IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 3 November 2011 (all times are UTC)
| 00:05 | mordocai has joined IRC (mordocai!~mordocai@mail.mordocai.net) | |
| 00:13 | alkisg has left IRC (alkisg!alkisg@ubuntu/member/alkisg, Ping timeout: 240 seconds) | |
| 00:29 | lifeboy has joined IRC (lifeboy!~lifeboy@196-210-157-244.dynamic.isadsl.co.za) | |
| 00:37 | vagrantc has left IRC (vagrantc!~vagrant@75-150-46-245-Oregon.hfc.comcastbusiness.net, Ping timeout: 244 seconds) | |
| 00:56 | cliebow_ has joined IRC (cliebow_!~cliebow@66.63.66.218) | |
| 01:02 | hughessd has joined IRC (hughessd!~hughessd@71-20-34-17.war.clearwire-wmx.net) | |
| 01:04 | hughessd has left IRC (hughessd!~hughessd@71-20-34-17.war.clearwire-wmx.net, Client Quit) | |
| 01:09 | mordocai has left IRC (mordocai!~mordocai@mail.mordocai.net, Quit: Leaving) | |
| 01:24 | loather has joined IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net) | |
| 01:45 | jammcq has joined IRC (jammcq!~jam@70-91-230-209-BusName-Michigan.hfc.comcastbusiness.net) | |
| 01:46 | <cliebow_> hiho jammcq
| |
| 01:52 | <jammcq> hey cliebow_
| |
| 01:53 | <lifeboy> Anyone awake here?
| |
| 01:53 | <cliebow_> nice to see you here..not hthe same without you
| |
| 01:53 | been real quiet..Gadi still looks live
| |
| 01:53 | <jammcq> yeah, I try to pop in when I can
| |
| 01:54 | also I was checking if sbalneav was in yet. I dropped him at the airport around 2pm
| |
| 01:54 | he should be arriving home about now
| |
| 01:54 | <cliebow_> me too..ohhh airport
| |
| 01:55 | <lifeboy> Booting LTSP with PXE using NFS, I get an error:
| |
| 01:55 | /scripts/nfs-bottom/ltsp: line 47: can't create /root/etc/hostname: Read-only file system
| |
| 01:55 | <cliebow_> im still trying to get a handle on the changes coming..seems tied to sshpam auth
| |
| 01:55 | <lifeboy> and then
| |
| 01:55 | init: hostname main process (1039) terminated with status 1
| |
| 01:56 | Anyone know what's causing this?
| |
| 01:56 | Gadi: you still around?
| |
| 01:56 | <jammcq> lifeboy: sorry, it's way beyond my current knowledge of ltsp
| |
| 01:56 | <cliebow_> lifeboy, pretty late here
| |
| 01:56 | <lifeboy> Here too, 4am...
| |
| 01:57 | No rest for the wicked... :D
| |
| 01:57 | <cliebow_> wow..worked all night ..again
| |
| 02:00 | i cant pretend to know how to help..i can only ask at what stage you see this error..
| |
| 02:04 | <lifeboy> It's in line 47 of /usr/share/initramfs-tools/scripts/nfs-bottom/ltsp which is:
| |
| 02:04 | cat /proc/sys/kernel/hostname > ${rootmnt}/etc/hostname || true
| |
| 02:05 | I think {rootmnt} is readonly, but I can't find anything via google that says what I should do
| |
| 02:05 | to fix that
| |
| 02:05 | I don't even know what {rootmnt}'s value is once the client has booted
| |
| 02:06 | But I do know where the coffee is in my kitchen. I'll be back...
| |
| 02:06 | <cliebow_> so cant write the hostname..must be root..like before a switchroot?
| |
| 02:06 | <lifeboy> It says before that in the console: Begin: Running scripts/nfs-bottom ...
| |
| 02:07 | and a few lines up: Mounting root file system ...
| |
| 02:08 | <cliebow_> i cant tell whether you are in the initramfs and the nbd filesystem hasnt mounted...
| |
| 02:08 | <lifeboy> The fact that it says in the error: "can't create /root/etc/hostname" means it attempted to mount /root, but it's readonly and shouldn't be
| |
| 02:09 | I'm not using NBD... long story
| |
| 02:10 | <cliebow_> sorry.. i am blowing smoke from memory..oh..nfs
| |
| 02:10 | sure nfs is set up rw?
| |
| 02:10 | <lifeboy> that's why nfs-bottom is being run
| |
| 02:10 | <cliebow_> God it has been so long
| |
| 02:11 | Well the serve has to be set up rw for nfs
| |
| 02:11 | <lifeboy> I followed the instructions at https://help.ubuntu.com/community/UbuntuLTSP/LTSPWithoutNFS
| |
| 02:12 | and that says I should have /etc/export
| |
| 02:12 | /opt/ltsp/i386 *(ro,no_subtree_check,no_root_squash)
| |
| 02:12 | Let me change that to rw
| |
| 02:13 | <cliebow_> this looks ancient...primrdial even
| |
| 02:14 | <lifeboy> You're not that old... ;-)
| |
| 02:14 | I restarted nfs-kernel-server after making it rw, but it still give the same error
| |
| 02:14 | <cliebow_> heh..the document..
| |
| 02:16 | i can look at my Sanscrit configs at school tomorrow..nfs can be an odd duck..stale mounts and stuff
| |
| 02:19 | i guess originally /opt/ltsp/i386 was set to ro..so everyone wouldnt write all over the servers filesystem..
| |
| 02:19 | i am shutting up..s my memory is as stale as the doc you are working from
| |
| 02:20 | <lifeboy> I'm checking to see if I can find something for 10.04 specifically
| |
| 02:23 | <cliebow_> get some sleep!! see ya
| |
| 02:26 | <lifeboy> cheers
| |
| 02:39 | cliebow_ has left IRC (cliebow_!~cliebow@66.63.66.218, Quit: Leaving) | |
| 02:39 | Parker955_Away is now known as Parker955 | |
| 03:16 | Parker955 is now known as Parker955_Away | |
| 03:32 | F-GT has left IRC (F-GT!~phantom@ppp121-44-71-230.lns20.syd6.internode.on.net, Ping timeout: 276 seconds) | |
| 03:43 | miks has joined IRC (miks!~chatzilla@50-39-205-135.bvtn.or.frontiernet.net) | |
| 03:46 | <miks> Hey everyone, I could use some help getting thin clients to shut down on Edubuntu 11.10
| |
| 03:47 | I'd like to get automated shutdown working, AND give them a button the desktop to shutdown, if possible. I'd settle for EITHER at the moment
| |
| 03:47 | I set CRON up in the chroot and tried to do it that way, but nothing happened at the appointed time.
| |
| 03:49 | I then used the sample ltsp-shutdown.desktop file recommended the other day, but when I run it nothing happens. I tried it from a terminal, and got the result "command not found"
| |
| 03:50 | Can anyone point me in the right direction, or tell me how to troubleshoot?
| |
| 03:51 | Frustratingly enough, I don't even have a "Shut down" option in the menu on the thin client. Only "logout"
| |
| 04:00 | lifeboy has left IRC (lifeboy!~lifeboy@196-210-157-244.dynamic.isadsl.co.za) | |
| 04:10 | F-GT has joined IRC (F-GT!~phantom@ppp121-44-137-22.lns20.syd7.internode.on.net) | |
| 04:18 | miks has left IRC (miks!~chatzilla@50-39-205-135.bvtn.or.frontiernet.net, Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238]) | |
| 04:45 | staffencasa has left IRC (staffencasa!~staffenca@128-193-147-224.oregonstate.edu, Ping timeout: 252 seconds) | |
| 05:20 | joshson has joined IRC (joshson!~bakytn@airport.kg) | |
| 05:20 | joshson is now known as bakytn | |
| 05:21 | <bakytn> Hey! Currently I have couple of LTSP clients running through rdesktop. If I run ltsp-update-image. Would I break their work?
| |
| 05:24 | btw, I am also getting: "kernel: FAT: bogus number of reserved sectors" when ThinClient is booting up. This never happened before..
| |
| 06:57 | Gadi has left IRC (Gadi!~romm@ool-4571ca04.dyn.optonline.net, Read error: Connection reset by peer) | |
| 07:14 | Gadi has joined IRC (Gadi!~romm@ool-4571ca04.dyn.optonline.net) | |
| 07:57 | loather has left IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net, Quit: This computer has gone to sleep) | |
| 07:57 | loather has joined IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net) | |
| 08:00 | loather has left IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net, Read error: Operation timed out) | |
| 08:15 | <bakytn> Hello! Is there any way to provide use USB Flash Printers as a local ones via RDP. Unfortunately only a few printers support JetDirect
| |
| 08:19 | dobber has joined IRC (dobber!~dobber@213.169.45.222) | |
| 08:49 | vmlintu has left IRC (vmlintu!~vmlintu@nblzone-240-143.nblnetworks.fi, Ping timeout: 252 seconds) | |
| 09:21 | lifeboy has joined IRC (lifeboy!~lifeboy@196-210-157-244.dynamic.isadsl.co.za) | |
| 09:21 | <lifeboy> Hi all!
| |
| 09:21 | Steve_the_Pirate has joined IRC (Steve_the_Pirate!~Gary@host86-166-104-72.range86-166.btcentralplus.com) | |
| 09:23 | <lifeboy> Is there anyone here that knows how to use NFS for a thin client rather than NBD that can help me troubleshoot a readonly root mount issue?
| |
| 09:25 | The US is still sleeping...
| |
| 09:50 | Steve_the_Pirate has left IRC (Steve_the_Pirate!~Gary@host86-166-104-72.range86-166.btcentralplus.com, Ping timeout: 260 seconds) | |
| 09:58 | * bakytn I wonder if it should be read only normally | |
| 10:04 | Gremble has joined IRC (Gremble!~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) | |
| 10:13 | cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg) | |
| 10:15 | bakytn has left IRC (bakytn!~bakytn@airport.kg, Quit: Leaving) | |
| 10:17 | vmlintu has joined IRC (vmlintu!~vmlintu@dsl-hkibrasgw2-ff78c300-68.dhcp.inet.fi) | |
| 10:24 | toscalix has joined IRC (toscalix!~toscalix@31.4.43.72) | |
| 10:25 | alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg) | |
| 10:30 | <lifeboy> lifeboy 03/11/2011 11:23:04
| |
| 10:30 | Is there anyone here that knows how to use NFS for a thin client rather than NBD that can help me troubleshoot a readonly root mount issue?
| |
| 10:31 | <alkisg> https://help.ubuntu.com/community/UbuntuLTSP/LTSPWithoutNFS
| |
| 10:40 | <lifeboy> alkisg: If it was that simple, I would had this sorted out last week already!
| |
| 10:41 | <alkisg> I'm not sure I understand, do you mean it's difficult to switch from nbd to nfs?
| |
| 10:42 | <lifeboy> No, I've got NFS all in place (as per that how to), but the root file system get's mounted as read only, preventing the client startup process from continuing
| |
| 10:43 | <alkisg> lifeboy: nbd is supposed to be mounted read-only
| |
| 10:43 | Then aufs + tmpfs are used to get a writeable root
| |
| 10:43 | As Gadi said yesterday, you're probably not using the ltsp_nbd script?
| |
| 10:44 | <lifeboy> I can confidently say I have explored all the commonly documented options :-(
| |
| 10:45 | <alkisg> Erm, can you paste your pxelinux.cfg/default ?
| |
| 10:45 | [GuS] has left IRC ([GuS]!~gustavo@186.19.173.126, Remote host closed the connection) | |
| 10:46 | <lifeboy> default ltsp
| |
| 10:46 | label ltsp
| |
| 10:46 | kernel vmlinuz
| |
| 10:46 | i# append ro initrd=initrd.img nbdsrv=172.16.0.1 nbdport=2000 nbdroot=/dev/nbd
| |
| 10:46 | append ro initrd=initrd.img boot=nfs nfsroot=/opt/ltsp/i386
| |
| 10:47 | sorry, the 4th line is commented out. (not starting with an i)
| |
| 10:47 | <alkisg> lifeboy: ah ok I thought you were still trying NBD
| |
| 10:48 | I'm not very familiar with the required nfs command line
| |
| 10:48 | <lifeboy> The problem with NBD is this:
| |
| 10:48 | I can't compile the r6040 module under Ubuntu 10.04
| |
| 10:49 | If I use the vortex86 kernel, I have r6040 already in there, but NBD seems to be the upstream version, not the Ubuntu one
| |
| 10:49 | And I can't figure out how to setup that NBD
| |
| 10:49 | <alkisg> You misunderstood
| |
| 10:49 | "but NBD seems to be the upstream version, not the Ubuntu one"
| |
| 10:49 | we don't have a special nbd module in ltsp
| |
| 10:49 | We're using the upstream one
| |
| 10:50 | But
| |
| 10:50 | <lifeboy> Ah, that explains why the sources are the same...
| |
| 10:50 | <alkisg> We do have an initramfs script to load the module
| |
| 10:50 | That initramfs script is called ltsp_nbd
| |
| 10:50 | What Gadi was telling you yesterday is that instead of using the ltsp_nbd script on pxelinux.cfg/default, you're using the upstream "nbd" script
| |
| 10:51 | That's just a small configuration change
| |
| 10:51 | <lifeboy> I have ltsp_nbd. So should I copy that into the vortex client?
| |
| 10:51 | <alkisg> Let me say some theory again
| |
| 10:52 | <lifeboy> Please
| |
| 10:52 | <alkisg> ltsp_nbd is installed with ltsp-client in the chroot
| |
| 10:52 | So "I have ltsp_nbd. So should I copy that into the vortex client?" => doesn't make sense
| |
| 10:52 | It should already be there in the vortex client
| |
| 10:52 | Unless you used some weird method to generate the chroot
| |
| 10:53 | Now, there are some scripts called "initramfs hooks"
| |
| 10:53 | Those hooks run when you run `update-initramfs -u` inside the chroot
| |
| 10:53 | And these hooks make sure that the nbd module and the ltsp_nbd script are copied inside the initramfs
| |
| 10:53 | So
| |
| 10:53 | Even if you're using a custom kernel, your initramfs should automatically be ltsp-ready
| |
| 10:54 | That means that you're using some weird method somewhere, preventing ltsp from copying the necessary stuff to the initramfs
| |
| 10:54 | So, what exactly are you doing, so that we can pinpoint the problem in your method?
| |
| 10:54 | <lifeboy> the only way I know to get the vortex chroot is to make a directory and copy the extracted archive into it. Then I check everything manually.
| |
| 10:55 | <alkisg> The shouldn't even be a "vortex chroot" concept
| |
| 10:55 | vortex kernel, ok
| |
| 10:55 | But vorted chroot, or vorted initrd, no
| |
| 10:56 | <lifeboy> So what do I call the client setup in /opt/ltsp/i386 ?
| |
| 10:56 | <alkisg> So, start from "I have an ltsp chroot at /opt/ltsp/i386, and a vorted kernel at /somedir/vmlinuz"
| |
| 10:56 | And mention the steps you take to put the vortex kernel in the ltsp chroot
| |
| 10:56 | <lifeboy> Ok, let me go through it all step by step.
| |
| 10:57 | <alkisg> Before you start
| |
| 10:58 | You're actually using a chroot generated by ltsp-build-client, right?
| |
| 10:59 | lifeboy has left IRC (lifeboy!~lifeboy@196-210-157-244.dynamic.isadsl.co.za, Remote host closed the connection) | |
| 11:06 | alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Ping timeout: 260 seconds) | |
| 11:06 | Trixboxer has joined IRC (Trixboxer!~Trixboxer@office.supportdepartment.net) | |
| 11:22 | lifeboy_ has joined IRC (lifeboy_!~roland@196-210-157-244.dynamic.isadsl.co.za) | |
| 11:23 | lifeboy_ has left IRC (lifeboy_!~roland@196-210-157-244.dynamic.isadsl.co.za, Client Quit) | |
| 11:25 | lifeboy has joined IRC (lifeboy!~lifeboy@196-210-157-244.dynamic.isadsl.co.za) | |
| 11:26 | lifeboy has left IRC (lifeboy!~lifeboy@196-210-157-244.dynamic.isadsl.co.za) | |
| 11:27 | lifeboy has joined IRC (lifeboy!~roland@196-210-157-244.dynamic.isadsl.co.za) | |
| 11:28 | FrozenZia has left IRC (FrozenZia!pbrown@shell.paivola.fi, Quit: Changing server) | |
| 11:30 | <lifeboy> alkisg: I'm building a new chroot with that yes.
| |
| 11:33 | Steve_the_Pirate has joined IRC (Steve_the_Pirate!~Gary@host86-166-104-72.range86-166.btcentralplus.com) | |
| 11:36 | Da-Geek has joined IRC (Da-Geek!~Da-Geek@80.235.229.178) | |
| 11:40 | alkisg has joined IRC (alkisg!alkisg@ubuntu/member/alkisg) | |
| 11:56 | pbrown has joined IRC (pbrown!pbrown@shell.paivola.fi) | |
| 11:56 | pbrown is now known as FrozenZia | |
| 11:59 | jammcq has left IRC (jammcq!~jam@70-91-230-209-BusName-Michigan.hfc.comcastbusiness.net, Quit: leaving) | |
| 12:00 | <lifeboy> alkisg: Once the ltsp-build-client is complete, I add the some apt configs to make sources available and enable apt-cacher proxy to the chroot
| |
| 12:00 | <alkisg> lifeboy: no need
| |
| 12:00 | You've already compiled your module, right?
| |
| 12:01 | Just put your kernel + the module in the stock ltsp chroot
| |
| 12:01 | And run update-initramfs -u in the chroot and ltsp-update-kernels/ltsp-update-image outside it
| |
| 12:01 | <lifeboy> alkisg: No, I haven't, but I don't need to, since the vortex sources have it.
| |
| 12:01 | <alkisg> lifeboy: I mean, you don't need to compile again since you compiled the previous time
| |
| 12:02 | So no need to go though kernel compilation again
| |
| 12:02 | Just copy the result that you have from your last try
| |
| 12:02 | <lifeboy> Yes, I understand
| |
| 12:02 | <alkisg> But anyway if you feel better recompiling, np, do so
| |
| 12:03 | Also you don't need apt-cacher
| |
| 12:03 | <lifeboy> I'll first try the previously build kernel
| |
| 12:03 | <alkisg> There's a "mount package cache" option that reuses the package cache from your server
| |
| 12:03 | Both in ltsp-build-client and in ltsp-chroot
| |
| 12:03 | <lifeboy> I know I don't, but I don't want to waster bandwidth when I update the chroot now or later
| |
| 12:03 | waster = waste
| |
| 12:04 | <alkisg> Yes, I'm telling you of an easier method to not waste bandwidth
| |
| 12:04 | <lifeboy> But if the server is 10.10 64bit, if won't help, will it?
| |
| 12:05 | <alkisg> It will do the same thing apt-cacher does, but without using apt-cacher
| |
| 12:05 | OK anyway details, go on...
| |
| 12:05 | * alkisg also goes back to programming epoptes... | |
| 12:06 | <lifeboy> Ok, ltsp-build-client is complete
| |
| 12:27 | now I copied: linux-image-2.6.34.1-vortex86-sg_1.0_i386 to /opt/ltsp/i386/usr/src
| |
| 12:27 | then ltsp-chroot --arch=i386
| |
| 12:28 | and mount -t proc proc /proc
| |
| 12:28 | <pscheie> good morning all
| |
| 12:28 | <lifeboy> apt-get install gcc
| |
| 12:29 | I ave recompile the vortex kernel... I cleaned up the old stuff last night :-(
| |
| 12:30 | typo... I have to recompile...
| |
| 12:33 | copied .config for vortex kernel to /usr/src/linux-image-2.6.34.1-vortex86-sg_1.0_i386
| |
| 12:34 | <alkisg> Good morning pscheie
| |
| 12:34 | <pscheie> good morning alkisg. Are you still in Orlando?
| |
| 12:34 | <alkisg> pscheie: so are those lsb/init-functions available in some package?
| |
| 12:34 | Yup
| |
| 12:35 | If so, I can always use them, and just have the centos/fedora package depend on that lsb package...
| |
| 12:36 | <pscheie> There is the redhat-lsb package which probably will do what you want, and you could just make that a dependency on Centos.
| |
| 12:36 | Let me install it just to make sure it's got what you need.
| |
| 12:37 | <alkisg> script = /lib/lsb/init-functions, functions = log_begin_msg, log_warning_msg, log_end_msg
| |
| 12:38 | <lifeboy> linux-image-2.6.34.1-vortex86-sg_1.0_i386 was the wrong one, the correct one is linux-2.6.34.1-vortex86-sg
| |
| 12:40 | <pscheie> /lib/lsb-init-functions has a log_warning.msg, but no log_begin_msg nor log_end_msg :-(
| |
| 12:40 | It says it's LSB spec 1.1.0, fwiw.
| |
| 12:40 | <alkisg> :(
| |
| 12:41 | <pscheie> http://pastebin.ubuntu.com/727217/
| |
| 12:41 | It's a small file.
| |
| 12:42 | <lifeboy> apt-get install fakeroot kernel-wedge build-essential makedumpfile kernel-package libncurses5 libncurses5-dev
| |
| 12:43 | <alkisg> pscheie: so I'd better just use the alias hack we used at the BTS :)
| |
| 12:43 | <pscheie> Yes, I think you'll have to roll your own.
| |
| 12:44 | <lifeboy> make ARCH=i386 menuconfi
| |
| 12:44 | make ARCH=i386 menuconfig
| |
| 12:45 | and add -nbd to the localversion and enable NBD support
| |
| 12:45 | brunolambert has joined IRC (brunolambert!~bruno@nat/revolutionlinux/x-fxngeoqyrjyrprhv) | |
| 12:46 | <lifeboy> make-kpkg clean
| |
| 12:49 | time fakeroot make-kpkg --initrd kernel-image kernel-headers
| |
| 12:50 | This will give me kernel with NBD and r6040 compiled in without requiring modules
| |
| 12:51 | toscalix has left IRC (toscalix!~toscalix@31.4.43.72, Read error: Connection reset by peer) | |
| 12:58 | Parker955_Away is now known as Parker955 | |
| 13:03 | <lifeboy> Ok, compile is done.
| |
| 13:04 | Parker955 is now known as Parker955_Away | |
| 13:05 | <lifeboy> alkisg: I now have a linux-image-....deb after the compile. Do Install that in the chroot, or do I extract that somewhere and only copy the kernel to /boot?
| |
| 13:05 | I meant "Do I install..."
| |
| 13:06 | <alkisg> Install it
| |
| 13:07 | <lifeboy> Ok.
| |
| 13:09 | alkisg: There an image and a header .deb. I only install the image, right?
| |
| 13:09 | dpkg -i linux-image-2.6.34.1-vortex86-sg-nbd_2.6.34.1-vortex86-sg-nbd-10.00.Custom_i386.deb
| |
| 13:09 | <alkisg> Right, unless you also want to compile something else in the chroot, like e.g. the nvidia proprietary drivers
| |
| 13:10 | <lifeboy> ok, done.
| |
| 13:12 | Gremble has left IRC (Gremble!~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com, Quit: I Leave) | |
| 13:15 | Parker955_Away is now known as Parker955 | |
| 13:15 | <lifeboy> alkisg: what's the command again that links vmlinuz et al to the correct source file? It currently links to the other kernel
| |
| 13:15 | <alkisg> lifeboy: there's no need to do that
| |
| 13:15 | <lifeboy> Not?
| |
| 13:16 | <alkisg> You mean to link $CHROOT/vmlinuz, right?
| |
| 13:16 | <lifeboy> yes
| |
| 13:16 | <alkisg> No, that's not used. Whatever you put to pxelinux.cfg/default is used
| |
| 13:16 | Also, if the kernel you installed is newer than the one you had, that symlink should automatically be generated
| |
| 13:16 | Unless vortex has a broken package
| |
| 13:18 | <lifeboy> It still has vmlinuz -> vmlinuz-2.6.32-34-generic
| |
| 13:18 | not vmlinuz-2.6.34.1-vortex86-sg-nbd
| |
| 13:18 | <alkisg> OK. Anyway ignore that
| |
| 13:18 | <lifeboy> Ok
| |
| 13:19 | Is there anything else I need to do before I edit pxelinux.cfg/default?
| |
| 13:21 | <alkisg> Put the output of the installation to pastebin
| |
| 13:21 | dpkg -i linux-imageetc
| |
| 13:21 | To see if the triggers were executed, the initramfs updated etc
| |
| 13:23 | Parker955 is now known as Parker955_Away | |
| 13:29 | <lifeboy> update-initramfs -c -k 2.6.34.1-vortex86-sg-nbd
| |
| 13:30 | <alkisg> I didn't say to run a command
| |
| 13:30 | I just asked to paste the output of the commands you already ran
| |
| 13:30 | You shouldn't have to run update-initramfs manually, it should happen automatically
| |
| 13:33 | <lifeboy> one sec
| |
| 13:33 | ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood) | |
| 13:34 | ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de) | |
| 13:39 | bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at) | |
| 13:43 | dead_inside has joined IRC (dead_inside!~dead_insi@76.75.3.174) | |
| 13:43 | <lifeboy> alkisg: sorry, phonecall...
| |
| 13:43 | <alkisg> np I'm not watching the channel anyway, still programming... :)
| |
| 13:51 | Gremble has joined IRC (Gremble!~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com) | |
| 13:51 | <lifeboy> alkisg: http://pastebin.com/qKmaxumS
| |
| 13:51 | alkisg: Sorry about that, I didn't realise that was meant for me
| |
| 13:53 | alkisg: btw, I'm really looking forward to trying out epoptes in a business environment for remote support to ltsp users desktops
| |
| 13:54 | <alkisg> An edubuntu session starts in 5 minutes, bbl..
| |
| 14:19 | ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Ping timeout: 256 seconds) | |
| 14:22 | andygraybeal has left IRC (andygraybeal!~andy.gray@obsidian.casanueva.com, Quit: Ex-Chat) | |
| 14:22 | ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de) | |
| 14:37 | mgariepy has left IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy, Quit: Leaving) | |
| 14:51 | mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy) | |
| 14:56 | mgariepy has left IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy, Remote host closed the connection) | |
| 15:00 | mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy) | |
| 15:12 | Gremble has left IRC (Gremble!~Ben@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com, Quit: I Leave) | |
| 15:19 | <lifeboy> alkisg: When you get back, I have put everything I did after compiling into here: http://pastebin.com/FxUzQbPY
| |
| 15:19 | staffencasa has joined IRC (staffencasa!~staffenca@128-193-149-148.oregonstate.edu) | |
| 15:21 | loather has joined IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net) | |
| 15:21 | <alkisg> lifeboy: and what happens now if you try to boot the client?
| |
| 15:27 | mgariepy has left IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy, Remote host closed the connection) | |
| 15:28 | <lifeboy> I get the error that I had yesterday: "No Init found. Try passing init= bootarg"
| |
| 15:32 | mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy) | |
| 15:37 | loather has left IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net, Quit: Leaving) | |
| 15:40 | Da-Geek has left IRC (Da-Geek!~Da-Geek@80.235.229.178, Quit: Leaving) | |
| 15:41 | <alkisg> lifeboy: put your /var/lib/tftpboot/ltsp/i386/pxelinux.cfg/default to pastebin
| |
| 15:49 | cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 240 seconds) | |
| 15:51 | cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg) | |
| 15:55 | vmlintu has left IRC (vmlintu!~vmlintu@dsl-hkibrasgw2-ff78c300-68.dhcp.inet.fi, Ping timeout: 258 seconds) | |
| 15:56 | loather has joined IRC (loather!~khudson@wsip-98-175-250-115.sd.sd.cox.net) | |
| 16:07 | <lifeboy> I'm back. Here it is http://pastebin.com/eWzWnPth
| |
| 16:08 | alkisg: I edited that to reflect the actual filenams to be used
| |
| 16:08 | ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood) | |
| 16:08 | pscheie has left IRC (pscheie!~pscheie@74-95-141-141-Minnesota.hfc.comcastbusiness.net, Ping timeout: 252 seconds) | |
| 16:09 | pscheie has joined IRC (pscheie!~pscheie@74-95-141-141-Minnesota.hfc.comcastbusiness.net) | |
| 16:09 | <alkisg> lifeboy: it seems that the chroot kernel hooks didn't execute, because the vortex package is semi-broken, you need to call them manually
| |
| 16:09 | Those hooks update the initramfs, put ltsp_nbd in it etc
| |
| 16:09 | Update the vmlinuz symlink...
| |
| 16:10 | cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 240 seconds) | |
| 16:10 | <alkisg> /etc/kernel/postinst.d/ltsp-update-kernels
| |
| 16:10 | ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de) | |
| 16:12 | cyberorg has joined IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg) | |
| 16:13 | <lifeboy> alkisg: This file is empty
| |
| 16:13 | <alkisg> lifeboy: you mean that you don't have that script in your chroot?
| |
| 16:13 | mnemoc has left IRC (mnemoc!~amery@shell.opensde.net, Quit: Lost terminal) | |
| 16:13 | <alkisg> It's part of ltsp-client-core, if you don't have it then you don't have ltsp
| |
| 16:13 | <lifeboy> in the chroot (/opt/ltsp/i386) right?
| |
| 16:13 | <alkisg> Yes
| |
| 16:14 | What's the output of this command?
| |
| 16:14 | sudo chroot /opt/ltsp/i386 dpkg -l ltsp-client
| |
| 16:14 | <lifeboy> The file is there, but it contains nothing. Unless the dpkg -i overwrote it?
| |
| 16:16 | <alkisg> If you don't have ltsp-client installed, *don't install it now*.
| |
| 16:16 | Do you?
| |
| 16:16 | <lifeboy> dpkg -l ltsp-client
| |
| 16:16 | Desired=Unknown/Install/Remove/Purge/Hold
| |
| 16:16 | | Status=Not/Inst/Cfg-files/Unpacked/Failed-cfg/Half-inst/trig-aWait/Trig-pend
| |
| 16:16 | |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
| |
| 16:16 | ||/ Name Version Description
| |
| 16:16 | +++-==========================================-==========================================-====================================================================================================
| |
| 16:16 | ii ltsp-client 5.2.1-0ubuntu9 LTSP client environment
| |
| 16:16 | huh? Surely lstp-build-client should have done that?
| |
| 16:19 | <alkisg> No, no package is supposed to overwrite scripts with empty files
| |
| 16:19 | cyberorg has left IRC (cyberorg!~cyberorg@opensuse/member/Cyberorg, Ping timeout: 248 seconds) | |
| 16:19 | <alkisg> I can only assume that the vortex kernel package is so horribly broken that it deletes other packages's scripts in case they won't work with that kernel...
| |
| 16:20 | Try restoring the missing script + running it
| |
| 16:21 | <lifeboy> Just checked the image.deb and it does contain the /etc/kernel/* directories, but not files in there, so this is not the case it seems to me
| |
| 16:21 | <alkisg> lifeboy: ah, maybe you didn't follow the symlink correctly
| |
| 16:21 | Did you try to check that file from within the chroot, or from outside the chroot?
| |
| 16:21 | $ ls -lha /opt/ltsp/i386/etc/kernel/postinst.d/ltsp-update-kernels
| |
| 16:21 | lrwxrwxrwx 1 root root 30 2011-09-01 08:48 /opt/ltsp/i386/etc/kernel/postinst.d/ltsp-update-kernels -> /usr/share/ltsp/update-kernels
| |
| 16:22 | It's a symlink, so you're supposed to run it from within the chroot, not open it from outside...
| |
| 16:23 | <lifeboy> Ah, yes, I see the link is to /usr/share/ltsp/update-kernels, which does exist
| |
| 16:23 | in the chroot
| |
| 16:24 | <alkisg> lifeboy: and maybe the main problem was that when you installed the kernel, you didn't have /proc mounted
| |
| 16:24 | Did you?
| |
| 16:25 | <lifeboy> No it did have it mounted. Let me check my history
| |
| 16:25 | <alkisg> So anyway chroot, mount /proc, and run that postinst kernels hook
| |
| 16:28 | <lifeboy> Here's my history http://pastebin.com/Qy8HxFSw
| |
| 16:29 | Ok
| |
| 16:30 | run /etc/kernel/postinst.d/ltsp-update-kernels?
| |
| 16:31 | <alkisg> Yes, in the chroot, with /proc mounted
| |
| 16:31 | And put the output to pastebin
| |
| 16:31 | And check if it fixed the /boot/vmlinuz symlink
| |
| 16:32 | <lifeboy> /usr/share/ltsp/update-kernels run, the previous one (the symlink) does nothing
| |
| 16:32 | <alkisg> ....are you sure you're inside the chroot?
| |
| 16:32 | Both of them should be the same file
| |
| 16:32 | <lifeboy> Yes. Only three lines
| |
| 16:32 | Skipping yaboot configuration. install yaboot package if you need it.
| |
| 16:32 | Skipping netabootwrap images. Install the aboot package if you need them.
| |
| 16:32 | Skipping sparc piggyback images. Install the sparc-utils package if you need them.
| |
| 16:32 | dobber has left IRC (dobber!~dobber@213.169.45.222, Remote host closed the connection) | |
| 16:33 | hughessd has joined IRC (hughessd!~hughessd@173-164-117-109-Oregon.hfc.comcastbusiness.net) | |
| 16:34 | <alkisg> (06:31:48 μμ) alkisg: And check if it fixed the /boot/vmlinuz symlink
| |
| 16:34 | <lifeboy> Yes, the links are correct and the the pxelinux.cfg/default is correct as well
| |
| 16:35 | <alkisg> OK, now within the chroot again, run: update-initramfs -u
| |
| 16:35 | <lifeboy> update-initramfs: Generating /boot/initrd.img-2.6.34.1-vortex86-sg-nbd
| |
| 16:36 | <alkisg> After that unmount proc and exit the chroot
| |
| 16:36 | <lifeboy> ok
| |
| 16:36 | ltsp-update-image and first and then ltsp-update-kernels?
| |
| 16:37 | <alkisg> Not yet
| |
| 16:37 | Do these:
| |
| 16:37 | (outside the chroot)
| |
| 16:37 | <lifeboy> ok
| |
| 16:37 | <alkisg> mkdir /tmp/init-vortex
| |
| 16:37 | cd /tmp/init-vortex/
| |
| 16:37 | gunzip -c /opt/ltsp/i386/boot/initrd | cpio -i -d -H newc --no-absolute-filenames
| |
| 16:37 | ls -lha -R
| |
| 16:38 | And put the result to pastebin
| |
| 16:38 | With that, we'll check that all the files got indeed inside your initramfs
| |
| 16:38 | Sorry I forgot an .img extension
| |
| 16:38 | gunzip -c /opt/ltsp/i386/boot/initrd.img | cpio -i -d -H newc --no-absolute-filenames
| |
| 16:41 | <lifeboy> http://pastebin.com/K5H93dtY
| |
| 16:41 | mgariepy has left IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy, Remote host closed the connection) | |
| 16:45 | <alkisg> lifeboy: I don't see nbd.ko in it, is it compiled into the kernel?
| |
| 16:47 | But anyway, continue with the update-image + kernel
| |
| 16:48 | mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy) | |
| 16:49 | jvin has joined IRC (jvin!~jvin@108-82-19-151.lightspeed.livnmi.sbcglobal.net) | |
| 16:50 | <lifeboy> Yes. I said 'y' instead of 'm'
| |
| 16:50 | <alkisg> OK, continue with the update-image + kernel + reboot client, bbl..
| |
| 16:55 | alkisg has left IRC (alkisg!alkisg@ubuntu/member/alkisg, Ping timeout: 240 seconds) | |
| 16:57 | <lifeboy> alkisg: I still get the same error! "No init found. Try passing init= bootarg."
| |
| 16:58 | mgariepy has left IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy, Quit: Leaving) | |
| 17:02 | * lifeboy is away: Away at the moment | |
| 17:09 | lifeboy has left IRC (lifeboy!~roland@196-210-157-244.dynamic.isadsl.co.za, Quit: Leaving) | |
| 17:10 | lifeboy has joined IRC (lifeboy!~roland@196-210-157-244.dynamic.isadsl.co.za) | |
| 17:16 | mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy) | |
| 17:23 | vagrantc has joined IRC (vagrantc!~vagrant@c-76-105-252-50.hsd1.or.comcast.net) | |
| 17:34 | vagrantc has left IRC (vagrantc!~vagrant@c-76-105-252-50.hsd1.or.comcast.net, Ping timeout: 256 seconds) | |
| 17:41 | alkisg has joined IRC (alkisg!alkisg@ubuntu/member/alkisg) | |
| 17:42 | <alkisg> lifeboy: do you get an initramfs shell after that?
| |
| 17:43 | <lifeboy> yes
| |
| 17:44 | <alkisg> lifeboy: there, try: ls /dev/nbd*
| |
| 17:44 | Do you see the nbd devices?
| |
| 17:45 | <lifeboy> yes
| |
| 17:45 | <alkisg> OK, try: nbd-client server-ip 2000 /dev/nbd0
| |
| 17:45 | <lifeboy> /dev/nbd0 to /dev/nbd15
| |
| 17:46 | <alkisg> Output?
| |
| 17:46 | <lifeboy> Error: Cannot open NBD: No such file or directory
| |
| 17:46 | oh, wait
| |
| 17:47 | I typed dbd0 instead of nbd0
| |
| 17:47 | Now:
| |
| 17:48 | Negotiation: ..size = 1235132KB bs=1024, sz=1235132
| |
| 17:49 | <alkisg> OK, now: mount -o loop /dev/nbd0 /mnt
| |
| 17:49 | Or something similar, you may not have /mnt already so you'd need to create it
| |
| 17:50 | And another command: echo $BOOT
| |
| 17:51 | <lifeboy> mount is successful after creating /mnt
| |
| 17:51 | $BOOT = ltsp_nbd
| |
| 17:51 | <alkisg> ls -l /mnt init
| |
| 17:51 | err sbin/init
| |
| 17:51 | ls -l /mnt/sbin/init
| |
| 17:52 | Ahh you're in 10.04 so you're probably using nbd-proxy by default, that could be a problem as well...
| |
| 17:53 | <lifeboy> SQUASHFS error: unable to read metadata cache entry [4b6141a5]
| |
| 17:53 | <alkisg> Ah
| |
| 17:53 | <lifeboy> ls: /mnt/sbin/init: Input/output error
| |
| 17:54 | <alkisg> Yeah, that's your problem
| |
| 17:54 | So nbd isn't working correctly
| |
| 17:54 | <lifeboy> nbd-proxy?
| |
| 17:54 | <alkisg> We tried manually now so we didn't use nbd-proxy
| |
| 17:54 | So no, nbd-proxy isn't your problem here, it's a problem with the nbd module not working
| |
| 17:55 | <lifeboy> but pxelinux.cfg/default does have nbd_proxy=false
| |
| 17:55 | <alkisg> It doesn't take effect in 10.4
| |
| 17:55 | You'd need to patch ltsp_nbd to make it effect
| |
| 17:55 | But anyway let's not talk about nbd-proxy anymore, your problem is the nbd module
| |
| 17:56 | Which isn't working. You succesfully mounted the server image with nbd, and you couldn't `ls` it
| |
| 17:56 | So either your server nbd image is broken, or nbd isn't working, or squashfs isn't working
| |
| 17:56 | <lifeboy> hmm... Ik read something that somewhere... the nbd_proxy that was causing trouble
| |
| 17:56 | <alkisg> Yeah but we didn't use nbd-proxy
| |
| 17:57 | So it's not possible for it to cause problems when we're not using
| |
| 17:57 | it
| |
| 17:57 | Steve_the_Pirate has left IRC (Steve_the_Pirate!~Gary@host86-166-104-72.range86-166.btcentralplus.com, Ping timeout: 260 seconds) | |
| 17:57 | <lifeboy> I had a stgraber ppa that I install an never ltsp-client from in an earlier attempt. That won't help, will it?
| |
| 17:58 | It's the nbd.c in the vortex kernel, is it?
| |
| 17:58 | <alkisg> (07:56:34 μμ) alkisg: So either your server nbd image is broken, or nbd isn't working, or squashfs isn't working
| |
| 17:59 | <lifeboy> Can we find that out in another way or should I compare the sources with the standard ubuntu 10.04 kernel?
| |
| 18:01 | I have booted the standard i386 kernel via PXE from my notebook repeatedly, so I suppose that means the server is ok? Is is that a different kernel and doesn't mean anything here?
| |
| 18:01 | Or is that... I meant
| |
| 18:01 | <alkisg> The i386 kernel with today's nbd image?
| |
| 18:01 | If so, then yes it would mean that your server is ok
| |
| 18:02 | <lifeboy> Both with the vortex kernel, with the ubuntu 10.04 kernel
| |
| 18:02 | Sorry: Not with the vortex kernel, with the ubuntu 10.04 kernel
| |
| 18:02 | <alkisg> Today?
| |
| 18:03 | <lifeboy> No, earlier in the week
| |
| 18:03 | <alkisg> Try now
| |
| 18:03 | <lifeboy> Ok, gimme a sec
| |
| 18:03 | <alkisg> Just modify your pxelinux.cfg to point to the 10.04 kernel
| |
| 18:03 | <lifeboy> yes
| |
| 18:03 | <alkisg> If that makes your laptop boot, then your server is ok
| |
| 18:10 | Steve_the_Pirate has joined IRC (Steve_the_Pirate!~Gary@84.45.15.228) | |
| 18:10 | ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood) | |
| 18:12 | ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de) | |
| 18:17 | vagrantc has joined IRC (vagrantc!~vagrant@75-150-46-245-Oregon.hfc.comcastbusiness.net) | |
| 18:25 | <alkisg> vagrantc: I pushed openssl support in epoptes
| |
| 18:25 | I also put a command line param in epoptes-client to fetch the certificate from the server, epoptes-client -c
| |
| 18:26 | (so that ssh/scp aren't required, which would make it harder for some cases)
| |
| 18:26 | If it's acceptable, i'd like to put a "epoptes-client -c localhost" on epoptes-client.postinst, if we're in an LTSP chroot
| |
| 18:28 | vmlintu has joined IRC (vmlintu!~vmlintu@nblzone-240-143.nblnetworks.fi) | |
| 18:33 | <Hyperbyte> alkisg, you rock1
| |
| 18:33 | !
| |
| 18:33 | <vagrantc> alkisg: tested it last night, thouggh couldn't get vnc connections to work
| |
| 18:33 | <alkisg> Hyperbyte: let's hope it also works :)
| |
| 18:34 | <vagrantc> found a few more dependencies, but didn't have time to commit
| |
| 18:35 | <alkisg> vagrantc: I'll check if it's better to switch to remmina, if it is, no point in troubleshooting vinagre in lxde then
| |
| 18:36 | But afaik it worked for me, maybe some firewall?
| |
| 18:36 | I wonder if we should use forward vnc connections instead of reverse...
| |
| 18:37 | (battery end, bbl, will read logs)
| |
| 18:39 | lifeboy has left IRC (lifeboy!~roland@196-210-157-244.dynamic.isadsl.co.za, Ping timeout: 260 seconds) | |
| 18:40 | lifeboy has joined IRC (lifeboy!~roland@41.183.26.104) | |
| 18:41 | <lifeboy> alkisg: I can boot from my notebook, although it doesn't allow me to log in, but that a different problem I suppose
| |
| 18:42 | So the nbd server is fine then, right?
| |
| 18:42 | alkisg has left IRC (alkisg!alkisg@ubuntu/member/alkisg, Ping timeout: 256 seconds) | |
| 18:47 | Mip5 has joined IRC (Mip5!~chatzilla@208.72.71.69) | |
| 18:48 | <lifeboy> alkisg: I'll continue tomorrow. I was up far too long already last night. Thanks for all the help! You're really appreciated over here!
| |
| 18:52 | Steve_the_Pirate has left IRC (Steve_the_Pirate!~Gary@84.45.15.228, Ping timeout: 260 seconds) | |
| 18:58 | ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood) | |
| 18:59 | <vagrantc> alkisg: i'll have time to really test tomorrow... hopefully you'll be around?
| |
| 18:59 | ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de) | |
| 19:01 | alkisg has joined IRC (alkisg!alkisg@ubuntu/member/alkisg) | |
| 19:04 | <alkisg> Yup, tomorrow I'll be around, then I'll be travelling back
| |
| 19:04 | * alkisg checks at remmina... | |
| 19:06 | <alkisg> Meh, 569 is a well-defined port for ms-roma, we need to change it...
| |
| 19:07 | <Hyperbyte> 1533
| |
| 19:07 | <alkisg> No we'd better use a privileged port, < 1024
| |
| 19:07 | <Hyperbyte> :(
| |
| 19:07 | But I like 1533!
| |
| 19:07 | <alkisg> http://www.iana.org/assignments/service-names-port-numbers/service-names-port-numbers.xml
| |
| 19:07 | Hehe
| |
| 19:08 | OK then I'll put an "if hostname=hyperbyte, select 1533" in it - anyone other has some specific request? :P
| |
| 19:08 | <Hyperbyte> Rather put
| |
| 19:08 | if hostname!=alkisg
| |
| 19:08 | :P
| |
| 19:09 | <alkisg> virtual-places 1533 tcp Virtual Places Software [Avshalom_Houri] [Avshalom_Houri]
| |
| 19:09 | Sir, your port is also reserved
| |
| 19:09 | <Hyperbyte> Avshalom Houri?
| |
| 19:09 | lol
| |
| 19:10 | That's the worst name I've ever heard for.... well... basically anything.
| |
| 19:10 | Sounds like something you order at a Turkish restaurant.
| |
| 19:11 | So I guess for a Turkish dish it'd be an ok name.
| |
| 19:12 | Ok... now I'm hungry.
| |
| 19:12 | <alkisg> I hope he doesn't read the irc logs though :P
| |
| 19:12 | <Hyperbyte> Avshalom Houri is a person?
| |
| 19:13 | And he reserved port 1533 just for himself?
| |
| 19:13 | <alkisg> Service Name Port Number Transport Protocol Description Assignee Contact
| |
| 19:14 | www-http 80 tcp World Wide Web HTTP [Tim_Berners_Lee] [Tim_Berners_Lee]
| |
| 19:15 | So yeah Tim_Bernes_Lee has the whole web reserved for himself :P
| |
| 19:15 | <Hyperbyte> Oh that's just contact person
| |
| 19:15 | I thought it was the name of some software.
| |
| 19:15 | For a person name it's okay too I guess! Sorry Avshalom!
| |
| 19:15 | <alkisg> Nah, it's the name of the person that filed the port application
| |
| 19:15 | * Hyperbyte hides. | |
| 19:15 | <alkisg> Haha
| |
| 19:15 | <Hyperbyte> In all seriousness... I need Turkish food now.
| |
| 19:15 | Back later.
| |
| 19:16 | <alkisg> bb
| |
| 19:16 | nbd 10809 tcp Linux Network Block Device [Wouter_Verhelst] [Wouter_Verhelst] 2010-08-02
| |
| 19:16 | Goooood Wouter :)
| |
| 19:21 | <vagrantc> alkisg: there's no real reason to use a "privledged" port
| |
| 19:21 | alkisg: and the software needs to be able to support arbitrary ports
| |
| 19:22 | <alkisg> vagrantc: suppose for some reason the service hasn't started, and a person on the thin client runs a copy of his own for the daemon
| |
| 19:22 | Ah.
| |
| 19:22 | Keys :)
| |
| 19:22 | We now have keys :D
| |
| 19:23 | Well, there's a PORT setting in the configuration files, so it supports arbitrary ports
| |
| 19:23 | shawnp0wers has left IRC (shawnp0wers!~spowers@71-13-74-18.static.aldl.mi.charter.com, Quit: leaving) | |
| 19:23 | <alkisg> So anyway, we'll have to provide 1 default port, so should I select a system one, or one > 1024?
| |
| 19:23 | E.g. 585, 659 are currently available...
| |
| 19:23 | shawnp0wers has joined IRC (shawnp0wers!~spowers@71-13-74-18.static.aldl.mi.charter.com) | |
| 19:23 | shawnp0wers has joined IRC (shawnp0wers!~spowers@linuxjournal/staff/shawnp0wers) | |
| 19:25 | <vagrantc> alkisg: i need to look into debian policy as to how to handle arbitrary ports not assigned by IANNA
| |
| 19:25 | tomorrow :)
| |
| 19:26 | <alkisg> Nice
| |
| 19:29 | artista_frustrad has joined IRC (artista_frustrad!~fernando@200.247.43.2) | |
| 19:29 | <alkisg> 789 is unassigned, looks like a good number
| |
| 19:30 | komunista has joined IRC (komunista!~slavko@adsl-195-168-248-132.dynamic.nextra.sk) | |
| 19:30 | vagrantc has left IRC (vagrantc!~vagrant@75-150-46-245-Oregon.hfc.comcastbusiness.net, Ping timeout: 240 seconds) | |
| 19:32 | vagrantc has joined IRC (vagrantc!~vagrant@75-150-46-245-Oregon.hfc.comcastbusiness.net) | |
| 20:07 | vagrantc has left IRC (vagrantc!~vagrant@75-150-46-245-Oregon.hfc.comcastbusiness.net, Ping timeout: 260 seconds) | |
| 20:08 | Rafik has joined IRC (Rafik!Rafik@ubuntu/member/rafik) | |
| 20:09 | Parker955_Away is now known as Parker955 | |
| 20:11 | <alkisg> Hey Rafik :)
| |
| 20:11 | toscalix has joined IRC (toscalix!~toscalix@31.4.43.72) | |
| 20:13 | toscalix has left IRC (toscalix!~toscalix@31.4.43.72, Read error: Connection reset by peer) | |
| 20:16 | lifeboy has left IRC (lifeboy!~roland@41.183.26.104, Quit: Leaving) | |
| 20:22 | hughessd has left IRC (hughessd!~hughessd@173-164-117-109-Oregon.hfc.comcastbusiness.net, Quit: hughessd) | |
| 20:23 | hughessd has joined IRC (hughessd!~hughessd@173-164-117-109-Oregon.hfc.comcastbusiness.net) | |
| 20:33 | ogra_ has left IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de, Excess Flood) | |
| 20:35 | ogra_ has joined IRC (ogra_!~ogra@p5098ed03.dip0.t-ipconnect.de) | |
| 20:43 | hughessd has joined IRC (hughessd!~hughessd@173-164-117-109-Oregon.hfc.comcastbusiness.net) | |
| 20:44 | <Mip5> Have any of you folks used Xexit to run scripts for thin clients?
| |
| 20:45 | brunolambert has left IRC (brunolambert!~bruno@nat/revolutionlinux/x-fxngeoqyrjyrprhv, Ping timeout: 244 seconds) | |
| 20:52 | <Hyperbyte> Mip5, I take it you didn't get it working via pam?
| |
| 21:00 | komunista has left IRC (komunista!~slavko@adsl-195-168-248-132.dynamic.nextra.sk, Quit: Leaving.) | |
| 21:04 | alkisg has left IRC (alkisg!alkisg@ubuntu/member/alkisg, Ping timeout: 244 seconds) | |
| 21:05 | Rafik has left IRC (Rafik!Rafik@ubuntu/member/rafik, Read error: Operation timed out) | |
| 21:09 | <Mip5> Hyperbyte: No, I've put my script there, but it hasn't worked. I'm not sure though whether my script should be a full on bash script, or just a file with the command in it (rm -rf .gnome). Thoughts?
| |
| 21:14 | <Gadi> Mip5: did you actually read this? http://www.opinsys.fi/en/killing-user-processes-with-xexit-on-ltsp-servers
| |
| 21:15 | if you like, you can have it *only* run your script by modifying /etc/X11/Xsession.d/90xexit
| |
| 21:15 | so that it calls "xexit /path/to/your/script"
| |
| 21:16 | you should also be able to test it from command line
| |
| 21:16 | by running "xexit /path/to/your/script"
| |
| 21:16 | then check to see if xexit is running with ps
| |
| 21:16 | if it is, great
| |
| 21:16 | then logout of the thin client
| |
| 21:17 | and check to see if that xexit process goes away
| |
| 21:17 | and yeah, your script should be a chmod 755 bash script
| |
| 21:17 | <Mip5> Gadi: Yes, I did read the link. I must just be missing something.
| |
| 21:17 | <Gadi> :)
| |
| 21:18 | vmlintu has left IRC (vmlintu!~vmlintu@nblzone-240-143.nblnetworks.fi, Ping timeout: 252 seconds) | |
| 21:19 | <Mip5> Gadi: Thanks - the reason I asked about the 755 bash script is because the example script in the directory is *not* executable, nor is it a bash script - rather simply command
| |
| 21:19 | <Gadi> sorry. that question reads too sarcastically
| |
| 21:19 | :)
| |
| 21:19 | <Mip5> It's all good - there's tons I don't know, and you folks are truly a wonderful resource for folks like me
| |
| 21:19 | Rafik has joined IRC (Rafik!Rafik@conference/ubuntudevelopersummit/x-dihtehbvrvxzalwf) | |
| 21:19 | Rafik has joined IRC (Rafik!Rafik@ubuntu/member/rafik) | |
| 21:19 | <Gadi> I suspect that the files in xexit.d/ are sourced
| |
| 21:20 | but, if you actually call "xexit /script", /script should be executable
| |
| 21:20 | that's kind of our style in ltsp. To source .d/ directory scripts
| |
| 21:21 | <Mip5> Thanks - what does "sourced" mean in your previous comment?
| |
| 21:21 | <Gadi> so, I imagine that's how scotty coded xexit.d
| |
| 21:21 | ah
| |
| 21:21 | so, with shell scripts, you can source other files
| |
| 21:21 | like: . /my/file
| |
| 21:21 | <Mip5> ok....
| |
| 21:21 | <Gadi> the "." means source
| |
| 21:21 | which is like "include"
| |
| 21:22 | as opposed to "run"
| |
| 21:22 | every time you run a shell script from a shell script, it uses more memory
| |
| 21:22 | dead_inside has left IRC (dead_inside!~dead_insi@76.75.3.174, Read error: Operation timed out) | |
| 21:22 | <Gadi> because it spawns another shell
| |
| 21:22 | and it also takes a little longer
| |
| 21:22 | but, when you include,
| |
| 21:22 | the computer is only using one shell
| |
| 21:23 | and just executing the lines it finds in those other files
| |
| 21:23 | <Mip5> ah
| |
| 21:23 | <Gadi> so, as long as all the files are in shell,
| |
| 21:23 | it is faster to source than to execute
| |
| 21:23 | and sourced files don't need executable permissions, because they are not executed
| |
| 21:24 | only "included"
| |
| 21:24 | the only thing that was executed was the original shell
| |
| 21:24 | does that make sense?
| |
| 21:24 | * Gadi feels long-winded | |
| 21:24 | <Gadi> :D
| |
| 21:24 | <Mip5> ah - so sourced files are just included lines in something that is already being executed?
| |
| 21:24 | <Gadi> exactly
| |
| 21:24 | <Mip5> okay - that helps!
| |
| 21:25 | <Gadi> so, the only /bin/sh running is the first one in the master script
| |
| 21:25 | that is including all the rest
| |
| 21:26 | we are kinda obsessed here with minimizing our use of memory
| |
| 21:26 | as, well, at least inn the good ole days, we had very little to spare
| |
| 21:26 | but even these days, the less you waste, the faster things go
| |
| 21:29 | <Hyperbyte> Gadi, I wish all developers had your look on things. :)
| |
| 21:30 | <Mip5> ah - that makes sense. So, it seems the file in Xexit.d/ can just be a one liner (no shbang) that's not executable.
| |
| 21:30 | <Gadi> Hyperbyte: well, LTSP is a really good project for exposing waste in other projects
| |
| 21:30 | Mip5: that would be my guess
| |
| 21:31 | Mip5: to know for sure, you can look at that /usr/bin/EndXSession script
| |
| 21:31 | Mip5: and see if it uses "." to call the scripts in Xexit.d
| |
| 21:31 | or if it just calls them
| |
| 21:31 | brb
| |
| 21:36 | <Mip5> okay - I can't find /user/bin/EndXSession maybe that's the problem
| |
| 21:36 | I mean /usr/bin/EndXSession
| |
| 21:37 | Steve_the_Pirate has joined IRC (Steve_the_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com) | |
| 21:37 | <Mip5> Ah - I found it in /usr/share/Xexit/
| |
| 21:39 | Trixboxer has left IRC (Trixboxer!~Trixboxer@office.supportdepartment.net, Quit: "Achievement is not the end, its the beginning of new journey !!!") | |
| 21:43 | Steve_the_Pirate has left IRC (Steve_the_Pirate!~Gary@cpc16-haye17-2-0-cust169.haye.cable.virginmedia.com, Ping timeout: 260 seconds) | |
| 21:44 | darkpixel_ has joined IRC (darkpixel_!~darkpixel@curetheitch/staff/darkpixel) | |
| 21:48 | <Mip5> Gadi: Yeah, it looks to be sourced by the file /usr/share/Xexit/EndXSession. I'll keep playing with this. Thanks again for your help and explanations.
| |
| 21:53 | Curiously - when I ps aux | grep Xexit, I see several instances of the program running. However, all the instances of my ltsp logins show uid, and not login name. In contrast, the instance associated with my session shows my username.
| |
| 21:55 | <darkpixel_> I have two separate sites running LTSP under Ubuntu 10.04 (Lucid). About a week ago thin clients started not coming back up after rebooting. They simply flash like they are trying to start x but fail. The client logs show nbd-client giving an error message about an invalid partition. The server shows DHCP requests, and nbd connections followed by nbd disconnects.
| |
| 21:56 | Someone mentioned yesterday to try disabling compiz, that had no effect. The clients aren't getting far enough to log data to .xsession-errors. I've even tried rolling the nbd packages on my system back to the original archive version with no change. Does anyone have any pointers?
| |
| 21:58 | <Mip5> darkpixel_: Have you checked more mundane level switching? Do you still have good connectivity to your clients (ie, check the cable...)? Does it happen to all thin clients, or just some on some aspect of the network?
| |
| 21:59 | <darkpixel_> Mip5: All thin clients. One site has 13, the other has 3.
| |
| 22:00 | Mip5: Also, connectivity appears good. Pinging with large packets gives <1msec and no drops.
| |
| 22:00 | <Mip5> Had you run updates prior to the issue, or done anything else around the time of the issue? Any chance someone else has setup a dhcp server on your network?
| |
| 22:04 | Rafik has left IRC (Rafik!Rafik@ubuntu/member/rafik, Ping timeout: 240 seconds) | |
| 22:04 | <darkpixel_> Mip5: I usually connect in every 3-4 weeks and install updates. Unfortunately the server never gets rebooted (~240 days uptime). I have no clue how frequently the thin clients get rebooted. I did try rebooting the server to fix it. No effect. No rogue DHCP servers that I can see.
| |
| 22:06 | Mip5: I just disabled auto-login and my on-site tech tells me he gets a login screen now followed by "No response from server, restarting". I'm sure Google will be helpful with that one--I recall dealing with it last year at some point.
| |
| 22:06 | <Mip5> darkpixel_: have you tried rebuilding the image?
| |
| 22:06 | <darkpixel_> Mip5: Yes.
| |
| 22:09 | <Mip5> darkpixel_: sometimes packages can be updated and things can go wrong. Have you tried running apt-get to search for broken or mis/incomplete packages on your server, and in your chrooted environment?
| |
| 22:10 | <darkpixel_> Mip5: Just figured it out. Apparently one of the techs was making a change to our central puppet server and pushed out 'PasswordLogins No' to SSH on all our servers--not just our external servers.
| |
| 22:10 | <Mip5> Ah - excellent! Good find
| |
| 22:12 | <matrix3000> to use the NFS_HOME= definition to access an external NFS server i could do NFS_HOME=192.168.2.17:/home
| |
| 22:12 | ?
| |
| 22:13 | <darkpixel_> Mip5: Now to figure out how to get an account on the LTSP wiki so I can add that....
| |
| 22:13 | <Mip5> ;-)
| |
| 22:17 | vagrantc has joined IRC (vagrantc!~vagrant@75-150-46-245-Oregon.hfc.comcastbusiness.net) | |
| 22:19 | cliebow_ has joined IRC (cliebow_!~cliebow@66.63.66.218) | |
| 22:23 | bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Ping timeout: 255 seconds) | |
| 22:26 | <matrix3000> anyone?
| |
| 22:26 | to use the NFS_HOME= definition to access an external NFS server i could do NFS_HOME=192.168.2.17:/home
| |
| 22:30 | <vagrantc> hrm. apparently undocumented in lts.conf manpage...
| |
| 22:30 | * vagrantc makes a note to figure that out and update it later | |
| 22:32 | <Mip5> Well thanks gang. Talk to you tomorrow, g'night
| |
| 22:32 | Mip5 has left IRC (Mip5!~chatzilla@208.72.71.69) | |
| 22:46 | <matrix3000> vagrantc: thank you man, i know you can do the /home, but double mounting an nfs mount doesn't sound like a smart idea
| |
| 22:46 | <Gadi> matrix3000: that's right
| |
| 22:46 | it works just as you think
| |
| 22:46 | <matrix3000> sweet really!
| |
| 22:47 | <Gadi> yeah, if you leave out the IP, it just assumes the boot server
| |
| 22:47 | but, you can specify the IP to be somewhere else
| |
| 22:47 | <matrix3000> and if i put in the ip it looks for the NFS at the ip
| |
| 22:47 | <Gadi> right
| |
| 22:47 | <matrix3000> and i would use :/home at the end of it
| |
| 22:47 | <Gadi> yup
| |
| 22:47 | <matrix3000> like i was doing a reall nfs mount in fstab
| |
| 22:47 | <Gadi> yup
| |
| 22:48 | <matrix3000> sweet, yea, trying to finally conquer that Google Chrome singletonlock crap
| |
| 22:48 | <Gadi> there is also NFS_HOME_OPTIONS
| |
| 22:49 | <matrix3000> and what is that?
| |
| 22:49 | <Gadi> so, you may also want: NFS_HOME_OPTIONS="defaults,nolock"
| |
| 22:49 | guess
| |
| 22:49 | :)
| |
| 22:49 | <matrix3000> ah ok
| |
| 22:49 | any documentation on that?
| |
| 22:50 | <Gadi> ask alkisg - its his code ;)
| |
| 22:50 | <matrix3000> ah ok
| |
| 22:50 | yea, so prolly not
| |
| 22:50 | lol
| |
| 22:50 | and defaults no lock will do what?
| |
| 22:50 | <Gadi> mount -t nfs -o defaults,nolock $NFS_HOME /home
| |
| 22:50 | in essence
| |
| 22:50 | :)
| |
| 22:51 | <matrix3000> cause i never used any of that when i did the fstab
| |
| 22:51 | before
| |
| 22:52 | reading the manpage now for nfs
| |
| 22:52 | ah ok
| |
| 22:52 | i got it now
| |
| 23:05 | alkisg has joined IRC (alkisg!alkisg@ubuntu/member/alkisg) | |
| 23:13 | <darkpixel_> So how can I get edit access to the wiki to add information about my problem with SSH?
| |
| 23:18 | <alkisg> darkpixel_: the ltsp wiki or the ubuntultsp wiki?
| |
| 23:19 | <darkpixel_> alkisg: The one on sourceforge. Should I just type something up into pastebin and give the link to an admin to post?
| |
| 23:20 | <alkisg> Not sure, that wiki is kinda unmaintained, and Hyperbyte's planning to revive it
| |
| 23:20 | So maybe the best thing is to wait a couple of weeks until it's taken its new form
| |
| 23:21 | <darkpixel_> alkisg: Sounds good--I'll try back in a few weeks. Thanks for your help yesterday by the way. Tuned out our puppet server pushed 'PasswordAuthentication no' to all our ssh servers. (All our admins use ssh keys anyways) Finally figured out that affects LTSP. ;)
| |
| 23:22 | <alkisg> Yeah and unfortunately it doesn't display the ssh warnings/errors
| |
| 23:22 | Hope it will on the next version
| |
| 23:23 | <darkpixel_> alkisg: Yup. Oh well. I was about to fire up strace when I happened to glance over at the commit list from our management server and noticed the ssh change...
| |
| 23:24 | alkisg: Gotta reboot and tackle a stupid alsa/pulse/unity/gnome/webcam/ubuntu sound issue. I can't think of anything worse than audio issues in Ubuntu. ;)
| |
| 23:24 | Later
| |
| 23:24 | <alkisg> bb
| |
| 23:24 | darkpixel_ has left IRC (darkpixel_!~darkpixel@curetheitch/staff/darkpixel, Quit: Leaving.) | |
| 23:24 | <Gadi> alkisg: so, I think we should use /var/cache/ltsp/ltsp_config to store some initramfs vars for our new initramfs-scripts script to use
| |
| 23:25 | in other words, things like
| |
| 23:25 | <alkisg> Gadi, don't we store stuff there already?
| |
| 23:25 | <Gadi> NBD_ROOT_HOST, DNS_SERVER, etc
| |
| 23:25 | yeah
| |
| 23:25 | some
| |
| 23:25 | <alkisg> Ah you mean to put more stuff there
| |
| 23:25 | <Gadi> but, I am looking at our nfs-bottom script
| |
| 23:25 | <alkisg> Indeed, and boot_iface will be useful
| |
| 23:25 | <Gadi> and that should be reimplemented
| |
| 23:25 | yeah
| |
| 23:25 | <alkisg> nfs-bottom? why do we have that?
| |
| 23:26 | <Gadi> I feel like I am dismantling lots of stuff
| |
| 23:26 | :)
| |
| 23:26 | <alkisg> i mean, we need one script for premount, and one at init-bottom, right?
| |
| 23:26 | <Gadi> right
| |
| 23:26 | that's where we are headed
| |
| 23:26 | what we currently have is nfs-bottom
| |
| 23:26 | and ltsp_nbd
| |
| 23:26 | <alkisg> OK, let me see that...
| |
| 23:27 | <Gadi> we should add code to our new init-bottom script that writes certain vars to $rootmnt/var/cache/ltsp/ltsp_config
| |
| 23:28 | then, our initramfs-scripts script can use those vars to set the conf files
| |
| 23:28 | <alkisg> Everything at nfs-bottom should go away and be put in plugins, right?
| |
| 23:28 | <Gadi> right
| |
| 23:28 | and I suggest using that cache file to preserve certain env vars
| |
| 23:29 | (that will otherwise be lost)
| |
| 23:29 | <alkisg> Also, aren't those env vars available from the plugins?
| |
| 23:29 | <Gadi> no - they will be lost when we run chroot $rootmnt initramfs-scripts
| |
| 23:30 | gotta run for a bit
| |
| 23:30 | <alkisg> ?
| |
| 23:30 | <Gadi> but give it some thought
| |
| 23:30 | <alkisg> chroot preserves the vars for me
| |
| 23:30 | <Gadi> does it?
| |
| 23:30 | <alkisg> Yup
| |
| 23:30 | <Gadi> ok
| |
| 23:30 | well, then maybe we don't have to
| |
| 23:30 | <alkisg> So our bottom plugin can be really simple
| |
| 23:30 | *script
| |
| 23:30 | <Gadi> right
| |
| 23:30 | <alkisg> And most of the code can go in plugins
| |
| 23:30 | <Gadi> ok brb
| |
| 23:31 | <alkisg> OK, bb, cu later
| |
| 23:46 | Gadi has left IRC (Gadi!~romm@ool-4571ca04.dyn.optonline.net, Read error: Connection reset by peer) | |
| 23:53 | <matrix3000> http://screencast.com/t/oRdooeeIp7t
| |
| 23:53 | alkisg, where should i look to fix that, never seen that before
| |
| 23:54 | <alkisg> matrix3000: please ask in the channel, don't ping specific persons... I don't know I've never seen that either, what's your pxelinux.cfg/default like?
| |
| 23:56 | <matrix3000> ok sorry
| |
| 23:56 | ill just rebuild it
| |
| 23:57 | Matrix30001 has joined IRC (Matrix30001!~matrix300@69.58.116.65) | |