > /desktop/gnome/session/required_components/windowmanager metacity > > Sincerely, > Dave Hopkins If that's the case though, why does it log in ok using NFS? From burke at thealmquists.net Sun Aug 14 19:09:26 2011 From: burke at thealmquists.net (Burke Almquist) Date: Sun, 14 Aug 2011 14:09:26 -0500 Subject: [K12OSN] Why PA system mode? In-Reply-To: <4E456D28.1020904@siddall.name> References: <4E456D28.1020904@siddall.name> Message-ID: <7459551C-9981-4AF7-ACB6-064159A7ECE6@thealmquists.net> On Aug 12, 2011, at 1:12 PM, Jeff Siddall wrote: > Anyone know why the clients run pulseaudio in system mode? Here's an excerpt from the client system log: > > Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: OK, so you are running PA in system mode. Please note that you most likely shouldn't be doing that. > Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: If you do it nonetheless then it's your own fault if things don't work as expected. > Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: Please read http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode for an explanation why system mode is usually a bad idea > > I did read http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode and sure enough they say it is there for stuff like thin clients. However, the reasons for it are really not that good: we _do_ have a real user, and things like home directories typically _aren't_ flushed on each session termination. But we don't have a "real" user on the thin clients, the real user is logged into the server, not the client, so that doesn't help us, since PA needs to run on the client for it to output the sound. If you were doing DRBL, where you network boot the OS, but all the apps are actually local, then system mode doesn't make sense, but you still need it on thin clients, since the user's session is actually on a different machine. > > Further, since PA is really designed to be run as a user, could we remove some headache by not having it tun in system mode (ie: not have to fixup permissions etc.)? > > The reason I am interested is that it appears system mode causes problems with some pulse stuff like pulseaudio-equalizer. > > Thanks, > > Jeff > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From carl at snarlnet.com Sun Aug 14 21:49:13 2011 From: carl at snarlnet.com (Carl Keil) Date: Sun, 14 Aug 2011 14:49:13 -0700 Subject: [K12OSN] Need Quick Help - Any way to do a low level, complete copy of my boot HD? Message-ID: <4E4842D9.9030303@snarlnet.com> Hi, Sorry. My boot drive is apparently failing. Just got this in the mail after having to rescue my grub/boot info after waking up to a dead server. Device: /dev/hda, 51 Currently unreadable (pending) sectors The email before that said 44 Currently unreadable. So, is there a way to slap another 40 gig HD (that's how big the current one is) in there and type a command, from say the rescue CD's command line and just copy over what I have now, verbatim, to a new disk. I'm going to keep googling, but I trust this list more than most Linux info I've seen. Like copying at the /dev/hda level. Copying partitions, etc. Thanks, ck From carl at snarlnet.com Sun Aug 14 21:54:33 2011 From: carl at snarlnet.com (Carl Keil) Date: Sun, 14 Aug 2011 14:54:33 -0700 Subject: [K12OSN] Nevermind last message... Message-ID: <4E484419.7050309@snarlnet.com> Looks like dd is the way to go. Gonna try it now. Wish me luck. Thanks and sorry. I guess I was panicking. ck From warren at togami.com Sun Aug 14 22:02:24 2011 From: warren at togami.com (Warren Togami Jr.) Date: Sun, 14 Aug 2011 12:02:24 -1000 Subject: [K12OSN] Test NBD Root? In-Reply-To: <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> Message-ID: <4E4845F0.5070008@togami.com> On 8/12/2011 8:15 AM, Burke Almquist wrote: > > On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: > >> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >> >> Could someone please test if the NBD Root boot documentation still works? >> >> Warren >> > > My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on a fresh install of SL6.1. Even the network setup seemed to work fine for me. > When I enable NBD root instead of NFS, it gets to the login screen fine. But when I try to log in, it keeps going back to the login screen again. > Does NBD require more memory to work? Strange... on my system, the EL6.1 chroot with NBD boot gets stuck after: Adding udev persistent rules [ OK ] If you boot NFS the next visible thing there is starting the NBD swap, so I tried to turn off NBD swap with no apparent effect. I don't know what's going on. I will dig more later. Warren From jim.kinney at gmail.com Sun Aug 14 22:42:46 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Sun, 14 Aug 2011 18:42:46 -0400 Subject: [K12OSN] Need Quick Help - Any way to do a low level, complete copy of my boot HD? In-Reply-To: <4E4842D9.9030303@snarlnet.com> References: <4E4842D9.9030303@snarlnet.com> Message-ID: You are wanting the dd command or the ddrescue command. Dd is standard but will likely choke on failed sectors. Ddrescue will keep moving and retrying failed sectors until it suceeds or times out. On Aug 14, 2011 5:47 PM, "Carl Keil" wrote: > Hi, > > Sorry. My boot drive is apparently failing. Just got this in the mail > after having to rescue my grub/boot info after waking up to a dead server. > > Device: /dev/hda, 51 Currently unreadable (pending) sectors > > The email before that said 44 Currently unreadable. > > So, is there a way to slap another 40 gig HD (that's how big the current one is) in there and type a command, from say the rescue CD's command line and just copy over what I have now, verbatim, to a new disk. I'm going to keep googling, but I trust this list more than most Linux info I've seen. Like copying at the /dev/hda level. Copying partitions, etc. > > Thanks, > > ck > > > > _______________________________________________ > 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 k12ltsp at rwcinc.net Sun Aug 14 23:25:25 2011 From: k12ltsp at rwcinc.net (Patrick Fleming) Date: Sun, 14 Aug 2011 16:25:25 -0700 Subject: [K12OSN] Need Quick Help - Any way to do a low level, complete copy of my boot HD? In-Reply-To: <4E4842D9.9030303@snarlnet.com> References: <4E4842D9.9030303@snarlnet.com> Message-ID: <4E485965.1010707@rwcinc.net> I will second and third the dd_rescue tool. I have resurrected more than one failing drive on a variety of machines with it. On 08/14/11 14:49, Carl Keil wrote: > Hi, > > Sorry. My boot drive is apparently failing. Just got this in the mail > after having to rescue my grub/boot info after waking up to a dead server. > > Device: /dev/hda, 51 Currently unreadable (pending) sectors > > The email before that said 44 Currently unreadable. > > So, is there a way to slap another 40 gig HD (that's how big the current > one is) in there and type a command, from say the rescue CD's command > line and just copy over what I have now, verbatim, to a new disk. I'm > going to keep googling, but I trust this list more than most Linux info > I've seen. Like copying at the /dev/hda level. Copying partitions, etc. > > Thanks, > > ck > > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From cisna-barry at wc235.k12.il.us Mon Aug 15 12:54:50 2011 From: cisna-barry at wc235.k12.il.us (Barry Cisna) Date: Mon, 15 Aug 2011 07:54:50 -0500 Subject: [K12OSN] OT: Authentication server options Message-ID: <1313412890.10192.6.camel@hi2.wc235.k12.il.us> Dave, The MS AD authentication you are wanting to use,is the setup we have used at our k-12 school for 8 years now. It does in fact work fine with tested now,,,Windows 7,, Windows Vista as well. I have the Samba file server setup as an Domain member of course. As others have said I would like to try the Fedora 389 server to get MS out of the maintenance picture,,though. On my home server setup I use A Samba server as PDC and actually works better transparently for TC's and Windows TS sessions as the home folders are created somewhat more as a real looking Windows Desktop enviornment . Let us know how things go. Take Care, Barry From carl at snarlnet.com Mon Aug 15 15:28:19 2011 From: carl at snarlnet.com (Carl Keil) Date: Mon, 15 Aug 2011 08:28:19 -0700 Subject: [K12OSN] Need advice on recovering from failed boot disk Message-ID: <4E493B13.9010607@snarlnet.com> Thanks for helping me. So, I tried the dd command, twice and then tried CloneZilla. Twice. And the failing drive just kept throwing more and more errors. Even with the "move on after bad sectors" (or whatever it's called exactly) switch thrown. I think I need to give up on the miracle cure. How was I ever going to trust the copied disk anyway? Wouldn't that just be latent problems (half garbled config files, etc.) waiting to happen? Anyway, now I'm looking for a little more philosophical advice. I used to run k12ltsp on this server, but don't any more. Now it's a web/samba/mythtv box for my home. If it was you, and your old k12-centos 5.3 box ate it, but you had homes, web root, samba shares and your mythtv shows on separate drives, (and many complete backups via BackupPC) would you take the opportunity to upgrade to centos 5.6 or 6.0 or would you try to get back to 5.3 so you didn't break all that stuff via new versions of php, mysql, samba (prolly not an issue) and mythtv? What's the smart move here? Also, do you think there'll be a problem going to 64 bit, when the old install was 32 bit? I can't see why that would be an issue since basically I'm restoring functionality to content that I saved. If I go, for example from 32-bit Centos 5.3 to 64-bit Centos 6 will I still just be able to restore my /etc config files from backup and proceed on my merry way? Or will I have to go into each one and cut and paste relevant sections into new config files with differing formats, places on the drive, etc. Also, How do you tell if your hardware should get a 64 bit centos or not? I'm not positive about my system. I think it's a Pentium D dual core. Obviously I'd check for sure before proceeding. But what's considered the minimum CPU that actually works with 64 bit Centos? Thank you so much. ck > ------------------------------------------------------------------------ > > You are wanting the dd command or the ddrescue command. > Dd is standard but will likely choke on failed sectors. > Ddrescue will keep moving and retrying failed sectors until it suceeds > or times out. > > On Aug 14, 2011 5:47 PM, "Carl Keil" > wrote: > > Hi, > > > > Sorry. My boot drive is apparently failing. Just got this in the mail > > after having to rescue my grub/boot info after waking up to a dead > server. > > > > Device: /dev/hda, 51 Currently unreadable (pending) sectors > > > > The email before that said 44 Currently unreadable. > > > > So, is there a way to slap another 40 gig HD (that's how big the > current one is) in there and type a command, from say the rescue CD's > command line and just copy over what I have now, verbatim, to a new > disk. I'm going to keep googling, but I trust this list more than most > Linux info I've seen. Like copying at the /dev/hda level. Copying > partitions, etc. > > > > Thanks, > > > > ck > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From dean.jones at oregonstate.edu Mon Aug 15 15:41:11 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Mon, 15 Aug 2011 08:41:11 -0700 Subject: [K12OSN] Test NBD Root? In-Reply-To: <4E4845F0.5070008@togami.com> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4845F0.5070008@togami.com> Message-ID: <4E493E17.9030203@oregonstate.edu> On 8/14/11 3:02 PM, Warren Togami Jr. wrote: > On 8/12/2011 8:15 AM, Burke Almquist wrote: >> >> On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: >> >>> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >>> >>> Could someone please test if the NBD Root boot documentation still works? >>> >>> Warren >>> >> >> My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on a fresh install of SL6.1. Even the network setup seemed to work fine for me. >> When I enable NBD root instead of NFS, it gets to the login screen fine. But when I try to log in, it keeps going back to the login screen again. >> Does NBD require more memory to work? > > Strange... on my system, the EL6.1 chroot with NBD boot gets stuck after: > Adding udev persistent rules [ OK ] > > If you boot NFS the next visible thing there is starting the NBD swap, > so I tried to turn off NBD swap with no apparent effect. > > I don't know what's going on. I will dig more later. > > Warren > This is with the main EL6 release. With the release 11 the NBD root works fine. So these message comparisons are between those two versions. The only related messages in the logs are: /var/log/secure log errors upon login attempt: first three lines are normal for a successful login from a thin client NBD root mount. Aug 12 16:50:20 hostname123 sshd[2812]: Accepted password for ws504 from 172.31.100.11 port 44768 ssh2 Aug 12 16:50:20 hostname123 sshd[2812]: pam_unix(sshd:session): session opened for user ws504 by (uid=0) Aug 12 16:50:20 hostname123 sshd[2815]: subsystem request for sftp Aug 12 16:50:20 hostname123 sshd[2815]: Received disconnect from 172.31.100.11: 11: disconnected by user Aug 12 16:50:20 hostname123 sshd[2812]: pam_unix(sshd:session): session closed for user ws504 from /var/log/messages for the Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5569 from=::ffff:172.31.100.74 Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5569 duration=0(sec) Aug 12 13:07:51 hostname123 xinetd[5388]: START: ldminfod pid=5572 from=::ffff:172.31.100.74 Aug 12 13:07:51 hostname123 xinetd[5388]: EXIT: ldminfod status=0 pid=5572 duration=0(sec) From lesmikesell at gmail.com Mon Aug 15 16:12:53 2011 From: lesmikesell at gmail.com (Les Mikesell) Date: Mon, 15 Aug 2011 11:12:53 -0500 Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: <4E493B13.9010607@snarlnet.com> References: <4E493B13.9010607@snarlnet.com> Message-ID: <4E494585.3040602@gmail.com> On 8/15/2011 10:28 AM, Carl Keil wrote: > > Anyway, now I'm looking for a little more philosophical advice. I used > to run k12ltsp on this server, but don't any more. Now it's a > web/samba/mythtv box for my home. If it was you, and your old k12-centos > 5.3 box ate it, but you had homes, web root, samba shares and your > mythtv shows on separate drives, (and many complete backups via > BackupPC) would you take the opportunity to upgrade to centos 5.6 or 6.0 > or would you try to get back to 5.3 so you didn't break all that stuff > via new versions of php, mysql, samba (prolly not an issue) and mythtv? > What's the smart move here? Also, do you think there'll be a problem > going to 64 bit, when the old install was 32 bit? I can't see why that > would be an issue since basically I'm restoring functionality to content > that I saved. If I go, for example from 32-bit Centos 5.3 to 64-bit > Centos 6 will I still just be able to restore my /etc config files from > backup and proceed on my merry way? Or will I have to go into each one > and cut and paste relevant sections into new config files with differing > formats, places on the drive, etc. The first bit of advice would be to install on RAID1 mirrors if it holds your only copy of something, but backups are still important... If you are happy with the current functionality, installing a 32-bit 5.6 is probably the fastest way to get back in operation. There might be a surprise or two in the config files between 5.3 and 5.6 (which you deserve for not doing timely 'yum updates' but I'd expect it to be easily fixed. The switch to 6.0 would be more drastic and at this point you might as well wait for 6.1 and its fixes. Mythtv sounds like the only thing not from the standard repositories, so make sure you can get that working again first. > Also, How do you tell if your hardware should get a 64 bit centos or > not? I'm not positive about my system. I think it's a Pentium D dual > core. Obviously I'd check for sure before proceeding. But what's > considered the minimum CPU that actually works with 64 bit Centos? Boot from live or rescue media to a point where you can run 'cat /proc/cpuinfo' and look for 'lm' in the flags. The other interesting flag is 'vmx' (intel) or 'svm' (amd) which will tell you if it can run a 64-bit guest as a virtual machine. Some processors can run 64 bit natively but not virtualize them. -- Les Mikesell lesmikesell at gmail.com From news at siddall.name Mon Aug 15 18:57:18 2011 From: news at siddall.name (Jeff Siddall) Date: Mon, 15 Aug 2011 14:57:18 -0400 Subject: [K12OSN] Why PA system mode? In-Reply-To: <7459551C-9981-4AF7-ACB6-064159A7ECE6@thealmquists.net> References: <4E456D28.1020904@siddall.name> <7459551C-9981-4AF7-ACB6-064159A7ECE6@thealmquists.net> Message-ID: <4E496C0E.5050308@siddall.name> On 08/14/2011 03:09 PM, Burke Almquist wrote: > > On Aug 12, 2011, at 1:12 PM, Jeff Siddall wrote: > >> Anyone know why the clients run pulseaudio in system mode? Here's an excerpt from the client system log: >> >> Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: OK, so you are running PA in system mode. Please note that you most likely shouldn't be doing that. >> Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: If you do it nonetheless then it's your own fault if things don't work as expected. >> Aug 9 09:12:06 client-192 pulseaudio[1740]: main.c: Please read http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode for an explanation why system mode is usually a bad idea >> >> I did read http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode and sure enough they say it is there for stuff like thin clients. However, the reasons for it are really not that good: we _do_ have a real user, and things like home directories typically _aren't_ flushed on each session termination. > But we don't have a "real" user on the thin clients, the real user is logged into the server, not the client, so that doesn't help us, since PA needs to run on the client for it to output the sound. > If you were doing DRBL, where you network boot the OS, but all the apps are actually local, then system mode doesn't make sense, but you still need it on thin clients, since the user's session is actually on a different machine. Yes, but... The same user (both UID and name) logged in on the server is also present on the client -- with LTSP5 anyway. Not sure about what magic happens behind the scenes but a custom /etc/passwd containing the real user is mounted as a tmpfs file on the client. Have a look at a /etc/passwd on one of your logged in client to see what I mean. So really the user is on both the client and server and as such it seems like pulse could run as the real user. Maybe the system mode thing is a carryover from a time in the past (LTSP4 maybe?) where there were no real users on the client? Can anyone confirm/deny this? Thanks, Jeff From warren at togami.com Mon Aug 15 19:48:42 2011 From: warren at togami.com (Warren Togami Jr.) Date: Mon, 15 Aug 2011 09:48:42 -1000 Subject: [K12OSN] Test NBD Root? In-Reply-To: <4E493E17.9030203@oregonstate.edu> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4845F0.5070008@togami.com> <4E493E17.9030203@oregonstate.edu> Message-ID: <4E49781A.20605@togami.com> On 8/15/2011 5:41 AM, Dean Jones wrote: > On 8/14/11 3:02 PM, Warren Togami Jr. wrote: >> On 8/12/2011 8:15 AM, Burke Almquist wrote: >>> >>> On Aug 12, 2011, at 2:53 AM, Warren Togami Jr. wrote: >>> >>>> https://fedorahosted.org/k12linux/wiki/NBDRootConfiguration >>>> >>>> Could someone please test if the NBD Root boot documentation still >>>> works? >>>> >>>> Warren >>>> >>> >>> My Optiplex GX1 (PIII 450 192MB) works out of the box as a client on >>> a fresh install of SL6.1. Even the network setup seemed to work fine >>> for me. >>> When I enable NBD root instead of NFS, it gets to the login screen >>> fine. But when I try to log in, it keeps going back to the login >>> screen again. >>> Does NBD require more memory to work? >> >> Strange... on my system, the EL6.1 chroot with NBD boot gets stuck after: >> Adding udev persistent rules [ OK ] >> >> If you boot NFS the next visible thing there is starting the NBD swap, >> so I tried to turn off NBD swap with no apparent effect. >> >> I don't know what's going on. I will dig more later. >> >> Warren >> > > This is with the main EL6 release. With the release 11 the NBD root > works fine. So these message comparisons are between those two versions. > In my case it doesn't even get to X in EL6 with NBD boot. I suspect something during "Adding udev persistent rules [ OK ]" screwed up the nbd block device and it is just hanging. Warren From warren at togami.com Mon Aug 15 19:49:43 2011 From: warren at togami.com (Warren Togami Jr.) Date: Mon, 15 Aug 2011 09:49:43 -1000 Subject: [K12OSN] Why PA system mode? In-Reply-To: <4E496C0E.5050308@siddall.name> References: <4E456D28.1020904@siddall.name> <7459551C-9981-4AF7-ACB6-064159A7ECE6@thealmquists.net> <4E496C0E.5050308@siddall.name> Message-ID: <4E497857.9070603@togami.com> On 8/15/2011 8:57 AM, Jeff Siddall wrote: > The same user (both UID and name) logged in on the server is also > present on the client -- with LTSP5 anyway. Not sure about what magic > happens behind the scenes but a custom /etc/passwd containing the real > user is mounted as a tmpfs file on the client. Have a look at a > /etc/passwd on one of your logged in client to see what I mean. > > So really the user is on both the client and server and as such it seems > like pulse could run as the real user. > > Maybe the system mode thing is a carryover from a time in the past > (LTSP4 maybe?) where there were no real users on the client? Can anyone > confirm/deny this? > Convince LTSP upstream to change it and it will change in K12Linux. Warren From dahopkins429 at gmail.com Mon Aug 15 19:56:18 2011 From: dahopkins429 at gmail.com (David Hopkins) Date: Mon, 15 Aug 2011 15:56:18 -0400 Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: <4E493B13.9010607@snarlnet.com> References: <4E493B13.9010607@snarlnet.com> Message-ID: On Mon, Aug 15, 2011 at 11:28 AM, Carl Keil wrote: > Thanks for helping me. > > So, I tried the dd command, twice and then tried CloneZilla.? Twice.? And > the failing drive just kept throwing more and more errors.? Even with the > "move on after bad sectors" (or whatever it's called exactly) switch > thrown.? I think I need to give up on the miracle cure.? How was I ever > going to trust the copied disk anyway?? Wouldn't that just be latent > problems (half garbled config files, etc.) waiting to happen? > > Anyway, now I'm looking for a little more philosophical advice.? I used to > run k12ltsp on this server, but don't any more.? Now it's a web/samba/mythtv > box for my home.? If it was you, and your old k12-centos 5.3 box ate it, but > you had homes, web root, samba shares and your mythtv shows on separate > drives, (and many complete backups via BackupPC) would you take the > opportunity to upgrade to centos 5.6 or 6.0 or would you try to get back to > 5.3 so you didn't break all that stuff via new versions of php, mysql, samba > (prolly not an issue) and mythtv??? What's the smart move here?? Also, do > you think there'll be a problem going to 64 bit, when the old install was 32 > bit?? I can't see why that would be an issue since basically I'm restoring > functionality to content that I saved.?? If I go, for example from 32-bit > Centos 5.3 to 64-bit Centos 6 will I still just be able to restore my /etc > config files from backup and proceed on my merry way?? Or will I have to go > into each one and cut and paste relevant sections into new config files with > differing formats, places on the drive, etc. > > Also, How do you tell if your hardware should get a 64 bit centos or not? > I'm not positive about my system.? I think it's a Pentium D dual core. > Obviously I'd check for sure before proceeding.? But what's considered the > minimum CPU that actually works with 64 bit Centos? > I would upgrade to 5.6. I just installed a system using CentOS6. It installs ok but there are a number of changes in how to administer the system. In my case, I noticed that simple GUI utilities like system-config-bind-gui and system-config-network are no longer supported. Also, yumex doesn't seem to work quite right (crashes). If you're ok with either letting Network manager automagically set things or enjoy manually configuring the network interfaces and the DNS zone info then 6 isn't so bad. However, RHEL/CentOS5 is supported till 2017 (I think) which is quite long enough to build a test system to learn the in-and-outs of 6 before using it for production. Sincerely, Dave Hopkins From julius at turtle.com Mon Aug 15 20:30:01 2011 From: julius at turtle.com (Julius Szelagiewicz) Date: Mon, 15 Aug 2011 16:30:01 -0400 (EDT) Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: References: <4E493B13.9010607@snarlnet.com> Message-ID: <34794.216.216.171.235.1313440201.squirrel@216.216.171.235> > On Mon, Aug 15, 2011 at 11:28 AM, Carl Keil wrote: >> Thanks for helping me. >> >> So, I tried the dd command, twice and then tried CloneZilla.? Twice.? >> And >> the failing drive just kept throwing more and more errors.? Even with >> the >> "move on after bad sectors" (or whatever it's called exactly) switch >> thrown.? I think I need to give up on the miracle cure.? How was I ever >> going to trust the copied disk anyway?? Wouldn't that just be latent >> problems (half garbled config files, etc.) waiting to happen? >> >> Anyway, now I'm looking for a little more philosophical advice.? I used >> to >> run k12ltsp on this server, but don't any more.? Now it's a >> web/samba/mythtv >> box for my home.? If it was you, and your old k12-centos 5.3 box ate it, >> but >> you had homes, web root, samba shares and your mythtv shows on separate >> drives, (and many complete backups via BackupPC) would you take the >> opportunity to upgrade to centos 5.6 or 6.0 or would you try to get back >> to >> 5.3 so you didn't break all that stuff via new versions of php, mysql, >> samba >> (prolly not an issue) and mythtv??? What's the smart move here?? Also, >> do >> you think there'll be a problem going to 64 bit, when the old install >> was 32 >> bit?? I can't see why that would be an issue since basically I'm >> restoring >> functionality to content that I saved.?? If I go, for example from >> 32-bit >> Centos 5.3 to 64-bit Centos 6 will I still just be able to restore my >> /etc >> config files from backup and proceed on my merry way?? Or will I have to >> go >> into each one and cut and paste relevant sections into new config files >> with >> differing formats, places on the drive, etc. >> >> Also, How do you tell if your hardware should get a 64 bit centos or >> not? >> I'm not positive about my system.? I think it's a Pentium D dual core. >> Obviously I'd check for sure before proceeding.? But what's considered >> the >> minimum CPU that actually works with 64 bit Centos? >> > > I would upgrade to 5.6. I just installed a system using CentOS6. It > installs ok but there are a number of changes in how to administer the > system. In my case, I noticed that simple GUI utilities like > system-config-bind-gui and system-config-network are no longer > supported. Also, yumex doesn't seem to work quite right (crashes). If > you're ok with either letting Network manager automagically set things > or enjoy manually configuring the network interfaces and the DNS zone > info then 6 isn't so bad. However, RHEL/CentOS5 is supported till 2017 > (I think) which is quite long enough to build a test system to learn > the in-and-outs of 6 before using it for production. > > Sincerely, > Dave Hopkins Dave, I'm running 2 CentOS 6 database servers now. The helpful trick was to set one as a database server and the other as desktop environment. I ported the kernel params from database server to the "desktop" server and it became a nice server with the gui. Administration is much easier. I added gui stuff to the other server as well. julius From news at siddall.name Mon Aug 15 22:17:02 2011 From: news at siddall.name (Jeff Siddall) Date: Mon, 15 Aug 2011 18:17:02 -0400 Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: References: <4E493B13.9010607@snarlnet.com> Message-ID: <4E499ADE.9030208@siddall.name> On 08/15/2011 03:56 PM, David Hopkins wrote: > I would upgrade to 5.6. I just installed a system using CentOS6. It > installs ok but there are a number of changes in how to administer the > system. In my case, I noticed that simple GUI utilities like > system-config-bind-gui and system-config-network are no longer > supported. Also, yumex doesn't seem to work quite right (crashes). If > you're ok with either letting Network manager automagically set things > or enjoy manually configuring the network interfaces and the DNS zone > info then 6 isn't so bad. However, RHEL/CentOS5 is supported till 2017 > (I think) which is quite long enough to build a test system to learn > the in-and-outs of 6 before using it for production. Important correction: CentOS5 has already ended full updates. All updates, including maintenance, will end March 31st 2014. http://wiki.centos.org/FAQ/CentOS5 CentOS6 will reach EOL on 30 Nov 2017. Jeff From rowens at ptd.net Mon Aug 15 22:32:26 2011 From: rowens at ptd.net (Rob Owens) Date: Mon, 15 Aug 2011 18:32:26 -0400 Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: <4E493B13.9010607@snarlnet.com> References: <4E493B13.9010607@snarlnet.com> Message-ID: <20110815223226.GA26321@aurora.owens.net> On Mon, Aug 15, 2011 at 08:28:19AM -0700, Carl Keil wrote: > Thanks for helping me. > > So, I tried the dd command, twice and then tried CloneZilla. Twice. > And the failing drive just kept throwing more and more errors. Even > with the "move on after bad sectors" (or whatever it's called > exactly) switch thrown. I think I need to give up on the miracle > cure. How was I ever going to trust the copied disk anyway? > Wouldn't that just be latent problems (half garbled config files, > etc.) waiting to happen? > Did you try dd_rescue? If you need to get certain files (config files maybe) off that drive, give photorec a try. It can find files even if the MBR is trashed. > Anyway, now I'm looking for a little more philosophical advice. I > used to run k12ltsp on this server, but don't any more. Now it's a > web/samba/mythtv box for my home. If it was you, and your old > k12-centos 5.3 box ate it, but you had homes, web root, samba shares > and your mythtv shows on separate drives, (and many complete backups > via BackupPC) would you take the opportunity to upgrade to centos > 5.6 or 6.0 or would you try to get back to 5.3 so you didn't break > all that stuff via new versions of php, mysql, samba (prolly not an > issue) and mythtv? What's the smart move here? Also, do you think Have you compiled anything, or is everything from packages? Your MythTV installation may have required you to compile a video driver, lirc, or possibly drivers for your capture card(s). Don't forget about that stuff when you plan your rebuild. -Rob From carl at snarlnet.com Mon Aug 15 22:40:54 2011 From: carl at snarlnet.com (Carl Keil) Date: Mon, 15 Aug 2011 15:40:54 -0700 Subject: [K12OSN] Need advice on recovering from failed boot disk Message-ID: <4E49A076.3040707@snarlnet.com> Les Mikesell wrote: > (which you deserve for not doing timely 'yum updates' Y'know, I'd agree with you if I hadn't been doing regular yum updates. I was a little surprised to see that 5.6 was the latest stable when I started looking around for what to download. It's actually been hard to find a mirror with 5.3 on it. A bunch of the mirrors have an empty directory there. Well, thanks for all the advice. I guess I know what I'm doing today. Wish me luck. ck From lesmikesell at gmail.com Mon Aug 15 22:48:38 2011 From: lesmikesell at gmail.com (Les Mikesell) Date: Mon, 15 Aug 2011 17:48:38 -0500 Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: <4E49A076.3040707@snarlnet.com> References: <4E49A076.3040707@snarlnet.com> Message-ID: <4E49A246.2010404@gmail.com> On 8/15/2011 5:40 PM, Carl Keil wrote: > Les Mikesell wrote: >> (which you deserve for not doing timely 'yum updates' > > Y'know, I'd agree with you if I hadn't been doing regular yum updates. I > was a little surprised to see that 5.6 was the latest stable when I > started looking around for what to download. Did you reconfigure your yum repos? The default k12ltsp repos should still be working and up to date. Or are you getting a conflict that you never fixed? I sort-of remember having to make a few package changes over time but I thought is was mostly from new things in EPEL which I normally keep enabled. > It's actually been hard to > find a mirror with 5.3 on it. A bunch of the mirrors have an empty > directory there. Normally Centos will just update minor rev numbers automatically. If you need anything really old, you can find it http://vault.centos.org. -- Les Mikesell lesmikesell at gmail.com From peter at scheie.homedns.org Mon Aug 15 23:27:38 2011 From: peter at scheie.homedns.org (Peter Scheie) Date: Mon, 15 Aug 2011 18:27:38 -0500 Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: <4E49A076.3040707@snarlnet.com> References: <4E49A076.3040707@snarlnet.com> Message-ID: <4E49AB6A.7020701@scheie.homedns.org> Carl Keil wrote: > Les Mikesell wrote: >> (which you deserve for not doing timely 'yum updates' > > Y'know, I'd agree with you if I hadn't been doing regular yum updates. > I was a little surprised to see that 5.6 was the latest stable when I > started looking around for what to download. It's actually been hard to > find a mirror with 5.3 on it. A bunch of the mirrors have an empty > directory there. > > Well, thanks for all the advice. I guess I know what I'm doing today. > Wish me luck. > > ck > If you truly need, say, Centos 5.3, you can find any of the previous versions in vault.centos.org. The regular mirrors just keep the latest release of each version. Peter From odin at gnuskole.no Tue Aug 16 12:27:31 2011 From: odin at gnuskole.no (=?UTF-8?Q?Odin_N=C3=B8sen?=) Date: Tue, 16 Aug 2011 14:27:31 +0200 Subject: [K12OSN] Test NBD Root? In-Reply-To: <4E49781A.20605@togami.com> References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4845F0.5070008@togami.com> <4E493E17.9030203@oregonstate.edu> <4E49781A.20605@togami.com> Message-ID: > Could someone please test if the NBD Root boot documentation stillworks? NBD works for me. Booting from CentOS 6.0 with SL61-clients. NBD was much faster than NFS (which I expected). Still som strange problem with CAPS LOCK on the LTSP-clients - but that may be related to the NX-client that I use with the virtual machines. I will test this tomorrow on bare metal :-) I'm very happy with the progress in this project! Odin -------------- next part -------------- An HTML attachment was scrubbed... URL: From dahopkins429 at gmail.com Tue Aug 16 13:08:50 2011 From: dahopkins429 at gmail.com (David Hopkins) Date: Tue, 16 Aug 2011 09:08:50 -0400 Subject: [K12OSN] Need advice on recovering from failed boot disk In-Reply-To: <4E499ADE.9030208@siddall.name> References: <4E493B13.9010607@snarlnet.com> <4E499ADE.9030208@siddall.name> Message-ID: On Mon, Aug 15, 2011 at 6:17 PM, Jeff Siddall wrote: > On 08/15/2011 03:56 PM, David Hopkins wrote: >> >> I would upgrade to 5.6. I just installed a system using CentOS6. It >> installs ok but there are a number of changes in how to administer the >> system. In my case, I noticed that simple GUI utilities like >> system-config-bind-gui and system-config-network are no longer >> supported. Also, yumex doesn't seem to work quite right (crashes). ?If >> you're ok with either letting Network manager automagically set things >> or enjoy manually configuring the network interfaces and the DNS zone >> info then 6 isn't so bad. However, RHEL/CentOS5 is supported till 2017 >> (I think) which is quite long enough to build a test system to learn >> the in-and-outs of 6 before using it for production. > > Important correction: CentOS5 has already ended full updates. ?All updates, > including maintenance, will end March 31st 2014. > > http://wiki.centos.org/FAQ/CentOS5 > > CentOS6 will reach EOL on 30 Nov 2017. > Wow ... I definitely misread that page. Thanks for the correction. Sincerely, Dave Hopkins From dean.jones at oregonstate.edu Wed Aug 17 21:20:53 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Wed, 17 Aug 2011 14:20:53 -0700 Subject: [K12OSN] Test NBD Root? In-Reply-To: References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4845F0.5070008@togami.com> <4E493E17.9030203@oregonstate.edu> <4E49781A.20605@togami.com> Message-ID: Just another NBD note. I have had xinetd freak out a couple of times and stop serving nbdroot causing kernel panics on clients while they are booting. On Tue, Aug 16, 2011 at 5:27 AM, Odin N?sen wrote: >> Could someone please test if the NBD Root boot documentation stillworks? > > NBD works for me. Booting from CentOS 6.0 with SL61-clients. NBD was much > faster than NFS (which I expected). Still som strange problem with CAPS LOCK > on the LTSP-clients - but that may be related to the NX-client that I use > with the virtual machines. I will test this tomorrow on bare metal :-) > > I'm very happy with the progress in this project! > > > Odin > From lesmikesell at gmail.com Wed Aug 17 21:38:50 2011 From: lesmikesell at gmail.com (Les Mikesell) Date: Wed, 17 Aug 2011 16:38:50 -0500 Subject: [K12OSN] Test NBD Root? In-Reply-To: References: <4E44DBE3.2010204@togami.com> <36DE5A27-7B64-43F7-8E72-EC90CD421C5F@thealmquists.net> <4E4845F0.5070008@togami.com> <4E493E17.9030203@oregonstate.edu> <4E49781A.20605@togami.com> Message-ID: <4E4C34EA.1000308@gmail.com> On 8/17/2011 4:20 PM, Dean Jones wrote: > Just another NBD note. > > I have had xinetd freak out a couple of times and stop serving nbdroot > causing kernel panics on clients while they are booting. Does 'freak out' mean an error or just exceeding the default connection rate of 50 connections in a 10 second interval? You can change that with the 'cps' option for the service. -- Les Mikesell lesmikesell at gmail.com From jvermeulen at cawdekempen.be Thu Aug 18 07:23:37 2011 From: jvermeulen at cawdekempen.be (Johan Vermeulen) Date: Thu, 18 Aug 2011 09:23:37 +0200 Subject: [K12OSN] Fwd: comment on Documentation Message-ID: <4E4CBDF9.6020004@cawdekempen.be> Dear All, can I first say that I think it's very cool that this exists on Centos6. I 've been configuring ltsp on OpenSuse for some time, but that doesn's work out of the box, at least not with our hardware. K12 does, and in record time. So on facebook a noticed a request for comment on the documentation: I like it because if you follow it, your thinclients will boot. Sounds logical, but previous experience says it's not always so. But because this is used by people who may not know that much about the Linux system, maybe it would be a good idea to make an extra, separate part - not in between the commands- that explains more about the commands that are used. I myself spent some time searching for 'where is dhcp configured ', being used to do it manually. Also, I think that because there is a " ltsp server setup procedure" and "client set-up procedure", it might be useful to point out that they both happen on the same machine. Not everybody knows how pxe boot works, so people may think they need to do the client set-up on the thinclient. So that's my opinion, everybody feel free to knock me stupid if you feel I'm wrong. Can I just add a question here: I previously also tested Thinstation, http://www.thinstation.org/ I would like to compare the two in terms of bandwidth use on the lan, does anybody know how to do that? greetings, James Opensource Software is the future. From jomegat at jomegat.com Sun Aug 21 16:59:09 2011 From: jomegat at jomegat.com (Jomegat) Date: Sun, 21 Aug 2011 12:59:09 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E51370F.8070707@jomegat.com> References: <4E51370F.8070707@jomegat.com> Message-ID: <4E51395D.3050105@jomegat.com> On 08/21/2011 12:49 PM, Jomegat wrote: > I've been running a K12linux EL-5 64-bit system for three or four years > now with a pretty good measure of success. However, our teacher has > requested the ability to use a website that requires sound and mic > input. So I came in today to see if either of those work. No luck so far. > > I read through the thread where Jeff S had trouble with flash and sound, > and people were talkinga bout making sure aplay works. Well - I guess it > doesn't. > > [root at monarch ~]# aplay > ALSA lib confmisc.c:768:(parse_card) cannot find card '0' > ALSA lib conf.c:3513:(_snd_config_evaluate) function > snd_func_card_driver returned error: No such file or directory > ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings > ALSA lib conf.c:3513:(_snd_config_evaluate) function snd_func_concat > returned error: No such file or directory > ALSA lib confmisc.c:1251:(snd_func_refer) error evaluating name > ALSA lib conf.c:3513:(_snd_config_evaluate) function snd_func_refer > returned error: No such file or directory > ALSA lib conf.c:3985:(snd_config_expand) Evaluate error: No such file or > directory > ALSA lib pcm.c:2184:(snd_pcm_open_noupdate) Unknown PCM default > aplay: main:583: audio open error: No such file or directory > > I have some pretty old thin clients - I'm sure they're 586's, but can't > figure out the key sequence to get into the BIOS to check for sure (and > I don't know how to find that out for a TC). They do have sound > integrated on the mobo though. OK - I got into the bios. The TC's have Intel Celeron's running at 900MHz with 256MB of RAM. These are about a decade old. > > Any help I could get here would be very much appreciated. > -- Jim Thomas jomegat at jomegat.com Just because everything is different doesn't mean anything has changed. -- Irene Peter From jomegat at jomegat.com Sun Aug 21 16:49:19 2011 From: jomegat at jomegat.com (Jomegat) Date: Sun, 21 Aug 2011 12:49:19 -0400 Subject: [K12OSN] Sound problems Message-ID: <4E51370F.8070707@jomegat.com> I've been running a K12linux EL-5 64-bit system for three or four years now with a pretty good measure of success. However, our teacher has requested the ability to use a website that requires sound and mic input. So I came in today to see if either of those work. No luck so far. I read through the thread where Jeff S had trouble with flash and sound, and people were talkinga bout making sure aplay works. Well - I guess it doesn't. [root at monarch ~]# aplay ALSA lib confmisc.c:768:(parse_card) cannot find card '0' ALSA lib conf.c:3513:(_snd_config_evaluate) function snd_func_card_driver returned error: No such file or directory ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings ALSA lib conf.c:3513:(_snd_config_evaluate) function snd_func_concat returned error: No such file or directory ALSA lib confmisc.c:1251:(snd_func_refer) error evaluating name ALSA lib conf.c:3513:(_snd_config_evaluate) function snd_func_refer returned error: No such file or directory ALSA lib conf.c:3985:(snd_config_expand) Evaluate error: No such file or directory ALSA lib pcm.c:2184:(snd_pcm_open_noupdate) Unknown PCM default aplay: main:583: audio open error: No such file or directory I have some pretty old thin clients - I'm sure they're 586's, but can't figure out the key sequence to get into the BIOS to check for sure (and I don't know how to find that out for a TC). They do have sound integrated on the mobo though. Any help I could get here would be very much appreciated. -- Jim Thomas jomegat at jomegat.com Just because everything is different doesn't mean anything has changed. -- Irene Peter From robark at gmail.com Sun Aug 21 22:44:59 2011 From: robark at gmail.com (Robert Arkiletian) Date: Sun, 21 Aug 2011 15:44:59 -0700 Subject: [K12OSN] OT: LinuxCon 2011 Message-ID: I have many friends on this list so please excuse this email for being off topic. LinuxCon was just held in Vancouver Canada, my hometown. I really enjoyed it, especially the people. As Eben put it "There IS real strength in community". The food was great, the gala party was awesome too. What a band! Technically, I did learn a lot about filesystem best practices for various workloads. That was interesting. Also learned about some new and cool monitoring/benchmarking tools and some Python resources. Discovered OSUOSL http://osuosl.org/ and the people behind it http://osuosl.org/about/people and their story http://www.linux.com/news/enterprise/biz-enterprise/486133-a-look-at-oregon-state-universitys-open-source-lab I forgot to get a photo with Greg Kroah-Hartman (stable Kernel tree maintainer). Oh well maybe next time. But we did chat for a bit. Here are my photos https://picasaweb.google.com/robark/LinuxCon2011?authuser=0&feat=directlink Yes I finally met Linus. I had a permanent smile for 2 hours after that. Also having lunch with Jon "Maddog" Hall was fun too. Cheers -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From microman at cmosnetworks.com Mon Aug 22 10:49:58 2011 From: microman at cmosnetworks.com (Terrell Prude' Jr.) Date: Mon, 22 Aug 2011 06:49:58 -0400 Subject: [K12OSN] OT: LinuxCon 2011 In-Reply-To: References: Message-ID: <4E523456.3070304@cmosnetworks.com> Sounds like fun, and Vancouver, BC is a lovely city. Glad you enjoyed it! --TP Robert Arkiletian wrote: > I have many friends on this list so please excuse this email for being > off topic. > > LinuxCon was just held in Vancouver Canada, my hometown. I really > enjoyed it, especially the people. As Eben put it > "There IS real strength in community". > The food was great, the gala party was awesome too. What a band! > > Technically, I did learn a lot about filesystem best practices for > various workloads. That was interesting. Also learned about some new > and cool monitoring/benchmarking tools and some Python resources. > > Discovered OSUOSL http://osuosl.org/ > and the people behind it http://osuosl.org/about/people > and their story > http://www.linux.com/news/enterprise/biz-enterprise/486133-a-look-at-oregon-state-universitys-open-source-lab > > I forgot to get a photo with Greg Kroah-Hartman (stable Kernel tree > maintainer). Oh well maybe next time. But we did chat for a bit. > Here are my photos > https://picasaweb.google.com/robark/LinuxCon2011?authuser=0&feat=directlink > > Yes I finally met Linus. I had a permanent smile for 2 hours after > that. Also having lunch with Jon "Maddog" Hall was fun too. > > Cheers > > From ltsp at symbio-technologies.com Mon Aug 22 12:50:50 2011 From: ltsp at symbio-technologies.com (Gideon Romm) Date: Mon, 22 Aug 2011 08:50:50 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E51370F.8070707@jomegat.com> References: <4E51370F.8070707@jomegat.com> Message-ID: On Sun, Aug 21, 2011 at 12:49 PM, Jomegat wrote: > [root at monarch ~]# aplay > ALSA lib confmisc.c:768:(parse_card) cannot find card '0' It seems ALSA cannot find card 0. Did you run this command on the server? If YES => the alsa -> pulseaudio redirection is not configured and you need to modify your /etc/asound.conf If NO (you ran the command on the thin client) => The thin client cannot find an appropriate driver for the sound card. -Gadi From jomegat at jomegat.com Mon Aug 22 12:52:41 2011 From: jomegat at jomegat.com (Jomegat) Date: Mon, 22 Aug 2011 08:52:41 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E51395D.3050105@jomegat.com> References: <4E51370F.8070707@jomegat.com> <4E51395D.3050105@jomegat.com> Message-ID: <4E525119.7010807@jomegat.com> On 08/21/2011 12:59 PM, Jomegat wrote: > On 08/21/2011 12:49 PM, Jomegat wrote: >> I've been running a K12linux EL-5 64-bit system for three or four years >> now with a pretty good measure of success. However, our teacher has >> requested the ability to use a website that requires sound and mic >> input. So I came in today to see if either of those work. No luck so far. >> >> I read through the thread where Jeff S had trouble with flash and sound, >> and people were talkinga bout making sure aplay works. Well - I guess it >> doesn't. >> >> [root at monarch ~]# aplay >> ALSA lib confmisc.c:768:(parse_card) cannot find card '0' >> ALSA lib conf.c:3513:(_snd_config_evaluate) function >> snd_func_card_driver returned error: No such file or directory >> ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings >> ALSA lib conf.c:3513:(_snd_config_evaluate) function snd_func_concat >> returned error: No such file or directory >> ALSA lib confmisc.c:1251:(snd_func_refer) error evaluating name >> ALSA lib conf.c:3513:(_snd_config_evaluate) function snd_func_refer >> returned error: No such file or directory >> ALSA lib conf.c:3985:(snd_config_expand) Evaluate error: No such file or >> directory >> ALSA lib pcm.c:2184:(snd_pcm_open_noupdate) Unknown PCM default >> aplay: main:583: audio open error: No such file or directory >> >> I have some pretty old thin clients - I'm sure they're 586's, but can't >> figure out the key sequence to get into the BIOS to check for sure (and >> I don't know how to find that out for a TC). They do have sound >> integrated on the mobo though. > > OK - I got into the bios. The TC's have Intel Celeron's running at > 900MHz with 256MB of RAM. These are about a decade old. > >> >> Any help I could get here would be very much appreciated. OK, here's a little more info. /opt/ltsp/i386/etc/lts.conf contains the following: # enable sound by default SOUND = Y # choose either esd or nasd to be the default (esd only on x86_64) SOUND_DAEMON = "esd" # SOUND_DAEMON = "nasd" # default sound volume VOLUME = 75 ### For ISA sound cards, you have to specify the module to use: # SMODULE_01 = "sb io=0x220 irq=5 dma=1" Any ideas? -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Any sufficiently advanced technology is indistinguishable from magic. - Arthur C. Clarke From jomegat at jomegat.com Mon Aug 22 13:34:49 2011 From: jomegat at jomegat.com (Jomegat) Date: Mon, 22 Aug 2011 09:34:49 -0400 Subject: [K12OSN] Sound problems In-Reply-To: References: <4E51370F.8070707@jomegat.com> Message-ID: <4E525AF9.9080800@jomegat.com> On 08/22/2011 08:50 AM, Gideon Romm wrote: > On Sun, Aug 21, 2011 at 12:49 PM, Jomegat wrote: > >> [root at monarch ~]# aplay >> ALSA lib confmisc.c:768:(parse_card) cannot find card '0' > > It seems ALSA cannot find card 0. > > Did you run this command on the server? Good question. I was logged into a TC and opened an xterm. I THINK that means I ran the command on the server. I do not know how to run anything on the TC, but would appreciate enlightenment. > > If YES => the alsa -> pulseaudio redirection is not configured and you > need to modify your /etc/asound.conf This could be it - there is no /etc/asound.conf. Perhaps I have a missing package? > > If NO (you ran the command on the thin client) => The thin client > cannot find an appropriate driver for the sound card. Thanks! -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Any sufficiently advanced technology is indistinguishable from magic. - Arthur C. Clarke From news at siddall.name Mon Aug 22 14:52:16 2011 From: news at siddall.name (Jeff Siddall) Date: Mon, 22 Aug 2011 10:52:16 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E525AF9.9080800@jomegat.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> Message-ID: <4E526D20.5030807@siddall.name> On 08/22/2011 09:34 AM, Jomegat wrote: > On 08/22/2011 08:50 AM, Gideon Romm wrote: >> On Sun, Aug 21, 2011 at 12:49 PM, Jomegat wrote: >> >>> [root at monarch ~]# aplay >>> ALSA lib confmisc.c:768:(parse_card) cannot find card '0' >> >> It seems ALSA cannot find card 0. >> >> Did you run this command on the server? > > Good question. I was logged into a TC and opened an xterm. I THINK that > means I ran the command on the server. I do not know how to run anything > on the TC, but would appreciate enlightenment. > >> >> If YES => the alsa -> pulseaudio redirection is not configured and you >> need to modify your /etc/asound.conf > > This could be it - there is no /etc/asound.conf. Perhaps I have a > missing package? > >> >> If NO (you ran the command on the thin client) => The thin client >> cannot find an appropriate driver for the sound card. > > Thanks! Are you sure you are running K12Linux and not K12LTSP? K12Linux used LTSP5 and I thought pulseaudio was one of the things that didn't work in EL5. Most EL5 users run K12LTSP which was LTSP4. If it is LTSP4 I really have no idea how it is supposed to work because I have never run anything less than LTSP5. It looks like LTSP4 is using ESD for network audio but I don't know if the mic ever worked with that. You might end up having to go to K12Linux with Fedora or EL6. Jeff From roland at astrofoto.org Mon Aug 22 17:31:42 2011 From: roland at astrofoto.org (Roland Roberts) Date: Mon, 22 Aug 2011 13:31:42 -0400 Subject: [K12OSN] Mixed Network configuration woes Message-ID: <4E52927E.4000703@astrofoto.org> I've been reading through the page https://fedorahosted.org/k12linux/wiki/MixedNetworkSetup to try to get this working, but I've head a wall. Things *look* like they are working up to a point. My client gets the PXE info and downloads the files appearing to boot. But then when the display goes to initialize, I end up with an odd partial screen at which point nothing works. After a few seconds, the caps-lock key light begins to blink suggesting to me that the machine has locked up. First, I had everything working via a PXE-boot VM before moving on to trying to get this setup working. So at least I knew I had a working configuration. But it will be difficult if not impossible to set up separate networks for this site :-/ I have a dhcp server that has been working fine for general requests. Here's my stanza for my test host.... host ltspc0 { hardware ethernet f0:de:f1:79:d5:c4; option host-name "lstpc0.rlent.pnet"; option log-servers 192.168.3.3; option tcode "America/New_York"; next-server 192.168.3.36; option option-128 "192.168.3.36"; option option-129 "192.168.3.36"; filename "/ltsp/i386/pxelinux.0"; option root-path "192.168.3.36:/opt/ltsp/i386"; fixed-address 192.168.3.102; } I've tried with and without root-path, and options 128 and 129 but there is no difference. I'm not getting any log messages on the log-server to help track this down. the client is a Thinkpad 420s, a fairly new machine (actually newer than the server), so i'm wondering about the video driver. From what I've read, RHEL supposedly included this driver in 6.1, so I was assuming that should work, but maybe I'm wrong there. Or maybe it's something completely else. Can anyone point me to troubleshooting resources for setting this up? I can also post a YouTube video of the boot sequence if anyone thinks that would help. The last thing on the screen is before the half flat gray screen is: mtrr: no more MTRRs available [drm] MTRR allocation failed. graphics performance may suffer roland -- Roland Roberts, PhD http://www.astrofoto.org/ From roland at astrofoto.org Mon Aug 22 17:37:57 2011 From: roland at astrofoto.org (Roland Roberts) Date: Mon, 22 Aug 2011 13:37:57 -0400 Subject: [K12OSN] Mixed Network configuration woes In-Reply-To: <4E52927E.4000703@astrofoto.org> References: <4E52927E.4000703@astrofoto.org> Message-ID: <4E5293F5.7020209@astrofoto.org> On 8/22/2011 1:31 PM, Roland Roberts wrote: [...] > I have a dhcp server that has been working fine for general requests. > Here's my stanza for my test host.... > > host ltspc0 { > hardware ethernet f0:de:f1:79:d5:c4; > option host-name "lstpc0.rlent.pnet"; > option log-servers 192.168.3.3; > option tcode "America/New_York"; > next-server 192.168.3.36; > option option-128 "192.168.3.36"; > option option-129 "192.168.3.36"; > filename "/ltsp/i386/pxelinux.0"; > option root-path "192.168.3.36:/opt/ltsp/i386"; > fixed-address 192.168.3.102; > } > > I've tried with and without root-path, and options 128 and 129 but > there is no difference. I'm not getting any log messages on the > log-server to help track this down. How very bizarre. What I failed to notice is that when I added the root-path, and option-128/option-129, the capslock key was NOT left flashing. After walking away for a nature call and returning, what do I find but the k12osn login screen. So it appears to be working, but the whacky screen and long delay had me confused. The delay IS long, a few minutes. I'm running on gigabit ethernet, so I'm not sure why it would be so slow. roland -- Roland Roberts, PhD http://www.astrofoto.org/ From roland at astrofoto.org Mon Aug 22 17:41:01 2011 From: roland at astrofoto.org (Roland Roberts) Date: Mon, 22 Aug 2011 13:41:01 -0400 Subject: [K12OSN] Mixed Network configuration woes In-Reply-To: <4E5293F5.7020209@astrofoto.org> References: <4E52927E.4000703@astrofoto.org> <4E5293F5.7020209@astrofoto.org> Message-ID: <4E5294AD.4020902@astrofoto.org> On 8/22/2011 1:37 PM, Roland Roberts wrote: > How very bizarre. What I failed to notice is that when I added the > root-path, and option-128/option-129, the capslock key was NOT left > flashing. After walking away for a nature call and returning, what do > I find but the k12osn login screen. > > So it appears to be working, but the whacky screen and long delay had > me confused. The delay IS long, a few minutes. I'm running on gigabit > ethernet, so I'm not sure why it would be so slow. Okay, I'm taking a long walk now :-/. The k12osn login screen IS displaying, but it doesn't respond to keystrokes. And the screen is flickering significantly suggesting I still have a video issue. If anyone can give me a clue or point to some tests I can do, I'd appreciate it. roland -- Roland Roberts, PhD http://www.astrofoto.org/ From jomegat at jomegat.com Mon Aug 22 15:41:45 2011 From: jomegat at jomegat.com (Jomegat) Date: Mon, 22 Aug 2011 11:41:45 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E526D20.5030807@siddall.name> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> Message-ID: <4E5278B9.8020806@jomegat.com> On 08/22/2011 10:52 AM, Jeff Siddall wrote: > On 08/22/2011 09:34 AM, Jomegat wrote: >> On 08/22/2011 08:50 AM, Gideon Romm wrote: >>> On Sun, Aug 21, 2011 at 12:49 PM, Jomegat wrote: >>> >>>> [root at monarch ~]# aplay >>>> ALSA lib confmisc.c:768:(parse_card) cannot find card '0' >>> >>> It seems ALSA cannot find card 0. >>> >>> Did you run this command on the server? >> >> Good question. I was logged into a TC and opened an xterm. I THINK that >> means I ran the command on the server. I do not know how to run anything >> on the TC, but would appreciate enlightenment. >> >>> >>> If YES => the alsa -> pulseaudio redirection is not configured and you >>> need to modify your /etc/asound.conf >> >> This could be it - there is no /etc/asound.conf. Perhaps I have a >> missing package? >> >>> >>> If NO (you ran the command on the thin client) => The thin client >>> cannot find an appropriate driver for the sound card. >> >> Thanks! > > Are you sure you are running K12Linux and not K12LTSP? K12Linux used > LTSP5 and I thought pulseaudio was one of the things that didn't work in > EL5. Most EL5 users run K12LTSP which was LTSP4. > > If it is LTSP4 I really have no idea how it is supposed to work because > I have never run anything less than LTSP5. > > It looks like LTSP4 is using ESD for network audio but I don't know if > the mic ever worked with that. > > You might end up having to go to K12Linux with Fedora or EL6. Sorry - I typed K12Linux when I meant K12LTSP. I guess I need EL6 then. I have two servers right now - one of them is at the school and is slated for retirement. The other is here on my desk at the office. I could do an install from scratch. My main requirement is to support the old 900MHz Celeron's we're using for TC's. As I've been reading the list, it looks like I need to be careful about what I upgrade to. EL6 supports old clients, right? -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Any sufficiently advanced technology is indistinguishable from magic. - Arthur C. Clarke From roland at astrofoto.org Mon Aug 22 21:59:22 2011 From: roland at astrofoto.org (Roland Roberts) Date: Mon, 22 Aug 2011 17:59:22 -0400 Subject: [K12OSN] Mixed Network configuration woes In-Reply-To: <4E5294AD.4020902@astrofoto.org> References: <4E52927E.4000703@astrofoto.org> <4E5293F5.7020209@astrofoto.org> <4E5294AD.4020902@astrofoto.org> Message-ID: <4E52D13A.6010009@astrofoto.org> On 8/22/2011 1:41 PM, Roland Roberts wrote: > On 8/22/2011 1:37 PM, Roland Roberts wrote: >> How very bizarre. What I failed to notice is that when I added the >> root-path, and option-128/option-129, the capslock key was NOT left >> flashing. After walking away for a nature call and returning, what do >> I find but the k12osn login screen. >> >> So it appears to be working, but the whacky screen and long delay had >> me confused. The delay IS long, a few minutes. I'm running on gigabit >> ethernet, so I'm not sure why it would be so slow. > > Okay, I'm taking a long walk now :-/. The k12osn login screen IS > displaying, but it doesn't respond to keystrokes. And the screen is > flickering significantly suggesting I still have a video issue. Doh. USB keyboard came unplugged. I'm not sure what's causing the flickering, but when I use an external monitor, everything works fine. I still have that long boot delay with the whacky screen. I'm sure it's just a boot delay as packages load (well, pretty sure), but the screen is confusing. So, I'm going to declare success at this point. roland -- Roland Roberts, PhD http://www.astrofoto.org/ From warren at togami.com Tue Aug 23 00:46:52 2011 From: warren at togami.com (Warren Togami Jr.) Date: Mon, 22 Aug 2011 14:46:52 -1000 Subject: [K12OSN] Sound problems In-Reply-To: <4E5278B9.8020806@jomegat.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> Message-ID: <4E52F87C.5030603@togami.com> On 8/22/2011 5:41 AM, Jomegat wrote: > > My main requirement is to support the old 900MHz Celeron's we're using > for TC's. As I've been reading the list, it looks like I need to be > careful about what I upgrade to. EL6 supports old clients, right? > EL6 requires a minimum of i686. I believe Celeron was a minimum of i686, especially at 900MHz, so you will likely be fine. Warren From roland at astrofoto.org Tue Aug 23 05:09:02 2011 From: roland at astrofoto.org (Roland Roberts) Date: Tue, 23 Aug 2011 01:09:02 -0400 Subject: [K12OSN] Troubleshooting fuse? Message-ID: <4E5335EE.2050707@astrofoto.org> My local USB drives don't show up on my desktop. Apart from that obvious symptom, I'm wondering where I look to find out what might be the problem. I'm a member of group fuse, so the obvious is covered. My DHCP server is separate from my LTSP/TFTP server, and my home directory is an NFS mount. The NFS server is the same as the DHCP server. Oh yes, and accounts are all handled via NIS and the NIS master is yet another host. I'm not sure if any of that matters or can confuse how FUSE works, but thought I should mention it. Where do I look for errors? roland -- Roland Roberts, PhD http://www.astrofoto.org/ From dean.jones at oregonstate.edu Tue Aug 23 15:02:32 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Tue, 23 Aug 2011 08:02:32 -0700 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E5335EE.2050707@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> Message-ID: What version of K12 are you running? What fuse packages do you have installed on the server? On Mon, Aug 22, 2011 at 10:09 PM, Roland Roberts wrote: > My local USB drives don't show up on my desktop. Apart from that obvious > symptom, I'm wondering where I look to find out what might be the > problem. I'm a member of group fuse, so the obvious is covered. > > My DHCP server is separate from my LTSP/TFTP server, and my home > directory is an NFS mount. The NFS server is the same as the DHCP > server. Oh yes, and accounts are all handled via NIS and the NIS master > is yet another host. I'm not sure if any of that matters or can confuse > how FUSE works, but thought I should mention it. > > Where do I look for errors? > > roland > > -- > Roland Roberts, PhD > http://www.astrofoto.org/ > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From news at siddall.name Tue Aug 23 16:22:33 2011 From: news at siddall.name (Jeff Siddall) Date: Tue, 23 Aug 2011 12:22:33 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E5278B9.8020806@jomegat.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> Message-ID: <4E53D3C9.7090700@siddall.name> On 08/22/2011 11:41 AM, Jomegat wrote: > On 08/22/2011 10:52 AM, Jeff Siddall wrote: >> On 08/22/2011 09:34 AM, Jomegat wrote: >>> On 08/22/2011 08:50 AM, Gideon Romm wrote: >>>> On Sun, Aug 21, 2011 at 12:49 PM, Jomegat wrote: >>>> >>>>> [root at monarch ~]# aplay >>>>> ALSA lib confmisc.c:768:(parse_card) cannot find card '0' >>>> >>>> It seems ALSA cannot find card 0. >>>> >>>> Did you run this command on the server? >>> >>> Good question. I was logged into a TC and opened an xterm. I THINK that >>> means I ran the command on the server. I do not know how to run anything >>> on the TC, but would appreciate enlightenment. >>> >>>> >>>> If YES => the alsa -> pulseaudio redirection is not configured and you >>>> need to modify your /etc/asound.conf >>> >>> This could be it - there is no /etc/asound.conf. Perhaps I have a >>> missing package? >>> >>>> >>>> If NO (you ran the command on the thin client) => The thin client >>>> cannot find an appropriate driver for the sound card. >>> >>> Thanks! >> >> Are you sure you are running K12Linux and not K12LTSP? K12Linux used >> LTSP5 and I thought pulseaudio was one of the things that didn't work in >> EL5. Most EL5 users run K12LTSP which was LTSP4. >> >> If it is LTSP4 I really have no idea how it is supposed to work because >> I have never run anything less than LTSP5. >> >> It looks like LTSP4 is using ESD for network audio but I don't know if >> the mic ever worked with that. >> >> You might end up having to go to K12Linux with Fedora or EL6. > > Sorry - I typed K12Linux when I meant K12LTSP. I guess I need EL6 then. > I have two servers right now - one of them is at the school and is > slated for retirement. The other is here on my desk at the office. I > could do an install from scratch. > > My main requirement is to support the old 900MHz Celeron's we're using > for TC's. As I've been reading the list, it looks like I need to be > careful about what I upgrade to. EL6 supports old clients, right? No, unfortunately EL6 doesn't support "old" clients that well, but that is really dependent on your definition of "old". I would say it does not support _really_ old clients. Basically any CPU that will run i686 with PAE is good enough -- maybe even i686 without PAE. i586 and below (we are talking a decade old or more now) are probably a no-go. A Celeron 900 (mobile I assume) should be fine. Before you go to far you might want to try a test by downloading a SL6 live CD (i686 of course, not x86_64) and see if your clients will boot. You can burn it on a USB stick if your clients don't have CD drives. Of course it is also possible that mics can be made to work with LTSP4 but unfortunately someone else will have to help you with that. To be honest I have not even tried mics on LTSP5, but in principle pulseaudio should handle that. Jeff From roland at astrofoto.org Tue Aug 23 17:01:43 2011 From: roland at astrofoto.org (Roland Roberts) Date: Tue, 23 Aug 2011 13:01:43 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: References: <4E5335EE.2050707@astrofoto.org> Message-ID: <4E53DCF7.5000700@astrofoto.org> On 08/23/2011 11:02 AM, Dean Jones wrote: > What version of K12 are you running? I installed k12linux-release-5.2.17-1.el6.x86_64.rpm and went from there. The server is running SL 6.1. > What fuse packages do you have installed on the server? The server has the following: 256 roland> rpm -qa | grep fuse fuse-2.8.3-3.el6_1.x86_64 gvfs-fuse-1.4.3-12.el6.x86_64 fuse-sshfs-2.2-6.el6.x86_64 fuse-libs-2.8.3-3.el6_1.x86_64 I found these instructions https://wiki.ubuntu.com/DebugLocalDev and get through to the end with a manual attempt to mount the file system: -bash-4.1# ssh -X -S /var/run/ldm_socket_vt1_192.168.3.36 192.168.3.36 "/usr/sbin/ltspfsmounter /tmp add" root at 192.168.3.36's password: Wrote 842a4bb8bbb5126a62a2d5f5f07590fa size 32, waiting At which point nothing happens. However, I think I may have found the problem: https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/690050 According to the comment there ltspfs has a problem with 64-bit, which is what I'm running. It's an ubuntu-listed bug which seems to track back to this bug report: https://bugs.launchpad.net/ltsp/+bug/415952 So...do I need to revert to 32-bit images to get this to work? roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From jim.kinney at gmail.com Tue Aug 23 17:14:06 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Tue, 23 Aug 2011 13:14:06 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E53DCF7.5000700@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> Message-ID: I have not read the docs being used so I can only comment generically: using Ubuntu cookbook-style instructions is likely to not work on a RedHat based system. SL-6 is a RedHat based system. On Aug 23, 2011 1:04 PM, "Roland Roberts" wrote: > On 08/23/2011 11:02 AM, Dean Jones wrote: >> What version of K12 are you running? > > I installed k12linux-release-5.2.17-1.el6.x86_64.rpm and went from > there. The server is running SL 6.1. > >> What fuse packages do you have installed on the server? > > The server has the following: > > 256 roland> rpm -qa | grep fuse > fuse-2.8.3-3.el6_1.x86_64 > gvfs-fuse-1.4.3-12.el6.x86_64 > fuse-sshfs-2.2-6.el6.x86_64 > fuse-libs-2.8.3-3.el6_1.x86_64 > > I found these instructions https://wiki.ubuntu.com/DebugLocalDev and get > through to the end with a manual attempt to mount the file system: > > -bash-4.1# ssh -X -S /var/run/ldm_socket_vt1_192.168.3.36 > 192.168.3.36 "/usr/sbin/ltspfsmounter /tmp add" > root at 192.168.3.36's password: > Wrote 842a4bb8bbb5126a62a2d5f5f07590fa size 32, waiting > > At which point nothing happens. > > However, I think I may have found the problem: > https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/690050 > > According to the comment there ltspfs has a problem with 64-bit, which > is what I'm running. It's an ubuntu-listed bug which seems to track back > to this bug report: https://bugs.launchpad.net/ltsp/+bug/415952 > > So...do I need to revert to 32-bit images to get this to work? > > roland > > -- > PGP Key ID: 66 BC 3B CD > Roland B. Roberts, PhD RL Enterprises > roland at rlenter.com 6818 Madeline Court > roland at astrofoto.org Brooklyn, NY 11220 > > _______________________________________________ > 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 roland at astrofoto.org Tue Aug 23 17:21:20 2011 From: roland at astrofoto.org (Roland Roberts) Date: Tue, 23 Aug 2011 13:21:20 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> Message-ID: <4E53E190.4030001@astrofoto.org> On 08/23/2011 01:14 PM, Jim Kinney wrote: > > I have not read the docs being used so I can only comment generically: > using Ubuntu cookbook-style instructions is likely to not work on a > RedHat based system. SL-6 is a RedHat based system. > Unfortunately, there do not appear to be any RH based cookbooks nor anything I can find resembling debugging instructions. What I do know is that when I look at syslog on the client, the device IS being detected. And I know that, as the Ubuntu instructions note, that the device does then show up in /var/run/ltspfs_fstab. What I don't know is what goes wrong after that nor where else to work, especially if their next step of trying ltspfsmounter is wrong :-( roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From ltsp at symbio-technologies.com Tue Aug 23 19:48:51 2011 From: ltsp at symbio-technologies.com (Gideon Romm) Date: Tue, 23 Aug 2011 15:48:51 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E53E190.4030001@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> Message-ID: The thin client chroot needs to be 32-bit for ltspfs to work. The server can be 64-bit. On Tue, Aug 23, 2011 at 1:21 PM, Roland Roberts wrote: > On 08/23/2011 01:14 PM, Jim Kinney wrote: >> >> I have not read the docs being used so I can only comment generically: >> using Ubuntu cookbook-style instructions is likely to not work on a RedHat >> based system. SL-6 is a RedHat based system. >> > > Unfortunately, there do not appear to be any RH based cookbooks nor anything > I can find resembling debugging instructions. > > What I do know is that when I look at syslog on the client, the device IS > being detected. And I know that, as the Ubuntu instructions note, that the > device does then show up in /var/run/ltspfs_fstab. What I don't know is what > goes wrong after that nor where else to work, especially if their next step > of trying ltspfsmounter is wrong :-( > > roland > > -- > ? ? ? ? ? ? ? ? ? ? ? PGP Key ID: 66 BC 3B CD > Roland B. Roberts, PhD ? ? ? ? ? ? ? ? ? ? ? ? ? ? RL Enterprises > roland at rlenter.com ? ? ? ? ? ? ? ? ? ? ? ? ? ?6818 Madeline Court > roland at astrofoto.org ? ? ? ? ? ? ? ? ? ? ? ? ? Brooklyn, NY 11220 > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From roland at astrofoto.org Tue Aug 23 21:01:19 2011 From: roland at astrofoto.org (Roland Roberts) Date: Tue, 23 Aug 2011 17:01:19 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> Message-ID: <4E54151F.1080600@astrofoto.org> On 08/23/2011 03:48 PM, Gideon Romm wrote: > The thin client chroot needs to be 32-bit for ltspfs to work. The > server can be 64-bit. Bummer. Does the 32-bit client run the PAE kernel? Some of my "thin" clients have a lot of memory. roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From jim.kinney at gmail.com Tue Aug 23 21:23:21 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Tue, 23 Aug 2011 17:23:21 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E54151F.1080600@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> <4E54151F.1080600@astrofoto.org> Message-ID: PAE is not needed for 32 bit with under 4GB memory or any 64 bit system. RedHat defaulted to only PAE kernels to save confusion as using it did nothing if it was not needed. On Aug 23, 2011 5:03 PM, "Roland Roberts" wrote: > On 08/23/2011 03:48 PM, Gideon Romm wrote: >> The thin client chroot needs to be 32-bit for ltspfs to work. The >> server can be 64-bit. > > Bummer. Does the 32-bit client run the PAE kernel? Some of my "thin" > clients have a lot of memory. > > roland > > -- > PGP Key ID: 66 BC 3B CD > Roland B. Roberts, PhD RL Enterprises > roland at rlenter.com 6818 Madeline Court > roland at astrofoto.org Brooklyn, NY 11220 > > _______________________________________________ > 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 roland at astrofoto.org Tue Aug 23 21:34:02 2011 From: roland at astrofoto.org (Roland Roberts) Date: Tue, 23 Aug 2011 17:34:02 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> <4E54151F.1080600@astrofoto.org> Message-ID: <4E541CCA.1070108@astrofoto.org> On 08/23/2011 05:23 PM, Jim Kinney wrote: > > PAE is not needed for 32 bit with under 4GB memory or any 64 bit > system. RedHat defaulted to only PAE kernels to save confusion as > using it did nothing if it was not needed. > As it happens, the client I'm on right now has 8GB of memory, which was the cause for the question. This particular client is more of a convenience as it lets me boot into Linux w/o an install; dual boot on the cheap. Most of my other clients will not be so richly endowed. roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From roland at astrofoto.org Wed Aug 24 04:54:34 2011 From: roland at astrofoto.org (Roland Roberts) Date: Wed, 24 Aug 2011 00:54:34 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> Message-ID: <4E54840A.6080107@astrofoto.org> On 08/23/2011 03:48 PM, Gideon Romm wrote: > The thin client chroot needs to be 32-bit for ltspfs to work. The > server can be 64-bit. Well, I wiped my client install and redid it... ltsp-build-client --arch i386 set a root password on the client so I can log in, rebooted the client, not appears to be a 32-bit client -bash-4.1# uname -r 2.6.32-131.6.1.el6.i686.thinclient But still no joy on localdev. Among other things, the device itself does not mount on the thin client. "df -l" doesn't show it mounted anywhere. But there is in entry in /var/run/ltspfs_fstab: -bash-4.1# cat /var/run/ltspfs_fstab /dev/sdf1 /var/run/drives/usbdisk-sdf1 vfat defaults,utf8 0 0 > On Tue, Aug 23, 2011 at 1:21 PM, Roland Roberts wrote: >> On 08/23/2011 01:14 PM, Jim Kinney wrote: >>> I have not read the docs being used so I can only comment generically: >>> using Ubuntu cookbook-style instructions is likely to not work on a RedHat >>> based system. SL-6 is a RedHat based system Again, this is the best I've got. Here's the last step in that cookbook: -bash-4.1# ssh -X -S /var/run/ldm_socket_vt1_192.168.3.36 192.168.3.36 "/usr/sbin/ltspfsmounter /tmp add" root at 192.168.3.36's password: Wrote 281ddd7e5b4145e0540effae42e179db size 32, waiting Error: /media/root is not mounted Error: /media/root/tmp is not mounted but the directory is mounted. Note, if I try this: -bash-4.1# ssh -X -S /var/run/ldm_socket_2791_192.168.3.36 192.168.3.36 "/usr/sbin/ltspfsmounter /tmp add" Wrote 281ddd7e5b4145e0540effae42e179db size 32, waiting fusermount: mount failed: Operation not permitted Error: /tmp/.roland-ltspfs/tmp is not mounted This turns out to be the problem. On the server, /bin/fusermount is not setuid root or setgid. I made it setuid root and try again, pulling the USB stick and reinserting it. Again, /var/log/messages shows the device recognized, and it get put into /var/run/ltspfs_fstab. This time I do see *something* on the server: [root at copernicus bin]# find /media/ /media/ /media/roland /media/roland/usbdisk-sdf1 Okay, but the device is not mounted on the thin client so when I look on the server I see this: [root at copernicus bin]# ls -l /media/roland/ ls: cannot access /media/roland/usbdisk-sdf1: Permission denied total 0 d????????? ? ? ? ? ? usbdisk-sdf1 And one last attempt; on the client I do this: -bash-4.1# ssh -X -S /var/run/ldm_socket_2791_192.168.3.36 192.168.3.36 "/usr/sbin/ltspfsmounter /var/run/drives/usbdisk-sdf1 cleanup" -bash-4.1# mount /dev/sdf1 /var/run/drives/usbdisk-sdf1 -bash-4.1# ll /var/run/drives/usbdisk-sdf1/ total 32432 drwxr-xr-x 2 root root 8192 Apr 12 09:35 boot drwxr-xr-x 3 root root 8192 Oct 22 2010 EFI -rwxr-xr-x 1 root root 18092 Oct 22 2010 GPL drwxr-xr-x 2 root root 8192 Oct 22 2010 LiveOS -rwxr-xr-x 1 root root 33150608 Aug 21 18:54 MaxImDL515.exe drwxr-xr-x 2 root root 8192 Oct 22 2010 syslinux -bash-4.1# ssh -X -S /var/run/ldm_socket_2791_192.168.3.36 192.168.3.36 "/usr/sbin/ltspfsmounter /var/run/drives/usbdisk-sdf1 add" Wrote 281ddd7e5b4145e0540effae42e179db size 32, waiting -bash-4.1# ll /var/run/drives/usbdisk-sdf1/ total 0 So something in the process of trying to mount the drive makes it disappear. I'm so close, but I'm missing something. Why isn't the client mounting the drive? It should be, shouldn't it? Why does it disappear when I try to manually mount it? roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From bojali at yahoo.com Wed Aug 24 05:08:20 2011 From: bojali at yahoo.com (xbc bojali) Date: Tue, 23 Aug 2011 22:08:20 -0700 (PDT) Subject: [K12OSN] configuration questions In-Reply-To: <4E52D13A.6010009@astrofoto.org> References: <4E52927E.4000703@astrofoto.org> <4E5293F5.7020209@astrofoto.org> <4E5294AD.4020902@astrofoto.org> <4E52D13A.6010009@astrofoto.org> Message-ID: <1314162500.44065.YahooMailRC@web83602.mail.sp1.yahoo.com> Ok I am a new to this so I thought it would be a good idea to get some feed back. I am starting the conversion for my kids to k12lltsp. I have 3 Sun Ray 2FS Virtual Display Clients that I want to use for the thin clients, I have just bought a HP Quad core to build as the host/server for them. What is the Easiest way to build these? I started with Scientific Linux i86_64 but was not able to get the ltsp-server-tweaks, to work. And then read from Rolands posts that i might need to use the 32bit not the 64? any help would be appreciated. I just started this yesterday! jim -------------- next part -------------- An HTML attachment was scrubbed... URL: From burke at thealmquists.net Wed Aug 24 06:10:42 2011 From: burke at thealmquists.net (Burke Almquist) Date: Wed, 24 Aug 2011 01:10:42 -0500 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E54840A.6080107@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> <4E54840A.6080107@astrofoto.org> Message-ID: <49AAA53D-A0B5-4DD7-BF69-721651AD308A@thealmquists.net> On Aug 23, 2011, at 11:54 PM, Roland Roberts wrote: > On 08/23/2011 03:48 PM, Gideon Romm wrote: >> The thin client chroot needs to be 32-bit for ltspfs to work. The >> server can be 64-bit. > > Well, I wiped my client install and redid it... > > ltsp-build-client --arch i386 > > set a root password on the client so I can log in, rebooted the client, not appears to be a 32-bit client > > -bash-4.1# uname -r > 2.6.32-131.6.1.el6.i686.thinclient > > But still no joy on localdev. Among other things, the device itself does not mount on the thin client. "df -l" doesn't show it mounted anywhere. But there is in entry in /var/run/ltspfs_fstab: > media/roland > /media/roland/usbdisk-sdf1 > > This may be a dumb question, but is the user on the thin client a member of the "fuse" group? From roland at astrofoto.org Wed Aug 24 06:19:51 2011 From: roland at astrofoto.org (Roland Roberts) Date: Wed, 24 Aug 2011 02:19:51 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <49AAA53D-A0B5-4DD7-BF69-721651AD308A@thealmquists.net> References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> <4E54840A.6080107@astrofoto.org> <49AAA53D-A0B5-4DD7-BF69-721651AD308A@thealmquists.net> Message-ID: <4E549807.3030906@astrofoto.org> On 08/24/2011 02:10 AM, Burke Almquist wrote: > This may be a dumb question, but is the user on the thin client a > member of the "fuse" group? -bash-4.1# groups roland roland : roland fuse cvs fvis family iraf But....it turns out the client and the server disagreed on what the gid of fuse should be. I changed the ID on /opt/ltsp/i386/etc/group to match the server and rebooted the client. Bingo! My drive now shows up. Thank you! roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From warren at togami.com Wed Aug 24 06:41:05 2011 From: warren at togami.com (Warren Togami Jr.) Date: Tue, 23 Aug 2011 20:41:05 -1000 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E54151F.1080600@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> <4E54151F.1080600@astrofoto.org> Message-ID: <4E549D01.8070802@togami.com> On 8/23/2011 11:01 AM, Roland Roberts wrote: > On 08/23/2011 03:48 PM, Gideon Romm wrote: >> The thin client chroot needs to be 32-bit for ltspfs to work. The >> server can be 64-bit. > > Bummer. Does the 32-bit client run the PAE kernel? Some of my "thin" > clients have a lot of memory. > cd /opt/ltsp/i386 setarch i386 chroot . yum install kernel yum remove kernel-thinclient ltsp-rewrap-latest-kernel exit ltsp-update-kernels I think this will switch your EL6 chroot from the custom PAE-disabled kernel to the EL6 standard PAE kernel. Warren From warren at togami.com Wed Aug 24 06:41:28 2011 From: warren at togami.com (Warren Togami Jr.) Date: Tue, 23 Aug 2011 20:41:28 -1000 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E5335EE.2050707@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> Message-ID: <4E549D18.1040704@togami.com> On 8/22/2011 7:09 PM, Roland Roberts wrote: > My local USB drives don't show up on my desktop. Apart from that obvious > symptom, I'm wondering where I look to find out what might be the > problem. I'm a member of group fuse, so the obvious is covered. > > My DHCP server is separate from my LTSP/TFTP server, and my home > directory is an NFS mount. The NFS server is the same as the DHCP > server. Oh yes, and accounts are all handled via NIS and the NIS master > is yet another host. I'm not sure if any of that matters or can confuse > how FUSE works, but thought I should mention it. > > Where do I look for errors? > > roland > Did you follow the step where you add the user to the fuse group? Warren From roland at astrofoto.org Wed Aug 24 07:23:58 2011 From: roland at astrofoto.org (Roland Roberts) Date: Wed, 24 Aug 2011 03:23:58 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E549D18.1040704@togami.com> References: <4E5335EE.2050707@astrofoto.org> <4E549D18.1040704@togami.com> Message-ID: <4E54A70E.3030309@astrofoto.org> On 08/24/2011 02:41 AM, Warren Togami Jr. wrote: > On 8/22/2011 7:09 PM, Roland Roberts wrote: >> My local USB drives don't show up on my desktop. Apart from that obvious >> symptom, I'm wondering where I look to find out what might be the >> problem. I'm a member of group fuse, so the obvious is covered. >> >> My DHCP server is separate from my LTSP/TFTP server, and my home >> directory is an NFS mount. The NFS server is the same as the DHCP >> server. Oh yes, and accounts are all handled via NIS and the NIS master >> is yet another host. I'm not sure if any of that matters or can confuse >> how FUSE works, but thought I should mention it. >> >> Where do I look for errors? >> >> roland >> > > Did you follow the step where you add the user to the fuse group? Yes...when you catch up with my last note on the topic, you'll see the problem turned out to be that gid on the client and the server was different. I'm not sure why except that perhaps it has something to do with the group ID coming from NIS instead of being local. roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From ltsp at symbio-technologies.com Wed Aug 24 10:47:37 2011 From: ltsp at symbio-technologies.com (Gideon Romm) Date: Wed, 24 Aug 2011 06:47:37 -0400 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: <4E54840A.6080107@astrofoto.org> References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> <4E54840A.6080107@astrofoto.org> Message-ID: This is what you should expect to see. The /media/$USER folder on the server is only visible by $USER - it is not even visible by root. This is part of ltspfs's built in security. With ltspfs, the user -and only the user (not root) has permission to see his drives's content under /media. If you mount it again and switch user to the user "roland", you should be able to ls /media/roland/usbdisk-sda1 In terms of the drive being mounted on the thin client, this is also how ltspfs works. In "steady state", the drive remains unmounted on the thin client. Only when you access the /media/roland/usbdisk-sda1 does the drive actually get automounted on the thin client. It will remain mounted until 2 seconds after being idle and then be unmounted again. This is why you can wait 2 seconds and then simply pull out the USB stick from the thin client without caring to "safely remove the USB device". -Gadi On Wed, Aug 24, 2011 at 12:54 AM, Roland Roberts wrote: > On 08/23/2011 03:48 PM, Gideon Romm wrote: >> >> The thin client chroot needs to be 32-bit for ltspfs to work. The >> server can be 64-bit. > > Well, I wiped my client install and redid it... > > ?ltsp-build-client --arch i386 > > set a root password on the client so I can log in, rebooted the client, not > appears to be a 32-bit client > > ?-bash-4.1# uname -r > ?2.6.32-131.6.1.el6.i686.thinclient > > But still no joy on localdev. Among other things, the device itself does not > mount on the thin client. "df -l" doesn't show it mounted anywhere. But > there is in entry in /var/run/ltspfs_fstab: > > ?-bash-4.1# cat /var/run/ltspfs_fstab > ?/dev/sdf1 /var/run/drives/usbdisk-sdf1 vfat defaults,utf8 0 0 > >> On Tue, Aug 23, 2011 at 1:21 PM, Roland Roberts >> ?wrote: >>> >>> On 08/23/2011 01:14 PM, Jim Kinney wrote: >>>> >>>> I have not read the docs being used so I can only comment generically: >>>> using Ubuntu cookbook-style instructions is likely to not work on a >>>> RedHat >>>> based system. SL-6 is a RedHat based system > > Again, this is the best I've got. Here's the last step in that cookbook: > > ?-bash-4.1# ssh -X -S /var/run/ldm_socket_vt1_192.168.3.36 192.168.3.36 > "/usr/sbin/ltspfsmounter /tmp add" > ?root at 192.168.3.36's password: > ?Wrote 281ddd7e5b4145e0540effae42e179db size 32, waiting > ?Error: /media/root is not mounted > ?Error: /media/root/tmp is not mounted > > but the directory is mounted. > > Note, if I try this: > > ?-bash-4.1# ssh -X -S /var/run/ldm_socket_2791_192.168.3.36 192.168.3.36 > "/usr/sbin/ltspfsmounter /tmp add" > ?Wrote 281ddd7e5b4145e0540effae42e179db size 32, waiting > ?fusermount: mount failed: Operation not permitted > ?Error: /tmp/.roland-ltspfs/tmp is not mounted > > This turns out to be the problem. On the server, /bin/fusermount is not > setuid root or setgid. I made it setuid root and try again, pulling the USB > stick and reinserting it. > > Again, /var/log/messages shows the device recognized, and it get put into > /var/run/ltspfs_fstab. This time I do see *something* on the server: > > ?[root at copernicus bin]# find /media/ > ?/media/ > ?/media/roland > ?/media/roland/usbdisk-sdf1 > > Okay, but the device is not mounted on the thin client so when I look on the > server I see this: > > ?[root at copernicus bin]# ls -l /media/roland/ > ?ls: cannot access /media/roland/usbdisk-sdf1: Permission denied > ?total 0 > ?d????????? ? ? ? ? ? ? ? ? ? ?? usbdisk-sdf1 > > And one last attempt; on the client I do this: > > ?-bash-4.1# ssh -X -S /var/run/ldm_socket_2791_192.168.3.36 192.168.3.36 > "/usr/sbin/ltspfsmounter /var/run/drives/usbdisk-sdf1 cleanup" > ?-bash-4.1# mount /dev/sdf1 /var/run/drives/usbdisk-sdf1 > ?-bash-4.1# ll /var/run/drives/usbdisk-sdf1/ > ?total 32432 > ?drwxr-xr-x 2 root root ? ? 8192 Apr 12 09:35 boot > ?drwxr-xr-x 3 root root ? ? 8192 Oct 22 ?2010 EFI > ?-rwxr-xr-x 1 root root ? ?18092 Oct 22 ?2010 GPL > ?drwxr-xr-x 2 root root ? ? 8192 Oct 22 ?2010 LiveOS > ?-rwxr-xr-x 1 root root 33150608 Aug 21 18:54 MaxImDL515.exe > ?drwxr-xr-x 2 root root ? ? 8192 Oct 22 ?2010 syslinux > ?-bash-4.1# ssh -X -S /var/run/ldm_socket_2791_192.168.3.36 192.168.3.36 > "/usr/sbin/ltspfsmounter /var/run/drives/usbdisk-sdf1 add" > ?Wrote 281ddd7e5b4145e0540effae42e179db size 32, waiting > ?-bash-4.1# ll /var/run/drives/usbdisk-sdf1/ > ?total 0 > > So something in the process of trying to mount the drive makes it disappear. > > I'm so close, but I'm missing something. Why isn't the client mounting the > drive? It should be, shouldn't it? Why does it disappear when I try to > manually mount it? > > roland > > -- > ? ? ? ? ? ? ? ? ? ? ? PGP Key ID: 66 BC 3B CD > Roland B. Roberts, PhD ? ? ? ? ? ? ? ? ? ? ? ? ? ? RL Enterprises > roland at rlenter.com ? ? ? ? ? ? ? ? ? ? ? ? ? ?6818 Madeline Court > roland at astrofoto.org ? ? ? ? ? ? ? ? ? ? ? ? ? Brooklyn, NY 11220 > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From dean.jones at oregonstate.edu Wed Aug 24 14:59:52 2011 From: dean.jones at oregonstate.edu (Dean Jones) Date: Wed, 24 Aug 2011 07:59:52 -0700 Subject: [K12OSN] configuration questions In-Reply-To: <1314162500.44065.YahooMailRC@web83602.mail.sp1.yahoo.com> References: <4E52927E.4000703@astrofoto.org> <4E5293F5.7020209@astrofoto.org> <4E5294AD.4020902@astrofoto.org> <4E52D13A.6010009@astrofoto.org> <1314162500.44065.YahooMailRC@web83602.mail.sp1.yahoo.com> Message-ID: Hi, I think the Sun Ray 2fs clients are MIPS processor based and not likely to work with k12 LTSP without a lot of work. On Tue, Aug 23, 2011 at 10:08 PM, xbc bojali wrote: > > Ok I am a new ?to this so I?thought?it would be a good idea to get some feed > back. I am starting ?the conversion for my kids to k12lltsp. ?I have?3?Sun > Ray 2FS Virtual Display Clients that I want to use for the thin clients, I > have just bought a HP Quad core to build as the host/server for them. ?What > is the Easiest way to build these? I started with Scientific Linux i86_64 > but was not able to get the??ltsp-server-tweaks, to work. And then read from > Rolands posts that i might need to use the 32bit not the 64? > any help would be appreciated. I just started this yesterday! > jim > From jomegat at jomegat.com Wed Aug 24 15:16:11 2011 From: jomegat at jomegat.com (Jomegat) Date: Wed, 24 Aug 2011 11:16:11 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E53D3C9.7090700@siddall.name> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> <4E53D3C9.7090700@siddall.name> Message-ID: <4E5515BB.1010803@jomegat.com> On 08/23/2011 12:22 PM, Jeff Siddall wrote: > On 08/22/2011 11:41 AM, Jomegat wrote: >> On 08/22/2011 10:52 AM, Jeff Siddall wrote: >>> On 08/22/2011 09:34 AM, Jomegat wrote: >>>> On 08/22/2011 08:50 AM, Gideon Romm wrote: >>>>> On Sun, Aug 21, 2011 at 12:49 PM, Jomegat wrote: >>>>> >>>>>> [root at monarch ~]# aplay >>>>>> ALSA lib confmisc.c:768:(parse_card) cannot find card '0' >>>>> >>>>> It seems ALSA cannot find card 0. >>>>> >>>>> Did you run this command on the server? >>>> >>>> Good question. I was logged into a TC and opened an xterm. I THINK that >>>> means I ran the command on the server. I do not know how to run >>>> anything >>>> on the TC, but would appreciate enlightenment. >>>> >>>>> >>>>> If YES => the alsa -> pulseaudio redirection is not configured and you >>>>> need to modify your /etc/asound.conf >>>> >>>> This could be it - there is no /etc/asound.conf. Perhaps I have a >>>> missing package? >>>> >>>>> >>>>> If NO (you ran the command on the thin client) => The thin client >>>>> cannot find an appropriate driver for the sound card. >>>> >>>> Thanks! >>> >>> Are you sure you are running K12Linux and not K12LTSP? K12Linux used >>> LTSP5 and I thought pulseaudio was one of the things that didn't work in >>> EL5. Most EL5 users run K12LTSP which was LTSP4. >>> >>> If it is LTSP4 I really have no idea how it is supposed to work because >>> I have never run anything less than LTSP5. >>> >>> It looks like LTSP4 is using ESD for network audio but I don't know if >>> the mic ever worked with that. >>> >>> You might end up having to go to K12Linux with Fedora or EL6. >> >> Sorry - I typed K12Linux when I meant K12LTSP. I guess I need EL6 then. >> I have two servers right now - one of them is at the school and is >> slated for retirement. The other is here on my desk at the office. I >> could do an install from scratch. >> >> My main requirement is to support the old 900MHz Celeron's we're using >> for TC's. As I've been reading the list, it looks like I need to be >> careful about what I upgrade to. EL6 supports old clients, right? > > No, unfortunately EL6 doesn't support "old" clients that well, but that > is really dependent on your definition of "old". I would say it does not > support _really_ old clients. Basically any CPU that will run i686 with > PAE is good enough -- maybe even i686 without PAE. i586 and below (we > are talking a decade old or more now) are probably a no-go. A Celeron > 900 (mobile I assume) should be fine. > > Before you go to far you might want to try a test by downloading a SL6 > live CD (i686 of course, not x86_64) and see if your clients will boot. > You can burn it on a USB stick if your clients don't have CD drives. > > Of course it is also possible that mics can be made to work with LTSP4 > but unfortunately someone else will have to help you with that. To be > honest I have not even tried mics on LTSP5, but in principle pulseaudio > should handle that. > > Jeff > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > Thanks to everyone who helped on this. I installed Centos 6 and followed the instructions at https://fedorahosted.org/k12linux/wiki/InstallGuide - and everything is working swimmingly! Yay! I have so far tested USB flash drives, audio (in and out), flash (no nspluginwrapper). I've tried it with the one of the TC's we are using as well as with three other PC's in my office. It's a nice face lift too BTW. This evening I will deploy. I'd like to especially thank Warran for his work on making that happen. -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Having a smoking section in a restaurant is like having a peeing section in a swimming pool. From warren at togami.com Wed Aug 24 19:24:44 2011 From: warren at togami.com (Warren Togami Jr.) Date: Wed, 24 Aug 2011 09:24:44 -1000 Subject: [K12OSN] Troubleshooting fuse? In-Reply-To: References: <4E5335EE.2050707@astrofoto.org> <4E53DCF7.5000700@astrofoto.org> <4E53E190.4030001@astrofoto.org> <4E54840A.6080107@astrofoto.org> Message-ID: <4E554FFC.5080204@togami.com> On 8/24/2011 12:47 AM, Gideon Romm wrote: > This is what you should expect to see. > > The /media/$USER folder on the server is only visible by $USER - it is > not even visible by root. This is part of ltspfs's built in security. > With ltspfs, the user -and only the user (not root) has permission to > see his drives's content under /media. If you mount it again and > switch user to the user "roland", you should be able to ls > /media/roland/usbdisk-sda1 > > In terms of the drive being mounted on the thin client, this is also > how ltspfs works. In "steady state", the drive remains unmounted on > the thin client. Only when you access the /media/roland/usbdisk-sda1 > does the drive actually get automounted on the thin client. It will > remain mounted until 2 seconds after being idle and then be unmounted > again. This is why you can wait 2 seconds and then simply pull out > the USB stick from the thin client without caring to "safely remove > the USB device". > > -Gadi > Although "it is safe to pull the USB device" is generally a bad thing to teach users. The nautilus error message when they attempt to "properly" remove the device confuses the user, so I looked into modifying the GNOME stack to make that work. https://fedorahosted.org/k12linux/wiki/EL6Status#LOWPRIORITYTODO I wrote some notes here. Warren From warren at togami.com Wed Aug 24 19:26:51 2011 From: warren at togami.com (Warren Togami Jr.) Date: Wed, 24 Aug 2011 09:26:51 -1000 Subject: [K12OSN] configuration questions In-Reply-To: References: <4E52927E.4000703@astrofoto.org> <4E5293F5.7020209@astrofoto.org> <4E5294AD.4020902@astrofoto.org> <4E52D13A.6010009@astrofoto.org> <1314162500.44065.YahooMailRC@web83602.mail.sp1.yahoo.com> Message-ID: <4E55507B.3050301@togami.com> I am not familiar with 2fs specifically, but if I understand Sun Ray, the X server doesn't run on the clients themselves. The clients have a proprietary Sunray remote desktop protocol to a Sun server where the X servers run. In the past I've used a Sparc Solaris server solely as the Sunray X servers, to XDMCP into a Linux LTSP terminal server. It mostly worked, without sound and remote USB. Had some difficulties with fonts. Warren On 8/24/2011 4:59 AM, Dean Jones wrote: > Hi, > > I think the Sun Ray 2fs clients are MIPS processor based and not > likely to work with k12 LTSP without a lot of work. > > > On Tue, Aug 23, 2011 at 10:08 PM, xbc bojali wrote: >> >> Ok I am a new to this so I thought it would be a good idea to get some feed >> back. I am starting the conversion for my kids to k12lltsp. I have 3 Sun >> Ray 2FS Virtual Display Clients that I want to use for the thin clients, I >> have just bought a HP Quad core to build as the host/server for them. What >> is the Easiest way to build these? I started with Scientific Linux i86_64 >> but was not able to get the ltsp-server-tweaks, to work. And then read from >> Rolands posts that i might need to use the 32bit not the 64? >> any help would be appreciated. I just started this yesterday! >> jim From warren at togami.com Wed Aug 24 19:30:50 2011 From: warren at togami.com (Warren Togami Jr.) Date: Wed, 24 Aug 2011 09:30:50 -1000 Subject: [K12OSN] Sound problems In-Reply-To: <4E5515BB.1010803@jomegat.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> <4E53D3C9.7090700@siddall.name> <4E5515BB.1010803@jomegat.com> Message-ID: <4E55516A.5020409@togami.com> On 8/24/2011 5:16 AM, Jomegat wrote: > > Thanks to everyone who helped on this. I installed Centos 6 and followed > the instructions at https://fedorahosted.org/k12linux/wiki/InstallGuide > - and everything is working swimmingly! > > Yay! I have so far tested USB flash drives, audio (in and out), flash > (no nspluginwrapper). I've tried it with the one of the TC's we are > using as well as with three other PC's in my office. It's a nice face > lift too BTW. > > This evening I will deploy. > > I'd like to especially thank Warran for his work on making that happen. > https://fedorahosted.org/k12linux/wiki/2011FundDrive Donations are still being accepted. =) I am now in a MBA program, and as a class project I plan on incorporating K12Linux as a 501(c)3 non-profit. Future donations after that point will go 100% to school/library/community center deployments or to pay contractors (developers) who implement fixes to LTSP bugs. For example, I don't have the hardware here to be able to test LTSP cluster capability, so I may pay LTSP upstream to implement, test and document LTSP cluster for K12Linux. Warren From lesmikesell at gmail.com Wed Aug 24 19:45:41 2011 From: lesmikesell at gmail.com (Les Mikesell) Date: Wed, 24 Aug 2011 14:45:41 -0500 Subject: [K12OSN] Sound problems In-Reply-To: <4E55516A.5020409@togami.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> <4E53D3C9.7090700@siddall.name> <4E5515BB.1010803@jomegat.com> <4E55516A.5020409@togami.com> Message-ID: <4E5554E5.6060203@gmail.com> On 8/24/2011 2:30 PM, Warren Togami Jr. wrote: > > https://fedorahosted.org/k12linux/wiki/2011FundDrive > Donations are still being accepted. =) > > I am now in a MBA program, and as a class project I plan on > incorporating K12Linux as a 501(c)3 non-profit. Future donations after > that point will go 100% to school/library/community center deployments > or to pay contractors (developers) who implement fixes to LTSP bugs. > In case this doesn't work out or turns out to be too much trouble to maintain, there are umbrella 501c's that accept new projects, like: http://sfconservancy.org/ that was featured on a recent floss weekly podcast: http://twit.tv/floss171 -- Les Mikesell lesmikesell at gmail.com From jomegat at jomegat.com Wed Aug 24 20:20:56 2011 From: jomegat at jomegat.com (Jomegat) Date: Wed, 24 Aug 2011 16:20:56 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E55516A.5020409@togami.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> <4E53D3C9.7090700@siddall.name> <4E5515BB.1010803@jomegat.com> <4E55516A.5020409@togami.com> Message-ID: <4E555D28.3070908@jomegat.com> On 08/24/2011 03:30 PM, Warren Togami Jr. wrote: > On 8/24/2011 5:16 AM, Jomegat wrote: >> >> Thanks to everyone who helped on this. I installed Centos 6 and followed >> the instructions at https://fedorahosted.org/k12linux/wiki/InstallGuide >> - and everything is working swimmingly! >> >> Yay! I have so far tested USB flash drives, audio (in and out), flash >> (no nspluginwrapper). I've tried it with the one of the TC's we are >> using as well as with three other PC's in my office. It's a nice face >> lift too BTW. >> >> This evening I will deploy. >> >> I'd like to especially thank Warran for his work on making that happen. >> > > https://fedorahosted.org/k12linux/wiki/2011FundDrive > Donations are still being accepted. =) It's on my list of things to donate to. I just haven't yet figured out how to make the case to SWMBO (She Who Must Be Obeyed). Maybe I'll just donate and then take the heat. > I am now in a MBA program, and as a class project I plan on > incorporating K12Linux as a 501(c)3 non-profit. Future donations after > that point will go 100% to school/library/community center deployments > or to pay contractors (developers) who implement fixes to LTSP bugs. > > For example, I don't have the hardware here to be able to test LTSP > cluster capability, so I may pay LTSP upstream to implement, test and > document LTSP cluster for K12Linux. What do you mean by "LTSP cluster capability?" Multiple terminal servers? What would you need to gain that capability? -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Having a smoking section in a restaurant is like having a peeing section in a swimming pool. From jim.kinney at gmail.com Wed Aug 24 20:53:43 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Wed, 24 Aug 2011 16:53:43 -0400 Subject: [K12OSN] Sound problems In-Reply-To: <4E555D28.3070908@jomegat.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> <4E53D3C9.7090700@siddall.name> <4E5515BB.1010803@jomegat.com> <4E55516A.5020409@togami.com> <4E555D28.3070908@jomegat.com> Message-ID: On Wed, Aug 24, 2011 at 4:20 PM, Jomegat wrote: > > >> For example, I don't have the hardware here to be able to test LTSP >> cluster capability, so I may pay LTSP upstream to implement, test and >> document LTSP cluster for K12Linux. >> > > What do you mean by "LTSP cluster capability?" Multiple terminal servers? > What would you need to gain that capability? > Think about a number of servers in a closet that are ready to answer a PXE boot request and join any client to the network that asks. There's a couple of ways I was approaching this post APS-project: 1. multiple servers providing the DHCP/PXE capabilities with no load balancing at all. So which ever server answered first would take the client (only works in pairs and is rather a network noise problem). 2. multiple LTSP servers with a single (or failover pair of) DHCP/PXE server(s) that would load balance based on feedback from the LTSP servers. This would allow for a much larger stack of servers for the LTSP side and the ability to do enterprise things like NAS/SAN/Fiber Channel storage, etc. Also allows for hardware replacement with minimal downtime for clients - tell dhcp server to not use a system then send a reboot to the clients (in between classes). 3. Cluster the servers using a "big" tool like OpenSSI, Mosix, or Kerrighed. This is the opposite of taking a big machine and splitting it out into smaller virtual machines. It takes a bunch of physical machines and creates a single, large virtual machine with many processors and lots of ram. The cat's meow of this is the ability to migrate processes from a failing hardware node in order to fix it. Additionally, physical nodes that are not maxed out can take load automatically to do node-wise load balancing. Plus it's way cool for kids to be able to say they use a supercomputer every day at school :-) -- -- James P. Kinney III As long as the general population is passive, apathetic, diverted to consumerism or hatred of the vulnerable, then the powerful can do as they please, and those who survive will be left to contemplate the outcome. - *2011 Noam Chomsky http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From peter at scheie.homedns.org Wed Aug 24 20:26:10 2011 From: peter at scheie.homedns.org (Peter Scheie) Date: Wed, 24 Aug 2011 15:26:10 -0500 Subject: [K12OSN] Sound problems In-Reply-To: <4E555D28.3070908@jomegat.com> References: <4E51370F.8070707@jomegat.com> <4E525AF9.9080800@jomegat.com> <4E526D20.5030807@siddall.name> <4E5278B9.8020806@jomegat.com> <4E53D3C9.7090700@siddall.name> <4E5515BB.1010803@jomegat.com> <4E55516A.5020409@togami.com> <4E555D28.3070908@jomegat.com> Message-ID: <4E555E62.8090207@scheie.homedns.org> Jomegat wrote: > On 08/24/2011 03:30 PM, Warren Togami Jr. wrote: >> On 8/24/2011 5:16 AM, Jomegat wrote: >>> >>> Thanks to everyone who helped on this. I installed Centos 6 and followed >>> the instructions at https://fedorahosted.org/k12linux/wiki/InstallGuide >>> - and everything is working swimmingly! >>> >>> Yay! I have so far tested USB flash drives, audio (in and out), flash >>> (no nspluginwrapper). I've tried it with the one of the TC's we are >>> using as well as with three other PC's in my office. It's a nice face >>> lift too BTW. >>> >>> This evening I will deploy. >>> >>> I'd like to especially thank Warran for his work on making that happen. >>> >> >> https://fedorahosted.org/k12linux/wiki/2011FundDrive >> Donations are still being accepted. =) > > It's on my list of things to donate to. I just haven't yet figured out > how to make the case to SWMBO (She Who Must Be Obeyed). Maybe I'll just > donate and then take the heat. > >> I am now in a MBA program, and as a class project I plan on >> incorporating K12Linux as a 501(c)3 non-profit. Future donations after >> that point will go 100% to school/library/community center deployments >> or to pay contractors (developers) who implement fixes to LTSP bugs. >> >> For example, I don't have the hardware here to be able to test LTSP >> cluster capability, so I may pay LTSP upstream to implement, test and >> document LTSP cluster for K12Linux. > > What do you mean by "LTSP cluster capability?" Multiple terminal > servers? What would you need to gain that capability? > Yes, that's what clustering means in the LTSP context. The folks at RevolutionLinux do this, with Ubuntu. (They are also some of the primary developers for LTSP and Ubuntu.) I believe at least some of that code has been pushed into upstream, but Warren knows more about that I do, I'm sure. See https://www.ltsp-cluster.org/ for some background. Peter From redbranchwarrior at gmail.com Thu Aug 25 01:48:16 2011 From: redbranchwarrior at gmail.com (Matthew Carter) Date: Wed, 24 Aug 2011 21:48:16 -0400 Subject: [K12OSN] when running ltsp-server-tweaks Message-ID: <4E55A9E0.6000001@gmail.com> Hi. I'm running freshly installed CENT-OS 6.0 on a Dell PE 2850 following directions from the install guide. I get to the point of running ltsp-server-tweaks, and I receive this error: "Failed to access configuration source(s): Failed to contact configuration server; some possible causes are that you need to enable TCP/IP networking for ORBit, or you have stale NFS locks due to a system crash." I have restarted the installation, reloaded the OS from scratch. Has anyone seen this? V/r, Matthew From bfristen at shaw.ca Thu Aug 25 12:44:22 2011 From: bfristen at shaw.ca (Brian Fristensky) Date: Thu, 25 Aug 2011 07:44:22 -0500 Subject: [K12OSN] configuration questions In-Reply-To: <4E55507B.3050301@togami.com> References: <4E52927E.4000703@astrofoto.org> <4E5293F5.7020209@astrofoto.org> <4E5294AD.4020902@astrofoto.org> <4E52D13A.6010009@astrofoto.org> <1314162500.44065.YahooMailRC@web83602.mail.sp1.yahoo.com> <4E55507B.3050301@togami.com> Message-ID: <4E5643A6.3060605@shaw.ca> I can report having seen a teaching lab full of SunRays, in which the server ran CentOS, not Solaris. (CentOS is a spin of Red Hat Enterprise Linux, very similar to Scientific Linux.) That at least tells us that SunRay server is available for Linux, but I don't know if that has been open-sourced or not. The SunRays had full sound and USB capability and worked quite well. Warren is correct in that LTSP is probably not going to work on a SunRay. For what it's worth, the only thing I have in my office at work is a SunRay, and I do everything on our university Sun Solaris system. For my money, it's better than having a computer in my office. Warren Togami Jr. wrote: > I am not familiar with 2fs specifically, but if I understand Sun Ray, > the X server doesn't run on the clients themselves. The clients have > a proprietary Sunray remote desktop protocol to a Sun server where the > X servers run. In the past I've used a Sparc Solaris server solely as > the Sunray X servers, to XDMCP into a Linux LTSP terminal server. It > mostly worked, without sound and remote USB. Had some difficulties > with fonts. > > Warren > > On 8/24/2011 4:59 AM, Dean Jones wrote: >> Hi, >> >> I think the Sun Ray 2fs clients are MIPS processor based and not >> likely to work with k12 LTSP without a lot of work. >> >> >> On Tue, Aug 23, 2011 at 10:08 PM, xbc bojali wrote: >>> >>> Ok I am a new to this so I thought it would be a good idea to get >>> some feed >>> back. I am starting the conversion for my kids to k12lltsp. I have >>> 3 Sun >>> Ray 2FS Virtual Display Clients that I want to use for the thin >>> clients, I >>> have just bought a HP Quad core to build as the host/server for >>> them. What >>> is the Easiest way to build these? I started with Scientific Linux >>> i86_64 >>> but was not able to get the ltsp-server-tweaks, to work. And then >>> read from >>> Rolands posts that i might need to use the 32bit not the 64? >>> any help would be appreciated. I just started this yesterday! >>> jim > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- ============================================ Brian Fristensky 971 Somerville Avenue Winnipeg MB R3T 1B4 CANADA bfristen at shaw.ca 204-261-3960 ============================================ From bojali at yahoo.com Thu Aug 25 14:29:29 2011 From: bojali at yahoo.com (xbc bojali) Date: Thu, 25 Aug 2011 07:29:29 -0700 (PDT) Subject: [K12OSN] configuration questions In-Reply-To: <4E5643A6.3060605@shaw.ca> References: <4E52927E.4000703@astrofoto.org> <4E5293F5.7020209@astrofoto.org> <4E5294AD.4020902@astrofoto.org> <4E52D13A.6010009@astrofoto.org> <1314162500.44065.YahooMailRC@web83602.mail.sp1.yahoo.com> <4E55507B.3050301@togami.com> <4E5643A6.3060605@shaw.ca> Message-ID: <1314282569.11688.YahooMailRC@web83604.mail.sp1.yahoo.com> Thanks, I have started with Oracle Linux Server and with a few DVD's burned and a few mistakes, I have it loaded on the kids server. from there i needed to install the Sun Ray Software, (and a VM client that is part of the suite). So, I have to rebuild the OLS due to install requirements SRS has! So I will start a list of steps on what do to.. if someone else is crazy enough to try is also.. ;^) So far all of the software need is available once I registered. Just a bit painful to find on the Oracle Distrubution site. jim ________________________________ From: Brian Fristensky To: Support list for open source software in schools. Sent: Thu, August 25, 2011 5:44:22 AM Subject: Re: [K12OSN] configuration questions I can report having seen a teaching lab full of SunRays, in which the server ran CentOS, not Solaris. (CentOS is a spin of Red Hat Enterprise Linux, very similar to Scientific Linux.) That at least tells us that SunRay server is available for Linux, but I don't know if that has been open-sourced or not. The SunRays had full sound and USB capability and worked quite well. Warren is correct in that LTSP is probably not going to work on a SunRay. For what it's worth, the only thing I have in my office at work is a SunRay, and I do everything on our university Sun Solaris system. For my money, it's better than having a computer in my office. Warren Togami Jr. wrote: > I am not familiar with 2fs specifically, but if I understand Sun Ray, the X >server doesn't run on the clients themselves. The clients have a proprietary >Sunray remote desktop protocol to a Sun server where the X servers run. In the >past I've used a Sparc Solaris server solely as the Sunray X servers, to XDMCP >into a Linux LTSP terminal server. It mostly worked, without sound and remote >USB. Had some difficulties with fonts. > > Warren > > On 8/24/2011 4:59 AM, Dean Jones wrote: >> Hi, >> >> I think the Sun Ray 2fs clients are MIPS processor based and not >> likely to work with k12 LTSP without a lot of work. >> >> >> On Tue, Aug 23, 2011 at 10:08 PM, xbc bojali wrote: >>> >>> Ok I am a new to this so I thought it would be a good idea to get some feed >>> back. I am starting the conversion for my kids to k12lltsp. I have 3 Sun >>> Ray 2FS Virtual Display Clients that I want to use for the thin clients, I >>> have just bought a HP Quad core to build as the host/server for them. What >>> is the Easiest way to build these? I started with Scientific Linux i86_64 >>> but was not able to get the ltsp-server-tweaks, to work. And then read from >>> Rolands posts that i might need to use the 32bit not the 64? >>> any help would be appreciated. I just started this yesterday! >>> jim > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- ============================================ Brian Fristensky 971 Somerville Avenue Winnipeg MB R3T 1B4 CANADA bfristen at shaw.ca 204-261-3960 ============================================ _______________________________________________ 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 burke at thealmquists.net Thu Aug 25 17:52:48 2011 From: burke at thealmquists.net (Burke Almquist) Date: Thu, 25 Aug 2011 12:52:48 -0500 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: <4E55A9E0.6000001@gmail.com> References: <4E55A9E0.6000001@gmail.com> Message-ID: <6934A36C-E4D5-41CC-A6A6-5E4A8DBC3754@thealmquists.net> I did see this, but everything seemed to work just fine afterwards, so I ignored it. On Aug 24, 2011, at 8:48 PM, Matthew Carter wrote: > Hi. I'm running freshly installed CENT-OS 6.0 on a Dell PE 2850 following directions from the install guide. I get to the point of running ltsp-server-tweaks, and I receive this error: > > "Failed to access configuration source(s): Failed to contact configuration server; some possible causes are that you need to enable TCP/IP networking for ORBit, or you have stale NFS locks due to a system crash." > > I have restarted the installation, reloaded the OS from scratch. Has anyone seen this? > > V/r, > > Matthew > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From bojali at yahoo.com Thu Aug 25 18:16:32 2011 From: bojali at yahoo.com (xbc bojali) Date: Thu, 25 Aug 2011 11:16:32 -0700 (PDT) Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: <6934A36C-E4D5-41CC-A6A6-5E4A8DBC3754@thealmquists.net> References: <4E55A9E0.6000001@gmail.com> <6934A36C-E4D5-41CC-A6A6-5E4A8DBC3754@thealmquists.net> Message-ID: <1314296192.20197.YahooMailRC@web83607.mail.sp1.yahoo.com> I saw that on one of my tries to install Scientific Linux and thought I made a mistake. (one of a few thousand) so was going to reinstall again. ________________________________ From: Burke Almquist To: Support list for open source software in schools. Sent: Thu, August 25, 2011 10:52:48 AM Subject: Re: [K12OSN] when running ltsp-server-tweaks I did see this, but everything seemed to work just fine afterwards, so I ignored it. On Aug 24, 2011, at 8:48 PM, Matthew Carter wrote: > Hi. I'm running freshly installed CENT-OS 6.0 on a Dell PE 2850 following >directions from the install guide. I get to the point of running >ltsp-server-tweaks, and I receive this error: > > "Failed to access configuration source(s): Failed to contact configuration >server; some possible causes are that you need to enable TCP/IP networking for >ORBit, or you have stale NFS locks due to a system crash." > > I have restarted the installation, reloaded the OS from scratch. Has anyone >seen this? > > V/r, > > Matthew > > _______________________________________________ > 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 jomegat at jomegat.com Fri Aug 26 18:46:02 2011 From: jomegat at jomegat.com (Jomegat) Date: Fri, 26 Aug 2011 14:46:02 -0400 Subject: [K12OSN] New EL6 install Message-ID: <4E57E9EA.6010909@jomegat.com> Hi everyone, Just thought I'd report some status. I have switched my school over to a new server running K12Linux EL6-64. Before, we were running on another server with K12LTSP EL5-64. So far so good, but I have a little more work to do. As has been widely reported, Adobe Flash is terrible on TC's, and my case is no exception. If Hurricane Irene allows, I will spend some time this weekend enabling local applications. I will probably also run ooimpress (if not the entire oo suite) locally, as running several slide shows at once brought my old system to its knees. During an open house. It was not pretty! I will probably also upgrade Firefox to something a little more up-to-date than 3.6. I don't know if it's related to the poor flash performance or what, but FF 3.6 has a tendency to put itself in the offline mode at the least convenient times (such as when students are trying to use the Internet). I did see some really strange things during the initial setup. The real head-scratcher was that I bought a managed switch on eBay a while back and decided to try it out here. It was to replace the 10/100 switch between my TC's and terminal server. The new switch has 24 10/100 ports and a pair of gigE ports. The new switch appears to work in every regard except - no TC can boot through it. For now I am stuck with my old 10/100 switch. The managed switch is a Dell PowerConnect 3024 (if memory serves), so if anyone knows how to coax that into working between a cluster of TC's and a terminal server, I'm all ears. The other really bizarre thing is something I ought to be punished for (and... I guess I was). I got my hands on a used Netgear 5-port gigE switch a while back. It was reported to be dodgy, but I ignored the warning. I have been using this switch on the WAN side of my terminal server for about a year with no problems. It's my "green" network in my IPCop setup, meaning it sits between my terminal server and the firewall/gateway. As I said, it was working great - until I plugged a second gigE machine into it. The second gigE connection cannot get an ip address via DHCP. It can get it just fine if I disconnect the other gigE machine, but then THAT one can't get an IP. No matter what, a second gigE machine cannot get on. I should write a warning on its case with permanent marker. To get around that, I temporarily replaced the dodgy Netgear switch with an 8-port 10BaseT hub that I had lying around. Ouch! I would have used the eBay switch instead, but it was at my office and not at the school - until last night anyhow. Yesterday after work I brought the managed switch in and retired the 10BaseT hub. It all seems to work pretty OK now. The other adventure I had was in migrating my users from the old box to the new box. I added their accounts and then rsync'd their home directories over, excluding all the dot files. Logins failed. After an hour of chasing that down I found a recipe that worked. Delete the user accts, delete their home directories, adduser, and then log in allowing Gnome to create its magical files. THEN rsync the files from the old server. I guess rsync doesn't overwrite new files, so the gnome was all happy. I was able to login again after all that. So as soon as I can get Flash and FF running as local apps, I think we'll be golden. I will eventually move my webserver from the old server to the new (we're running openbiblio on that), and once that's done, I will have a spare, unused, and unneeded server. Warran - maybe you want it? It's a 1U blade from Monarch with two CPU's. Here are some selected bits from /proc/cpuinfo: model name : AMD Opteron(tm) Processor 246 HE cpu MHz : 1992.406 bogomips : 3987.22 and from /proc/meminfo: MemTotal: 4059588 kB It has only a single IDE drive in it, but it does have a SATA bus (no SATA drives). It has NINE fans in it, and they are... not quiet. I replaced the two noisiest ones which brought the noise level down to something more reasonable. It sounds like a vacuum cleaner now instead of like a jet engine. Maybe that could help you with LTS clustering? I will have a much easier time getting cash to ship something rather than cash to donate. For some reason, that's more palatable to the powers that be. I also have an 8-port 10BaseT hub I can give you ;-) If you don't want the server, I'll be looking for suggestions on how best to put it to use. -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Eternity is a very long time, especially towards the end. - Stephen Hawking From lesmikesell at gmail.com Fri Aug 26 19:05:12 2011 From: lesmikesell at gmail.com (Les Mikesell) Date: Fri, 26 Aug 2011 14:05:12 -0500 Subject: [K12OSN] New EL6 install In-Reply-To: <4E57E9EA.6010909@jomegat.com> References: <4E57E9EA.6010909@jomegat.com> Message-ID: On Fri, Aug 26, 2011 at 1:46 PM, Jomegat wrote: > I did see some really strange things during the initial setup. ?The real > head-scratcher was that I bought a managed switch on eBay a while back and > decided to try it out here. ?It was to replace the 10/100 switch between my > TC's and terminal server. ?The new switch has 24 10/100 ports and a pair of > gigE ports. ?The new switch appears to work in every regard except - no TC > can boot through it. ?For now I am stuck with my old 10/100 switch. ?The > managed switch is a Dell PowerConnect 3024 (if memory serves), so if anyone > knows how to coax that into working between a cluster of TC's and a terminal > server, I'm all ears. Switches are supposed to do a spanning-tree check as new nodes appear and may block for 30 seconds or so - which may be long enough for a PXE boot attempt to time out and restart the cycle. Look for a command to do 'rapid' spanning tree or to forward while learning. -- Les Mikesell lesmikesell at gmail.com From jomegat at jomegat.com Fri Aug 26 19:22:56 2011 From: jomegat at jomegat.com (Jomegat) Date: Fri, 26 Aug 2011 15:22:56 -0400 Subject: [K12OSN] New EL6 install In-Reply-To: References: <4E57E9EA.6010909@jomegat.com> Message-ID: <4E57F290.40508@jomegat.com> On 08/26/2011 03:05 PM, Les Mikesell wrote: > On Fri, Aug 26, 2011 at 1:46 PM, Jomegat wrote: > >> I did see some really strange things during the initial setup. The real >> head-scratcher was that I bought a managed switch on eBay a while back and >> decided to try it out here. It was to replace the 10/100 switch between my >> TC's and terminal server. The new switch has 24 10/100 ports and a pair of >> gigE ports. The new switch appears to work in every regard except - no TC >> can boot through it. For now I am stuck with my old 10/100 switch. The >> managed switch is a Dell PowerConnect 3024 (if memory serves), so if anyone >> knows how to coax that into working between a cluster of TC's and a terminal >> server, I'm all ears. > > Switches are supposed to do a spanning-tree check as new nodes appear > and may block for 30 seconds or so - which may be long enough for a > PXE boot attempt to time out and restart the cycle. Look for a > command to do 'rapid' spanning tree or to forward while learning. > Thanks! I have never used a managed switch before, so this is all new to me. I'll look for that. -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Eternity is a very long time, especially towards the end. - Stephen Hawking From jpeltier at sfu.ca Fri Aug 26 19:59:20 2011 From: jpeltier at sfu.ca (James A. Peltier) Date: Fri, 26 Aug 2011 12:59:20 -0700 (PDT) Subject: [K12OSN] New EL6 install In-Reply-To: <4E57F290.40508@jomegat.com> Message-ID: <353157261.366426.1314388760487.JavaMail.root@jaguar10.sfu.ca> ----- Original Message ----- | On 08/26/2011 03:05 PM, Les Mikesell wrote: | > On Fri, Aug 26, 2011 at 1:46 PM, Jomegat wrote: | > | >> I did see some really strange things during the initial setup. The | >> real | >> head-scratcher was that I bought a managed switch on eBay a while | >> back and | >> decided to try it out here. It was to replace the 10/100 switch | >> between my | >> TC's and terminal server. The new switch has 24 10/100 ports and a | >> pair of | >> gigE ports. The new switch appears to work in every regard except - | >> no TC | >> can boot through it. For now I am stuck with my old 10/100 switch. | >> The | >> managed switch is a Dell PowerConnect 3024 (if memory serves), so | >> if anyone | >> knows how to coax that into working between a cluster of TC's and a | >> terminal | >> server, I'm all ears. | > | > Switches are supposed to do a spanning-tree check as new nodes | > appear | > and may block for 30 seconds or so - which may be long enough for a | > PXE boot attempt to time out and restart the cycle. Look for a | > command to do 'rapid' spanning tree or to forward while learning. | > | | Thanks! I have never used a managed switch before, so this is all new | to me. I'll look for that. Different switches use different terms as well, you might see portfast (Cisco), client (Dell), other (Other) so read your documentation for your switch to determine how to properly set up Spanning Tree. We've been forced to set our CISCO switches to portfast for PXE to work, but our HPs don't have this issue. -- James A. Peltier IT Services - Research Computing Group Simon Fraser University - Burnaby Campus Phone : 778-782-6573 Fax : 778-782-3045 E-Mail : jpeltier at sfu.ca Website : http://www.sfu.ca/itservices http://blogs.sfu.ca/people/jpeltier I will do the best I can with the talent I have From roland at astrofoto.org Fri Aug 26 20:13:42 2011 From: roland at astrofoto.org (Roland Roberts) Date: Fri, 26 Aug 2011 16:13:42 -0400 Subject: [K12OSN] New EL6 install In-Reply-To: <4E57E9EA.6010909@jomegat.com> References: <4E57E9EA.6010909@jomegat.com> Message-ID: <4E57FE76.6090806@astrofoto.org> On 08/26/2011 02:46 PM, Jomegat wrote: > Just thought I'd report some status. I have switched my school over > to a new server running K12Linux EL6-64. Before, we were running on > another server with K12LTSP EL5-64. So far so good, but I have a > little more work to do. > > As has been widely reported, Adobe Flash is terrible on TC's, and my > case is no exception. If Hurricane Irene allows, I will spend some > time this weekend enabling local applications. I will probably also > run ooimpress (if not the entire oo suite) locally, as running several > slide shows at once brought my old system to its knees. During an > open house. It was not pretty! I installed the 64-bit Adobe plugin in my own firefox plugins directory and it *seems* to be working fine for me. At least I can do youtube without any problems. roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From jomegat at jomegat.com Fri Aug 26 20:22:48 2011 From: jomegat at jomegat.com (Jomegat) Date: Fri, 26 Aug 2011 16:22:48 -0400 Subject: [K12OSN] New EL6 install In-Reply-To: <4E57FE76.6090806@astrofoto.org> References: <4E57E9EA.6010909@jomegat.com> <4E57FE76.6090806@astrofoto.org> Message-ID: <4E580098.1030209@jomegat.com> On 08/26/2011 04:13 PM, Roland Roberts wrote: > On 08/26/2011 02:46 PM, Jomegat wrote: >> Just thought I'd report some status. I have switched my school over to >> a new server running K12Linux EL6-64. Before, we were running on >> another server with K12LTSP EL5-64. So far so good, but I have a >> little more work to do. >> >> As has been widely reported, Adobe Flash is terrible on TC's, and my >> case is no exception. If Hurricane Irene allows, I will spend some >> time this weekend enabling local applications. I will probably also >> run ooimpress (if not the entire oo suite) locally, as running several >> slide shows at once brought my old system to its knees. During an open >> house. It was not pretty! > > I installed the 64-bit Adobe plugin in my own firefox plugins directory > and it *seems* to be working fine for me. At least I can do youtube > without any problems. I had no trouble with youtube either. But there was another flash-heavy site that had real problems. I'm not sure what site the students were using, but it brought everything to its knees whatever it was. -- Jim Thomas Principal Applications Engineer Bittware, Inc jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 Eternity is a very long time, especially towards the end. - Stephen Hawking From roland at astrofoto.org Fri Aug 26 20:24:08 2011 From: roland at astrofoto.org (Roland Roberts) Date: Fri, 26 Aug 2011 16:24:08 -0400 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? Message-ID: <4E5800E8.1040602@astrofoto.org> Up until recently, my network was all Fedora 14. However, I turned on machine into SL 6.1 to act as an LTSP server. After doing so, I've run into some odd NFS performance issues. They show up as simply slow performance when doing builds, and I can't track down why. This is not really an LTSP issue, but I thought I'd check here since I thought people here might be more likely to have a larger heterogeneous network than the Fedora groups. I'll probably post there as well to see if anyone can give me some clues. For example, I when I did an RPM build of the 64-bit Sun Java 1.6.0.26 package, it consistently takes 16-18 minutes when running on my LTSP server from my home directory with my NFS mounted home directory. The NFS server shows no significant load, the client (which is the LTSP server) also shows little load but top report about 50% wait times. If I run the build using a directory local to the LTSP server, the time drops to 2-3 minutes. The LTSP server is a Core2 Quad 3.0 GHz host running SL 6.1. The NFS server is a Core2 Quad 2.4 GHz running Fedora 14 I have a second Core2 Quad 3.0 GHz host running SL 6.1. If I do the same set of tests there, the local file system build drops to under a minute, and the NFS build to 2-3 minutes. Clearly I get a performance hit going over the network, but the hit on the SL 6.1 LTSP server is rough to take. Does anyone here use NFS extensively for the LTSP and have encountered any performance problems? roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From monteslu at cox.net Fri Aug 26 20:53:42 2011 From: monteslu at cox.net (Luis Montes) Date: Fri, 26 Aug 2011 13:53:42 -0700 Subject: [K12OSN] New EL6 install In-Reply-To: <4E57FE76.6090806@astrofoto.org> References: <4E57E9EA.6010909@jomegat.com> <4E57FE76.6090806@astrofoto.org> Message-ID: <4E5807D6.9070804@cox.net> Roland Roberts wrote: > On 08/26/2011 02:46 PM, Jomegat wrote: >> Just thought I'd report some status. I have switched my school over >> to a new server running K12Linux EL6-64. Before, we were running on >> another server with K12LTSP EL5-64. So far so good, but I have a >> little more work to do. >> >> As has been widely reported, Adobe Flash is terrible on TC's, and my >> case is no exception. If Hurricane Irene allows, I will spend some >> time this weekend enabling local applications. I will probably also >> run ooimpress (if not the entire oo suite) locally, as running >> several slide shows at once brought my old system to its knees. >> During an open house. It was not pretty! > > I installed the 64-bit Adobe plugin in my own firefox plugins > directory and it *seems* to be working fine for me. At least I can do > youtube without any problems. > > roland > Have either of you guys tried Chrome on the thin clients yet? This is the main reason I want to update to SL6 or Centos6. Luis From jim.kinney at gmail.com Fri Aug 26 21:09:32 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Fri, 26 Aug 2011 17:09:32 -0400 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: <4E5800E8.1040602@astrofoto.org> References: <4E5800E8.1040602@astrofoto.org> Message-ID: Tuning NFS is an art form. First make sure to use v4. The block sizes need tweaks to max throughputs. There may also need to be a kernel param tweaked but I'm not near my notes. On Aug 26, 2011 4:28 PM, "Roland Roberts" wrote: > Up until recently, my network was all Fedora 14. However, I turned on > machine into SL 6.1 to act as an LTSP server. After doing so, I've run > into some odd NFS performance issues. They show up as simply slow > performance when doing builds, and I can't track down why. This is not > really an LTSP issue, but I thought I'd check here since I thought > people here might be more likely to have a larger heterogeneous network > than the Fedora groups. I'll probably post there as well to see if > anyone can give me some clues. > > For example, I when I did an RPM build of the 64-bit Sun Java 1.6.0.26 > package, it consistently takes 16-18 minutes when running on my LTSP > server from my home directory with my NFS mounted home directory. The > NFS server shows no significant load, the client (which is the LTSP > server) also shows little load but top report about 50% wait times. > > If I run the build using a directory local to the LTSP server, the time > drops to 2-3 minutes. > > The LTSP server is a Core2 Quad 3.0 GHz host running SL 6.1. The NFS > server is a Core2 Quad 2.4 GHz running Fedora 14 > > I have a second Core2 Quad 3.0 GHz host running SL 6.1. If I do the > same set of tests there, the local file system build drops to under a > minute, and the NFS build to 2-3 minutes. Clearly I get a performance > hit going over the network, but the hit on the SL 6.1 LTSP server is > rough to take. > > Does anyone here use NFS extensively for the LTSP and have encountered > any performance problems? > > roland > > -- > PGP Key ID: 66 BC 3B CD > Roland B. Roberts, PhD RL Enterprises > roland at rlenter.com 6818 Madeline Court > roland at astrofoto.org Brooklyn, NY 11220 > > _______________________________________________ > 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 roland at astrofoto.org Fri Aug 26 21:28:07 2011 From: roland at astrofoto.org (Roland Roberts) Date: Fri, 26 Aug 2011 17:28:07 -0400 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: References: <4E5800E8.1040602@astrofoto.org> Message-ID: <4E580FE7.7000505@astrofoto.org> On 08/26/2011 05:09 PM, Jim Kinney wrote: > > Tuning NFS is an art form. First make sure to use v4. The block sizes > need tweaks to max throughputs. There may also need to be a kernel > param tweaked but I'm not near my notes. > I'm trying to figure out a good tool for testing. I tried the simple dd if=/dev/zero .... but I get amazingly fast speeds with that. Of course, that could me its directory ops that are killing me, not raw write speeds. > > > > The LTSP server is a Core2 Quad 3.0 GHz host running SL 6.1. The NFS > > server is a Core2 Quad 2.4 GHz running Fedora 14 > > > > I have a second Core2 Quad 3.0 GHz host running SL 6.1. If I do the Oops, that was supposed to say, "...running Fedora 14...". This was my control with Fedora client/Fedora server. > > same set of tests there, the local file system build drops to under a > > minute, and the NFS build to 2-3 minutes. Clearly I get a performance > > hit going over the network, but the hit on the SL 6.1 LTSP server is > > rough to take. roalnd -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From robark at gmail.com Fri Aug 26 21:41:59 2011 From: robark at gmail.com (Robert Arkiletian) Date: Fri, 26 Aug 2011 14:41:59 -0700 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: References: <4E5800E8.1040602@astrofoto.org> Message-ID: On Fri, Aug 26, 2011 at 2:09 PM, Jim Kinney wrote: > Tuning NFS is an art form. First make sure to use v4. The block sizes need > tweaks to max throughputs. There may also need to be a kernel param tweaked > but I'm not near my notes. > Are you sure? The advice I see is to use v3. I have been using v3 in F11 and F13 with no problems. I haven't used v4 yet though. Here is a nice page about nfs optimizations. http://nfs.sourceforge.net/nfs-howto/ar01s05.html also use no_subtree_check in /etc/exports -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From roland at astrofoto.org Fri Aug 26 22:10:24 2011 From: roland at astrofoto.org (Roland Roberts) Date: Fri, 26 Aug 2011 18:10:24 -0400 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: References: <4E5800E8.1040602@astrofoto.org> Message-ID: <4E5819D0.4030801@astrofoto.org> On 08/26/2011 05:41 PM, Robert Arkiletian wrote: > On Fri, Aug 26, 2011 at 2:09 PM, Jim Kinney wrote: >> Tuning NFS is an art form. First make sure to use v4. The block sizes need >> tweaks to max throughputs. There may also need to be a kernel param tweaked >> but I'm not near my notes. >> > Are you sure? The advice I see is to use v3. I have been using v3 in > F11 and F13 with no problems. I haven't used v4 yet though. I think 4 is the default on F14 and EL6.x. I see it reported as available in rpcinfo -s. I was trying to see if there was some magic to force a version or prevent certain versions from running on the server to be sure of what's being used, but I can't find it. According to the docs, the version is negotiated starting at 4 and working down. > Here is a nice page about nfs optimizations. > http://nfs.sourceforge.net/nfs-howto/ar01s05.html I was staring at it even as your message arrived :-) > also use no_subtree_check in /etc/exports I did add this earlier today and looked to see what cruft I might have in my auto.home map. I keep bouncing the daemon on another exported filesystem so I can test. be happy with factor of 2 difference between local and nfs, but that factor of 6 is hurting. I'll try a few things and report back. roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From jomegat at jomegat.com Fri Aug 26 22:27:33 2011 From: jomegat at jomegat.com (Jomegat) Date: Fri, 26 Aug 2011 18:27:33 -0400 Subject: [K12OSN] New EL6 install In-Reply-To: <4E5807D6.9070804@cox.net> References: <4E57E9EA.6010909@jomegat.com> <4E57FE76.6090806@astrofoto.org> <4E5807D6.9070804@cox.net> Message-ID: <4E581DD5.6030407@jomegat.com> On 08/26/2011 04:53 PM, Luis Montes wrote: > Have either of you guys tried Chrome on the thin clients yet? This is > the main reason I want to update to SL6 or Centos6. > Luis I have not. It's a little too scary for me to have Google inside my browser. It's one thing when they collect your data when you visit their site. It's another when they collect it when you visit any site. Don't get me wrong - I think that what Chrome is doing for the browser now is what Firefox did five-ten years ago. Ever since they released it, all browsers have been focusing on speed - competition is good. -- Jim Thomas jomegat at jomegat.com Dyslexics of the world, UNTIE! From robark at gmail.com Fri Aug 26 23:04:54 2011 From: robark at gmail.com (Robert Arkiletian) Date: Fri, 26 Aug 2011 16:04:54 -0700 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: <4E5819D0.4030801@astrofoto.org> References: <4E5800E8.1040602@astrofoto.org> <4E5819D0.4030801@astrofoto.org> Message-ID: On Fri, Aug 26, 2011 at 3:10 PM, Roland Roberts wrote: > > I think 4 is the default on F14 and EL6.x. I see it reported as available in > rpcinfo -s. I was trying to see if there was some magic to force a version > or prevent certain versions from running on the server to be sure of what's > being used, but I can't find it. According to the docs, the version is > negotiated starting at 4 and working down. > you can use nfsvers=3 as one of the client mount options -- Robert Arkiletian Eric Hamber Secondary, Vancouver, Canada From monteslu at cox.net Fri Aug 26 23:23:33 2011 From: monteslu at cox.net (Luis Montes) Date: Fri, 26 Aug 2011 16:23:33 -0700 Subject: [K12OSN] New EL6 install In-Reply-To: <4E581DD5.6030407@jomegat.com> References: <4E57E9EA.6010909@jomegat.com> <4E57FE76.6090806@astrofoto.org> <4E5807D6.9070804@cox.net> <4E581DD5.6030407@jomegat.com> Message-ID: <4E582AF5.9050208@cox.net> Jomegat wrote: > On 08/26/2011 04:53 PM, Luis Montes wrote: >> Have either of you guys tried Chrome on the thin clients yet? This is >> the main reason I want to update to SL6 or Centos6. >> Luis > > I have not. It's a little too scary for me to have Google inside my > browser. It's one thing when they collect your data when you visit > their site. It's another when they collect it when you visit any site. > > Don't get me wrong - I think that what Chrome is doing for the browser > now is what Firefox did five-ten years ago. Ever since they released > it, all browsers have been focusing on speed - competition is good. > Well, I like it as a browser even when I don't sign into a google account. It's V8 javascript engine is amazing. The HTML5 stuff is gonna make flash obsolete. And I'm somehow, someway, someday, gonna get WebGL working on thin clients that have an openGL capable GPU. So I built a Centos6 server in vmware and a pxeclient as well. Ran the K12Linux setup, configured the bridge to be the host-only network. PXE client booted up! Then, I decided to follow the https://fedorahosted.org/k12linux/wiki/LocalAppsSupport guide to get firefox running. Firefox runs, but seems like localapp networking is broken for me. I think I have to fix some routing so that I can NAT through the ltsp server or something like that. Or maybe just kill the bridge and make the new ltsp server my dhcp server for the whole network. Tuxpaint also installed without any issue. The younger kids are gonna love having this as a local app. Next, I got the i386 version of Chrome installed. The rpm had some dependencies that I was able to fix with yum. The one that took a while to figure out was libXss.so. Apparently that one is bundled in libXScrnSvr which yum was ok with installing. Chrome is showing a bunch of stuff accelerated in about:gpu at least in my vmware client. Can't wait to try this out on real hardware. Long story short: Warren is doing an incredible job with this stuff. Good times ahead for LTSP! Luis From jim.kinney at gmail.com Sat Aug 27 00:19:28 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Fri, 26 Aug 2011 20:19:28 -0400 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: <4E5819D0.4030801@astrofoto.org> References: <4E5800E8.1040602@astrofoto.org> <4E5819D0.4030801@astrofoto.org> Message-ID: You can force a version in the mount command. Mount -t nfs4 will use v4 and fail if if only v3 is available. On Aug 26, 2011 6:14 PM, "Roland Roberts" wrote: > On 08/26/2011 05:41 PM, Robert Arkiletian wrote: >> On Fri, Aug 26, 2011 at 2:09 PM, Jim Kinney wrote: >>> Tuning NFS is an art form. First make sure to use v4. The block sizes need >>> tweaks to max throughputs. There may also need to be a kernel param tweaked >>> but I'm not near my notes. >>> >> Are you sure? The advice I see is to use v3. I have been using v3 in >> F11 and F13 with no problems. I haven't used v4 yet though. > > I think 4 is the default on F14 and EL6.x. I see it reported as > available in rpcinfo -s. I was trying to see if there was some magic to > force a version or prevent certain versions from running on the server > to be sure of what's being used, but I can't find it. According to the > docs, the version is negotiated starting at 4 and working down. > >> Here is a nice page about nfs optimizations. >> http://nfs.sourceforge.net/nfs-howto/ar01s05.html > > I was staring at it even as your message arrived :-) >> also use no_subtree_check in /etc/exports > > I did add this earlier today and looked to see what cruft I might have > in my auto.home map. I keep bouncing the daemon on another exported > filesystem so I can test. be happy with factor of 2 difference between > local and nfs, but that factor of 6 is hurting. > > I'll try a few things and report back. > > roland > > -- > PGP Key ID: 66 BC 3B CD > Roland B. Roberts, PhD RL Enterprises > roland at rlenter.com 6818 Madeline Court > roland at astrofoto.org Brooklyn, NY 11220 > > _______________________________________________ > 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 burke at thealmquists.net Sat Aug 27 00:58:46 2011 From: burke at thealmquists.net (burke at thealmquists.net) Date: Fri, 26 Aug 2011 19:58:46 -0500 Subject: [K12OSN] New EL6 install Message-ID: You might try a plugin that doesn't load flash until you click on it. That seems to fix the worst websites. Jomegat wrote: >On 08/26/2011 04:13 PM, Roland Roberts wrote: >> On 08/26/2011 02:46 PM, Jomegat wrote: >>> Just thought I'd report some status. I have switched my school over to >>> a new server running K12Linux EL6-64. Before, we were running on >>> another server with K12LTSP EL5-64. So far so good, but I have a >>> little more work to do. >>> >>> As has been widely reported, Adobe Flash is terrible on TC's, and my >>> case is no exception. If Hurricane Irene allows, I will spend some >>> time this weekend enabling local applications. I will probably also >>> run ooimpress (if not the entire oo suite) locally, as running several >>> slide shows at once brought my old system to its knees. During an open >>> house. It was not pretty! >> >> I installed the 64-bit Adobe plugin in my own firefox plugins directory >> and it *seems* to be working fine for me. At least I can do youtube >> without any problems. > >I had no trouble with youtube either. But there was another flash-heavy >site that had real problems. I'm not sure what site the students were >using, but it brought everything to its knees whatever it was. > > >-- >Jim Thomas Principal Applications Engineer Bittware, Inc >jthomas at bittware.com http://www.bittware.com (603) 226-0404 x536 >Eternity is a very long time, especially towards the end. - Stephen Hawking > >_______________________________________________ >K12OSN mailing list >K12OSN at redhat.com >https://www.redhat.com/mailman/listinfo/k12osn >For more info see From jomegat at jomegat.com Sun Aug 28 15:11:56 2011 From: jomegat at jomegat.com (Jomegat) Date: Sun, 28 Aug 2011 11:11:56 -0400 Subject: [K12OSN] Local Apps and Firefox 6 Message-ID: <4E5A5ABC.9020204@jomegat.com> Hi everyone, I'm planning to run Firefox as a local application and I'm looking for some advice. First I tried following the instructions at https://fedorahosted.org/k12linux/wiki/LocalAppsSupport but when I got to the yum part, there was no joy: [root at termserv i386]# setarch i386 chroot . bash-4.1# yum install firefox Error: Cannot retrieve repository metadata (repomd.xml) for repository: epel. Please verify its path and try again I thought - maybe I need to run ltsp-server-tweaks, since I do not recall having run that when I last rebooted (but everything seemed to work then, so... I dunno?) Here's what happens when I try running that: [root at termserv ~]# ltsp-server-tweaks Failed to access configuration source(s): Failed to contact configuration server; some possible causes are that you need to enable TCP/IP networking for ORBit, or you have stale NFS locks due to a system crash. See http://projects.gnome.org/gconf/ for information. (Details - 1: Not running within active session) Hmmm. Not sure what all that means. I tried all this while remotely logged in, as Tropical Storm Irene might make it a bit dodgy to get to the school (or get home afterwards). I don't think that's significant, but just in case it is, I'm disclosing it here. My next thought was that maybe I should try FF 6 instead of 3.6 (which is what yum will install). Is there a reason I should avoid that? Thanks in advance! -- Jim Thomas jomegat at jomegat.com Dyslexics of the world, UNTIE! From robark at gmail.com Sun Aug 28 18:10:46 2011 From: robark at gmail.com (Robert Arkiletian) Date: Sun, 28 Aug 2011 11:10:46 -0700 Subject: [K12OSN] Local Apps and Firefox 6 In-Reply-To: <4E5A5ABC.9020204@jomegat.com> References: <4E5A5ABC.9020204@jomegat.com> Message-ID: On Sun, Aug 28, 2011 at 8:11 AM, Jomegat wrote: > Hi everyone, > > I'm planning to run Firefox as a local application and I'm looking for some > advice. ?First I tried following the instructions at > https://fedorahosted.org/k12linux/wiki/LocalAppsSupport Not 100% sure as I can't verify this. It's been a while, but check if /proc is mounted inside the chroot. I think you used to have to mount /proc inside the chroot before you do yum install. > > but when I got to the yum part, there was no joy: > > [root at termserv i386]# setarch i386 chroot . > bash-4.1# yum install firefox > Error: Cannot retrieve repository metadata (repomd.xml) for repository: > epel. Please verify its path and try again > > I thought - maybe I need to run ltsp-server-tweaks, since I do not recall > having run that when I last rebooted (but everything seemed to work then, > so... I dunno?) > > Here's what happens when I try running that: > [root at termserv ~]# ltsp-server-tweaks > Failed to access configuration source(s): Failed to contact configuration > server; some possible causes are that you need to enable TCP/IP networking > for ORBit, or you have stale NFS locks due to a system crash. See > http://projects.gnome.org/gconf/ for information. (Details - ?1: Not running > within active session) > > Hmmm. ?Not sure what all that means. > > I tried all this while remotely logged in, as Tropical Storm Irene might > make it a bit dodgy to get to the school (or get home afterwards). ?I don't > think that's significant, but just in case it is, I'm disclosing it here. > > My next thought was that maybe I should try FF 6 instead of 3.6 (which is > what yum will install). ?Is there a reason I should avoid that? > > Thanks in advance! > > -- > Jim Thomas ? ? ? ? ? ?jomegat at jomegat.com > Dyslexics of the world, UNTIE! > > _______________________________________________ > 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 jomegat at jomegat.com Sun Aug 28 18:29:20 2011 From: jomegat at jomegat.com (Jomegat) Date: Sun, 28 Aug 2011 14:29:20 -0400 Subject: [K12OSN] Local Apps and Firefox 6 In-Reply-To: References: <4E5A5ABC.9020204@jomegat.com> Message-ID: <4E5A8900.7060709@jomegat.com> On 08/28/2011 02:10 PM, Robert Arkiletian wrote: > On Sun, Aug 28, 2011 at 8:11 AM, Jomegat wrote: >> Hi everyone, >> >> I'm planning to run Firefox as a local application and I'm looking for some >> advice. First I tried following the instructions at >> https://fedorahosted.org/k12linux/wiki/LocalAppsSupport > > Not 100% sure as I can't verify this. It's been a while, but check if > /proc is mounted inside the chroot. I think you used to have to > > mount /proc > > inside the chroot before you do yum install. If that helped, it didn't help enough: [root at termserv ~]# cd /opt/ltsp/i386/ [root at termserv i386]# setarch i386 chroot . bash-4.1# mount /proc bash-4.1# yum install firefox Error: Cannot retrieve repository metadata (repomd.xml) for repository: epel. Please verify its path and try again bash-4.1# > >> >> but when I got to the yum part, there was no joy: >> >> [root at termserv i386]# setarch i386 chroot . >> bash-4.1# yum install firefox >> Error: Cannot retrieve repository metadata (repomd.xml) for repository: >> epel. Please verify its path and try again >> >> I thought - maybe I need to run ltsp-server-tweaks, since I do not recall >> having run that when I last rebooted (but everything seemed to work then, >> so... I dunno?) >> >> Here's what happens when I try running that: >> [root at termserv ~]# ltsp-server-tweaks >> Failed to access configuration source(s): Failed to contact configuration >> server; some possible causes are that you need to enable TCP/IP networking >> for ORBit, or you have stale NFS locks due to a system crash. See >> http://projects.gnome.org/gconf/ for information. (Details - 1: Not running >> within active session) >> >> Hmmm. Not sure what all that means. >> >> I tried all this while remotely logged in, as Tropical Storm Irene might >> make it a bit dodgy to get to the school (or get home afterwards). I don't >> think that's significant, but just in case it is, I'm disclosing it here. >> >> My next thought was that maybe I should try FF 6 instead of 3.6 (which is >> what yum will install). Is there a reason I should avoid that? >> >> Thanks in advance! >> >> -- >> Jim Thomas jomegat at jomegat.com >> Dyslexics of the world, UNTIE! >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see >> > > > -- Jim Thomas jomegat at jomegat.com Dyslexics of the world, UNTIE! From jomegat at jomegat.com Mon Aug 29 00:55:31 2011 From: jomegat at jomegat.com (Jomegat) Date: Sun, 28 Aug 2011 20:55:31 -0400 Subject: [K12OSN] Local Apps and Firefox 6 In-Reply-To: <4E5A8900.7060709@jomegat.com> References: <4E5A5ABC.9020204@jomegat.com> <4E5A8900.7060709@jomegat.com> Message-ID: <4E5AE383.7070505@jomegat.com> Thank you Robert. Your advice was absolutely necessary, but I also needed to copy /etc/resolv.conf to /opt/ltsp/i386/etc - that solved the yum problem. I also found that ltsp-server-tweaks has to be run when logged into the server at the console. Running it while remotely logged in via ssh was not enough? (Irene abated sufficiently for me to drive to the school and fool around with the network.) I now have firefox (and flash, and tuxpaint) installed as local apps, and they seem to run OK. Now all I need to do is figure out how to apply that change to all the user accounts. I don't want to log into each one and go through gnome's "add launcher" dialog. There must be a way to script this. I made a stab by copying the files in my own ~/.gnome2/panel2.d/defaults/launcher to that same directory in everyone else's acct. I also chowned it and chmodded it - but for whatever reason, the applets don't show up on the menu. Does anyone know how to change the default applets? Right now I get firefox, evolution, and gnotes in the menu panel at the top. I'd much rather have localapp-firefox, tuxpaint, and OpenOffice up there. I will do some research while I wait for an answer. I don't use gnome much myself, but rather, have been using KDE since... forever. Thanks again for the help. You guys have been golden. On 08/28/2011 02:29 PM, Jomegat wrote: > On 08/28/2011 02:10 PM, Robert Arkiletian wrote: >> On Sun, Aug 28, 2011 at 8:11 AM, Jomegat wrote: >>> Hi everyone, >>> >>> I'm planning to run Firefox as a local application and I'm looking >>> for some >>> advice. First I tried following the instructions at >>> https://fedorahosted.org/k12linux/wiki/LocalAppsSupport >> >> Not 100% sure as I can't verify this. It's been a while, but check if >> /proc is mounted inside the chroot. I think you used to have to >> >> mount /proc >> >> inside the chroot before you do yum install. > > If that helped, it didn't help enough: > > [root at termserv ~]# cd /opt/ltsp/i386/ > [root at termserv i386]# setarch i386 chroot . > bash-4.1# mount /proc > bash-4.1# yum install firefox > Error: Cannot retrieve repository metadata (repomd.xml) for repository: > epel. Please verify its path and try again > bash-4.1# > >> >>> >>> but when I got to the yum part, there was no joy: >>> >>> [root at termserv i386]# setarch i386 chroot . >>> bash-4.1# yum install firefox >>> Error: Cannot retrieve repository metadata (repomd.xml) for repository: >>> epel. Please verify its path and try again >>> >>> I thought - maybe I need to run ltsp-server-tweaks, since I do not >>> recall >>> having run that when I last rebooted (but everything seemed to work >>> then, >>> so... I dunno?) >>> >>> Here's what happens when I try running that: >>> [root at termserv ~]# ltsp-server-tweaks >>> Failed to access configuration source(s): Failed to contact >>> configuration >>> server; some possible causes are that you need to enable TCP/IP >>> networking >>> for ORBit, or you have stale NFS locks due to a system crash. See >>> http://projects.gnome.org/gconf/ for information. (Details - 1: Not >>> running >>> within active session) >>> >>> Hmmm. Not sure what all that means. >>> >>> I tried all this while remotely logged in, as Tropical Storm Irene might >>> make it a bit dodgy to get to the school (or get home afterwards). I >>> don't >>> think that's significant, but just in case it is, I'm disclosing it >>> here. >>> >>> My next thought was that maybe I should try FF 6 instead of 3.6 >>> (which is >>> what yum will install). Is there a reason I should avoid that? >>> >>> Thanks in advance! >>> >>> -- >>> Jim Thomas jomegat at jomegat.com >>> Dyslexics of the world, UNTIE! >>> >>> _______________________________________________ >>> K12OSN mailing list >>> K12OSN at redhat.com >>> https://www.redhat.com/mailman/listinfo/k12osn >>> For more info see >>> >> >> >> > > -- Jim Thomas jomegat at jomegat.com Dyslexics of the world, UNTIE! From roland at astrofoto.org Mon Aug 29 02:20:23 2011 From: roland at astrofoto.org (Roland Roberts) Date: Sun, 28 Aug 2011 22:20:23 -0400 Subject: [K12OSN] Local Apps and Firefox 6 In-Reply-To: <4E5AE383.7070505@jomegat.com> References: <4E5A5ABC.9020204@jomegat.com> <4E5A8900.7060709@jomegat.com> <4E5AE383.7070505@jomegat.com> Message-ID: <4E5AF767.3040207@astrofoto.org> On 08/28/2011 08:55 PM, Jomegat wrote: > [...] Now all I need to do is figure out how to apply that change to > all the user accounts. I don't want to log into each one and go > through gnome's "add launcher" dialog. There must be a way to script > this. I think you want to look at http://developer.gnome.org/menu-spec/. I haven't tried this myself yet.... roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From roland at astrofoto.org Mon Aug 29 04:39:31 2011 From: roland at astrofoto.org (Roland Roberts) Date: Mon, 29 Aug 2011 00:39:31 -0400 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: <4E5819D0.4030801@astrofoto.org> References: <4E5800E8.1040602@astrofoto.org> <4E5819D0.4030801@astrofoto.org> Message-ID: <4E5B1803.6020303@astrofoto.org> On 8/26/2011 6:10 PM, Roland Roberts wrote: > On 08/26/2011 05:41 PM, Robert Arkiletian wrote: [...] >> also use no_subtree_check in /etc/exports > > I did add this earlier today and looked to see what cruft I might have > in my auto.home map. I keep bouncing the daemon on another exported > filesystem so I can test. be happy with factor of 2 difference between > local and nfs, but that factor of 6 is hurting. > > I'll try a few things and report back. Well, I'm ready to throw in the towel. I'm not sure what the pattern of usage is that causes it, but... I created a 4GB file from /dev/urandom on the NFS server. Then I mount the directory on the client can do a "time cat /mnt/junk.dat > /dev/null" on the client. Then I unmount the directory from both the client and server, remount on the server and repeat on another client. If I do the "cat..." test directly on the NFS server, which is a Fedora 14 32-bit machine, it takes 54 seconds. If I do it on a Fedora 14 64-bit client, it takes 70 seconds. If I do it on the SL 6.1 64-bit client, it takes 150 seconds. Multiple tries give results within a few seconds of those each time. NFS options are set the same: [root at copernicus ~]# mount -v archos:/data/video /mnt mount: no type was given - I'll assume nfs because of the colon mount.nfs: trying text-based options 'nfsvers=4,proto=tcp,noacl,ac,bg,nocto,lock,rsize=32768,wsize=32768,bsize=32768,mountproto=tcp,addr=192.168.3.3,clientaddr=192.168.3.36' archos:/data/video on /mnt type nfs (rw,nfsvers=4,proto=tcp,noacl,ac,bg,nocto,lock,rsize=32768,wsize=32768,bsize=32768,mountproto=tcp) The hardware between the two clients is not identical, but the LAN chipset is Realtek 8168B in all three cases but...the "bad" client has a rev 02 chip and the other two have a rev 01 chip. No idea if that matters, although I have found comments about the 8168B online that I'm not sure how to take. I do know that I can't enable a 4k MTU on these or NFS stalls *very* badly. As in I log in but hang trying to run my .bashrc. I'm about ready to take the SL 6.1 motherboard and trash it. It's an MSI board, the only one I've ever had and I had some problems with it being very picky about what memory was in it, so I'm starting to wonder.... roland -- Roland Roberts, PhD http://www.astrofoto.org/ From jim.kinney at gmail.com Mon Aug 29 21:49:41 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Mon, 29 Aug 2011 17:49:41 -0400 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: <4E55A9E0.6000001@gmail.com> References: <4E55A9E0.6000001@gmail.com> Message-ID: gconftool-2 requires that the gconfd be runing. But as a non-root user, you can't edit the /etc/gconf file space. So.... boot the server but DON'T login at the gui screen. Instead, hit ctl-alt-f2 and get a normal terminal shell instead. Login as root. now run: telinit 3 This will stop the X login process now run startx And you'll (eventually) get a X screen (and a scary warning that you're logged in as root.., bad things.., fire breathing dragons, etc). _NOW_ open a terminal window and run the ltsp-server-tweaks command and it will not barf errors. log out of the X session and run telinit 5 to restart the normal X login screen alt-f7 should put you back on the correct screen for X. Login as a normal user. Open a terminal and run gconf-editor and look under desktop->gnome->lockdown and you should see that "disable_user_switching" is now checked. If you try and uncheck it, you'll get an error message that it's not writeable. Yay! On Wed, Aug 24, 2011 at 9:48 PM, Matthew Carter wrote: > Hi. I'm running freshly installed CENT-OS 6.0 on a Dell PE 2850 following > directions from the install guide. I get to the point of running > ltsp-server-tweaks, and I receive this error: > > "Failed to access configuration source(s): Failed to contact configuration > server; some possible causes are that you need to enable TCP/IP networking > for ORBit, or you have stale NFS locks due to a system crash." > > I have restarted the installation, reloaded the OS from scratch. Has anyone > seen this? > > V/r, > > Matthew > > ______________________________**_________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/**mailman/listinfo/k12osn > For more info see > -- -- James P. Kinney III As long as the general population is passive, apathetic, diverted to consumerism or hatred of the vulnerable, then the powerful can do as they please, and those who survive will be left to contemplate the outcome. - *2011 Noam Chomsky http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From burke at thealmquists.net Tue Aug 30 03:00:31 2011 From: burke at thealmquists.net (Burke Almquist) Date: Mon, 29 Aug 2011 22:00:31 -0500 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: <4E5B1803.6020303@astrofoto.org> References: <4E5800E8.1040602@astrofoto.org> <4E5819D0.4030801@astrofoto.org> <4E5B1803.6020303@astrofoto.org> Message-ID: <76D04D25-1242-4AB0-88A1-24432524DD89@thealmquists.net> On Aug 28, 2011, at 11:39 PM, Roland Roberts wrote: > > Well, I'm ready to throw in the towel. I'm not sure what the pattern of usage is that causes it, but... > > I created a 4GB file from /dev/urandom on the NFS server. Then I mount the directory on the client can do a "time cat /mnt/junk.dat > /dev/null" on the client. Then I unmount the directory from both the client and server, remount on the server and repeat on another client. > > If I do the "cat..." test directly on the NFS server, which is a Fedora 14 32-bit machine, it takes 54 seconds. If I do it on a Fedora 14 64-bit client, it takes 70 seconds. If I do it on the SL 6.1 64-bit client, it takes 150 seconds. Multiple tries give results within a few seconds of those each time. > > NFS options are set the same: > > [root at copernicus ~]# mount -v archos:/data/video /mnt > mount: no type was given - I'll assume nfs because of the colon > mount.nfs: trying text-based options 'nfsvers=4,proto=tcp,noacl,ac,bg,nocto,lock,rsize=32768,wsize=32768,bsize=32768,mountproto=tcp,addr=192.168.3.3,clientaddr=192.168.3.36' > archos:/data/video on /mnt type nfs (rw,nfsvers=4,proto=tcp,noacl,ac,bg,nocto,lock,rsize=32768,wsize=32768,bsize=32768,mountproto=tcp) > > The hardware between the two clients is not identical, but the LAN chipset is Realtek 8168B in all three cases but...the "bad" client has a rev 02 chip and the other two have a rev 01 chip. No idea if that matters, although I have found comments about the 8168B online that I'm not sure how to take. I do know that I can't enable a 4k MTU on these or NFS stalls *very* badly. As in I log in but hang trying to run my .bashrc. > > I'm about ready to take the SL 6.1 motherboard and trash it. It's an MSI board, the only one I've ever had and I had some problems with it being very picky about what memory was in it, so I'm starting to wonder.... > > roland > > > https://www.redhat.com/mailman/listinfo/k12osn > For more info see SL 6 is mostly based on Fedora 12 and 13. That may be impacting this. Have you tried using a smaller value such as 8192 for your rsize, wsize, and bsize? From roland at astrofoto.org Tue Aug 30 03:35:28 2011 From: roland at astrofoto.org (Roland Roberts) Date: Mon, 29 Aug 2011 23:35:28 -0400 Subject: [K12OSN] SL 6.1 NFS client for FC 14, performance issues? In-Reply-To: <76D04D25-1242-4AB0-88A1-24432524DD89@thealmquists.net> References: <4E5800E8.1040602@astrofoto.org> <4E5819D0.4030801@astrofoto.org> <4E5B1803.6020303@astrofoto.org> <76D04D25-1242-4AB0-88A1-24432524DD89@thealmquists.net> Message-ID: <4E5C5A80.5070706@astrofoto.org> On 08/29/2011 11:00 PM, Burke Almquist wrote: > On Aug 28, 2011, at 11:39 PM, Roland Roberts wrote: >> [...] >> NFS options are set the same: >> >> [root at copernicus ~]# mount -v archos:/data/video /mnt >> mount: no type was given - I'll assume nfs because of the colon >> mount.nfs: trying text-based options 'nfsvers=4,proto=tcp,noacl,ac,bg,nocto,lock,rsize=32768,wsize=32768,bsize=32768,mountproto=tcp,addr=192.168.3.3,clientaddr=192.168.3.36' >> archos:/data/video on /mnt type nfs (rw,nfsvers=4,proto=tcp,noacl,ac,bg,nocto,lock,rsize=32768,wsize=32768,bsize=32768,mountproto=tcp) >> >> [...] > SL 6 is mostly based on Fedora 12 and 13. That may be impacting this. > Have you tried using a smaller value such as 8192 for your rsize, wsize, and bsize? I think that's what they were before I started messing with things, but I'll go back and try them explicitly. roland -- PGP Key ID: 66 BC 3B CD Roland B. Roberts, PhD RL Enterprises roland at rlenter.com 6818 Madeline Court roland at astrofoto.org Brooklyn, NY 11220 From burke at thealmquists.net Wed Aug 31 02:37:43 2011 From: burke at thealmquists.net (Burke Almquist) Date: Tue, 30 Aug 2011 21:37:43 -0500 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: References: <4E55A9E0.6000001@gmail.com> Message-ID: On Aug 29, 2011, at 4:49 PM, Jim Kinney wrote: > gconftool-2 requires that the gconfd be runing. But as a non-root user, you can't edit the /etc/gconf file space. So.... > > boot the server but DON'T login at the gui screen. You CAN log in as root at the GUI screen, you just need to enter the username and password manually. > Instead, hit ctl-alt-f2 and get a normal terminal shell instead. Login as root. > > now run: telinit 3 > > This will stop the X login process > > now run startx > From jim.kinney at gmail.com Wed Aug 31 11:43:46 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Wed, 31 Aug 2011 07:43:46 -0400 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: References: <4E55A9E0.6000001@gmail.com> Message-ID: Hmm. I tried direct login as root and was blocked. Thus the startx process. On Aug 30, 2011 10:45 PM, "Burke Almquist" wrote: > > On Aug 29, 2011, at 4:49 PM, Jim Kinney wrote: > >> gconftool-2 requires that the gconfd be runing. But as a non-root user, you can't edit the /etc/gconf file space. So.... >> >> boot the server but DON'T login at the gui screen. > You CAN log in as root at the GUI screen, you just need to enter the username and password manually. > > > >> Instead, hit ctl-alt-f2 and get a normal terminal shell instead. Login as root. >> >> now run: telinit 3 >> >> This will stop the X login process >> >> now run startx >> > > > > > _______________________________________________ > 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 gianugo.altieri at gmail.com Wed Aug 31 11:59:26 2011 From: gianugo.altieri at gmail.com (Gianugo Altieri) Date: Wed, 31 Aug 2011 13:59:26 +0200 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: References: <4E55A9E0.6000001@gmail.com> Message-ID: <4E5E221E.5070607@gmail.com> On 08/31/2011 01:43 PM, Jim Kinney wrote: > > Hmm. I tried direct login as root and was blocked. Thus the startx > process. > > On Aug 30, 2011 10:45 PM, "Burke Almquist" > wrote: > > > > On Aug 29, 2011, at 4:49 PM, Jim Kinney wrote: > > > >> gconftool-2 requires that the gconfd be runing. But as a non-root > user, you can't edit the /etc/gconf file space. So.... > >> > >> boot the server but DON'T login at the gui screen. > > You CAN log in as root at the GUI screen, you just need to enter the > username and password manually. > > > > > > > >> Instead, hit ctl-alt-f2 and get a normal terminal shell instead. > Login as root. > >> > >> now run: telinit 3 > >> > >> This will stop the X login process > >> > >> now run startx > >> > > > > > > > > > > _______________________________________________ > > 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 When I need a graphical login as root, I edit this file: /etc/pam.d/gdm and I comment (#) this line: # auth required pam_succeed_if.so user != root quiet then I edit this file: /etc/pam.d/gdm-password and I comment (#) this line: # auth required pam_succeed_if.so user != root quiet Then I can graphically login as root both from server and from terminals Best Gianugo Altieri -------------- next part -------------- An HTML attachment was scrubbed... URL: From jim.kinney at gmail.com Wed Aug 31 12:23:05 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Wed, 31 Aug 2011 08:23:05 -0400 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: <4E5E221E.5070607@gmail.com> References: <4E55A9E0.6000001@gmail.com> <4E5E221E.5070607@gmail.com> Message-ID: Bear in mind that blocking direct root login to X and gdm was implemented because of the huge security issues exposed. X already runs with many root privileges due to how X interacts with the hardware layer. Opening pam to allow root login from terminals is flat out dangerous as the security of the password process over the network is an exposure that's not balanced by the convenience. With the exception of gconf editing now requiring an active X session to work, there is no reason for root to ever login anywhere except the actual console of the server and only at the command line. Current Linux distros basically should never have a need for direct root login unless the system is being put into single user mode for repairs. On Aug 31, 2011 8:03 AM, "Gianugo Altieri" wrote: > On 08/31/2011 01:43 PM, Jim Kinney wrote: >> >> Hmm. I tried direct login as root and was blocked. Thus the startx >> process. >> >> On Aug 30, 2011 10:45 PM, "Burke Almquist" > > wrote: >> > >> > On Aug 29, 2011, at 4:49 PM, Jim Kinney wrote: >> > >> >> gconftool-2 requires that the gconfd be runing. But as a non-root >> user, you can't edit the /etc/gconf file space. So.... >> >> >> >> boot the server but DON'T login at the gui screen. >> > You CAN log in as root at the GUI screen, you just need to enter the >> username and password manually. >> > >> > >> > >> >> Instead, hit ctl-alt-f2 and get a normal terminal shell instead. >> Login as root. >> >> >> >> now run: telinit 3 >> >> >> >> This will stop the X login process >> >> >> >> now run startx >> >> >> > >> > >> > >> > >> > _______________________________________________ >> > 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 > > When I need a graphical login as root, I edit this file: > /etc/pam.d/gdm > and I comment (#) this line: > # auth required pam_succeed_if.so user != root quiet > > then I edit this file: > /etc/pam.d/gdm-password > and I comment (#) this line: > # auth required pam_succeed_if.so user != root quiet > > Then I can graphically login as root both from server and from terminals > > Best > Gianugo Altieri -------------- next part -------------- An HTML attachment was scrubbed... URL: From burke at thealmquists.net Wed Aug 31 18:16:38 2011 From: burke at thealmquists.net (Burke Almquist) Date: Wed, 31 Aug 2011 13:16:38 -0500 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: References: <4E55A9E0.6000001@gmail.com> <4E5E221E.5070607@gmail.com> Message-ID: <8C54A7C5-FB73-432A-B363-3F984A2A33E9@thealmquists.net> On Aug 31, 2011, at 7:23 AM, Jim Kinney wrote: > Bear in mind that blocking direct root login to X and gdm was implemented because of the huge security issues exposed. X already runs with many root privileges due to how X interacts with the hardware layer. Opening pam to allow root login from terminals is flat out dangerous as the security of the password process over the network is an exposure that's not balanced by the convenience. With the exception of gconf editing now requiring an active X session to work, there is no reason for root to ever login anywhere except the actual console of the server and only at the command line. > Current Linux distros basically should never have a need for direct root login unless the system is being put into single user mode for repairs. FYI, I didn't try logging in as root on the terminals, just on the server. From jim.kinney at gmail.com Wed Aug 31 18:27:44 2011 From: jim.kinney at gmail.com (Jim Kinney) Date: Wed, 31 Aug 2011 14:27:44 -0400 Subject: [K12OSN] when running ltsp-server-tweaks In-Reply-To: <8C54A7C5-FB73-432A-B363-3F984A2A33E9@thealmquists.net> References: <4E55A9E0.6000001@gmail.com> <4E5E221E.5070607@gmail.com> <8C54A7C5-FB73-432A-B363-3F984A2A33E9@thealmquists.net> Message-ID: On Wed, Aug 31, 2011 at 2:16 PM, Burke Almquist wrote: > > On Aug 31, 2011, at 7:23 AM, Jim Kinney wrote: > > > Bear in mind that blocking direct root login to X and gdm was implemented > because of the huge security issues exposed. X already runs with many root > privileges due to how X interacts with the hardware layer. Opening pam to > allow root login from terminals is flat out dangerous as the security of the > password process over the network is an exposure that's not balanced by the > convenience. With the exception of gconf editing now requiring an active X > session to work, there is no reason for root to ever login anywhere except > the actual console of the server and only at the command line. > > Current Linux distros basically should never have a need for direct root > login unless the system is being put into single user mode for repairs. > FYI, I didn't try logging in as root on the terminals, just on the server. > Good to hear as that cuts down on some of the security issues. However, as all (most) thin client processes run on the server itself, and X has many data leaks, logging as root in X on the server is still a security nightmare asking for exploitation. And with a flock of curious kids looking for "fun things to do", a political problem waiting to happen :-) Besides, EVERYTHING needed for administration of a Linux server can be done with su, sudo or the gui's with root password. OK. So single user mode to fix a disk that has failed and a few repartitioning things requires root login in runlevel 1. But other than that, and the gconf thing, but, really, nothing else .... :-) Be wary of middle and high school students. They be crafty little imps! Fork-bombs are a fun way to bring class to a halt! > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- -- James P. Kinney III As long as the general population is passive, apathetic, diverted to consumerism or hatred of the vulnerable, then the powerful can do as they please, and those who survive will be left to contemplate the outcome. - *2011 Noam Chomsky http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: