[03:00] drewr (~drew@adsl-065-013-142-013.sip.bna.bellsouth.net) left irc: Read error: Connection reset by peer [05:49] oz_ (~oz@anubis.coresystems.de) left irc: Server closed connection [05:49] oz_ (~oz@anubis.coresystems.de) joined #fai. [08:19] sep_ (~sep@170.217-149-115.enivest.net) joined #fai. [08:21] sep_ (~sep@170.217-149-115.enivest.net) left irc: Remote host closed the connection [08:29] sep_ (~sep@170.217-149-115.enivest.net) joined #fai. [08:31] sepski (~sep@raserv.aasen.cx) left irc: Server closed connection [08:31] sepski (~sep@raserv.aasen.cx) joined #fai. [08:53] nik0la_ (~nik0la@admin.objectifgroove.com) left irc: Read error: Connection reset by peer [08:56] \sh (~shermann@newzealand.sourcecode.de) joined #fai. [08:56] <\sh> moins [08:58] nik0la (~nik0la@admin.objectifgroove.com) joined #fai. [09:04] incognitu (incognitu@nettuning.nl) left irc: Ping timeout: 480 seconds [09:31] Winkie (~urmom@ur.fa.gs) joined #fai. [09:31] oh hey my PC crashed :o [09:44] sep_ (~sep@170.217-149-115.enivest.net) left irc: Quit: Leaving [09:52] Action: oz_ @ fai softupdates [10:00] Mrfai (~lange@suenner.informatik.uni-koeln.de) joined #fai. [10:06] so i was talking about installing ubuntu using a debian FAI server the other day, and i realised something [10:06] i've never done a FAI softupdate [10:06] and i don't know what the procedure would be with ubuntu installed [10:06] anyone want to enlighten me? [10:06] i don't see why it should be any different than with debian [10:06] mgoetze: i don't even know how softupdates actually work :D [10:08] on a server which was installed with fai and has a correct /etc/fai/fai.conf, you simply run "fai softupdate" and it will update the software, go through the scripts again, etc., so that, if you did everything right, you will end up with a system just like if you had installed it from the current configuration [10:08] ...just faster. [10:09] and without significant downtime, generally [10:09] Action: oz_ really likes softupdates [10:09] ah i see, well i'll check it out once i've gotten the debian installation image booting on this machine [10:14] sep_ (~sep@174.62-50-191.enivest.net) joined #fai. [10:41] edoreld (~Adium@p57A0987D.dip0.t-ipconnect.de) joined #fai. [10:53] verwilst (~verwilst@router.begen1.office.netnoc.eu) joined #fai. [10:54] Nic0 (~courtel@fw.tls.cena.fr) left irc: Remote host closed the connection [10:57] Nic0 (~courtel@fw.tls.cena.fr) joined #fai. [11:09] I can see this class "DHCPC" in almost every kind of installation ( atom*, demohost, gnomehost, etc...), but I can't see any file with it's name in $FAICONFIG...does it do anything? [11:10] edoreld: man grep [11:10] mgoetze: :-))) [11:12] edoreld: "find /srv/fai/config -exec grep -l DHCPC {} \; " will should you the files which contain DHCPC [11:12] That worked ;) [11:12] used "grep -R "DHCPC" /srv/fai/config [11:12] " [11:12] ah, that's even simpler [11:12] Action: oz_ still is solaris infected [11:13] but I like that infection. ;) [11:14] heh ;) [11:33] simNIX (~simNIX@156-60.bbned.dsl.internl.net) left irc: Quit: Ik ga weg [12:09] bugger [12:09] well i got 2.6.29 booting [12:09] but looks like i should have included aufs with it [12:09] because it fails to mount with a no such device error [12:09] yeah, you should :) [12:09] i used debian's 2.6.29 config :( [12:09] is there a fai config somewhere i can steal? [12:09] it doesn't include aufs? if so, we'll have to file a bug report... [12:10] well, it might do, but the error i got was mounting aufs on /root failed: No such device [12:10] one second [12:10] debian-fai:/usr/src/linux# grep -i aufs .config [12:10] debian-fai:/usr/src/linux# [12:10] i'm going with no [12:11] ah, it's obviously not 'aufs' cause that isn't in 2.6.26 either [12:11] Winkie: you can alsoway use the simple examples in /usr/share/doc/fai-doc/examples/ [12:11] oz_: those are configuration files? [12:11] Winkie: If I would give you my config, I'd prolly get sacked from my job :) [12:12] Winkie: yes, those are config examples. [12:12] oz_: FAI configuration is not a problem, getting a working booting kernel is the problem [12:12] this machine will not boot on < 2.6.28 [12:15] so you need this kernel in the nfsroot and on the client. [12:16] yes [12:16] i'm building a defaulted upgrade from the 2.6.26 config now [12:16] lets see what happens [12:16] (in about an hour) [12:17] i only have a little under 6 hours to finish off here, and i've managed to do none of the tasks needed lol [12:17] ah well [12:28] in Debian aufs is a seperate module. It's included in the package aufs-modules-2.6.26-2-486 and similar [12:31] Winkie: I would try this: get the linux-image packages for 2.6.29 from sid, and also the aufs-modules-2.6.29.... package. add this kernel and auf-modules to your nfsroot, that should work [12:32] Mrfai: oh bugger it's a modules package? typical of me to miss that [12:33] i'm guessing it doesn't need to be in the initrd [12:33] cheers :) [12:38] haha Mrfai i need linux-latest-modules-2.6.29-2-686 apparently, which doesn't exist, don't you love computing [12:39] forced it through as that's a virtual package [12:42] <\sh> Mrfai: alaaf :) it's live-initramfs which adds the Debian Live Initramfs Hooks [12:42] <\sh> pool/main/l/live-initramfs/live-initramfs_1.156.1+1.57.1-1_all.deb (from latest lenny) [12:44] <\sh> Mrfai: btw..whoever is involved in setup_storage -> a big big hug from me it's much better then the old way [12:49] ah Mrfai it looks like aufs does need to be in the initrd [12:50] i've updated it with the module installed, lets see! [12:50] 00 [12:51] nop failure :( [12:52] I was gonna say 1000111 01101111 01101111 01100100 00100000 01101100 01110101 01100011 01101011 00100001 but I guess not -.- [12:52] haha sorry that was because i'm lagged as hell on this terminal [12:52] switching screens with ^A-0 :D [12:52] anyway now i need to figure out how to stuff aufs into this initrd [12:54] hahaha oh man [12:54] i got aufs loaded, and then got the exact same segfault i was having in ubuntu [12:54] Action: Winkie bangs head on keyboard [12:55] i'm going to have a nice drink and a sit down [12:56] i would really appreciate some advice on this, it was a known problem when i was talking about it yesterday, but i have no good solution if the debian version also segfaults :( [12:57] <\sh> Winkie: when I checked the source yesterday, there was one entry in /etc/fai/live.conf to be set...UNIONFS=aufs or something like this, dunno if this is right [12:57] <\sh> export UNIONTYPE="aufs" [12:57] <\sh> ah there it is...already set [12:58] \sh: i'm actually using Lenny, I guess maybe I should ugprade my boot environment to sid? [12:58] oh wait live.conf [12:58] i don't even have a live.conf :( [12:58] <\sh> Winkie: which version of FAI are you using?= [12:58] \sh: the default in Lenny [12:58] 3.2.17~lenny1 [12:59] <\sh> hmm [12:59] <\sh> 3.2.20 here...but I'm using FAIs archive from http://www.informatik.uni-koeln.de/fai [12:59] \sh: my problem is that i have these Dell Optiplex 320s, which require >=2.6.28 to boot [12:59] and so i've installed the sid kernel package (2.6.29) and aufs-modules for it [12:59] and i am getting the same segfault as i had when i tried to use ubuntu as a FAI base [13:00] <\sh> can you make a screenshot or something like that? [13:00] this is definitely a known problem, there were discussions on it yesterday, but I don't know where the problem occurs or what it encompasses [13:00] \sh: i can take a picture on my phone if you'd like? :) [13:01] <\sh> Winkie: I want to read the errors ;) sure :) [13:02] \sh: pm me your email address [13:03] <\sh> Winkie: no need to pm: sh@sourcecode.de :) [13:06] \sh: sending now [13:06] \sh: on ubuntu this occured every time on every boot on every machine [13:06] i have 3 (actually 4 but the fourth is irrelevant) different machines i'm trying this on, only one needs 2.6.28 but it seems wise to try and use it everywhere [13:10] right well email should have been sent [13:10] i know that the ubuntu CD version boots fine, so i'm reasonably sure it's an aufs bug of some sort, but god knows what [13:10] let me find my logs from yesterday [13:11] haha oh it was you talking :) [13:11] excellent [13:12] <\sh> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/382842 [13:13] <\sh> I think that this bug report is important here ;) [13:14] perhaps, although this is an actual panic i think. did you get my email? [13:14] <\sh> yepp [13:15] excellent, out of interest did it come from my mob@gmail account? [13:17] <\sh> Winkie: the bug and your crash could be related (ubuntu wise) [13:17] <\sh> Winkie: I would like to test it myself...but I won't come to the point of testing until next week... [13:18] \sh: it's possible, i could look at attempting to upgrade aufs myself, if there's anything i can do to give you access to help me out let me know [13:18] you're welcome to have SSH access to the virtual environment i'm preparing this on [13:19] <\sh> Winkie: no need...I just have enough hardware to test...I only need to setup an ubuntu box with fai ... that's straight forward...but my time is limited today and tomorrow [13:19] i see, well thanks for the help anyway, i've been waiting months for ubuntu to even boot on these machines so a few more days won't hurt, i'm just in the office today so hoped to get it finished [13:20] i'll see if i can update aufs and recompile a module for it [13:20] not done that with debian's build system before [13:29] Winkie are you trying to install ubuntu from debian fai-server? [13:30] edoreld: ideally yes, but i haven't gotten as far as debian booting and installing debian yet [13:32] Winkie: ouh [13:32] Winkie: just wanted to know the quirks & twists of what booting ubuntu from debian require ^^ [13:33] edoreld: i'm not actually booting ubuntu, i'm booting debian and then going to copy a ubuntu bootstrap to the hard drive i think [13:35] aah yeah that sounds easier :) [13:53] \sh: you mean live-initramfs_1.156.1+1.57.1-1_all.deb is needed to fix the problems you described in your mail? But this version is already in lenny, so IMO I do not need to add this into my repository [13:54] <\sh> Mrfai: no...it's the package to be fixed :) the patch mentioned by Carsten needs to go in [13:55] <\sh> BBL...TTBB: 2h [13:56] Mrfai: any immediate suggestions for me? ;) [13:57] Winkie: no panic [13:57] ;-) [13:58] Mrfai: that could mean "panic" or "don't panic" :( [13:58] I would try to boot your machine using the debian kernel and aufs from Debian sid. [13:58] don't panic ;-) [13:59] Mrfai: that's exactly what i did, and it panics :) [13:59] i sent the images to \sh but i'll upload them, one minute [14:01] Winkie: did you every tried this with some other piece of hardware? [14:02] Mrfai: i did not, but this crash is exactly the same as i got using ubuntu's fai-server and i did try that on other hardware, would you like me to go and test this version on other hardware too? [14:06] Mrfai: http://i42.tinypic.com/14jx0r9.jpg http://i40.tinypic.com/2gtd75j.jpg http://i43.tinypic.com/291ynty.jpg [14:06] this is up to you. I don't have the overview which fai version you are using exactly. [14:06] IMO there are two possiblilities [14:06] 1) your hardware and the new kernel causes the kernel panic [14:07] 2) the new kernel and aufs is brken on all hardware [14:07] Mrfai: i'm using the lenny default version, which is 3.2.17~lenny1, with a 2.6.29 kernel from debian sid [14:07] Action: Mrfai must go to a meeting [14:07] now [14:07] i'm thinking #2 also [14:07] thanks for your help [14:08] drewr (~drew@adsl-065-013-142-013.sip.bna.bellsouth.net) joined #fai. [14:11] Mrfai: just checked, the same crash occurs on a different machine [14:44] siggg (~jean@tra78-5-88-172-128-32.fbx.proxad.net) joined #fai. [14:48] hm, i have a script files/etc/icecast2/icecast.xml/postinst, which is supposed to stop and start icecast [14:48] during a softupdate, the stopping works fine, but the starting doesn't :( [14:49] sep_ (~sep@174.62-50-191.enivest.net) left irc: Remote host closed the connection [14:49] here's an excerpt from shell.log: http://paste.debian.net/38023/ [14:50] any hints? [15:21] mgoetze: i am no fai expert, but i can't see anything obvious there, there are all sorts of possibilities for why icecast might be failing to start, have you considered modifying the init file to write logs somewhere / does it write logs by default? [15:23] I wonder why the output is so mangled... [15:23] and I remember icecast being a bitching piece of software in regard to restarts and so [15:39] yes, in fact i generally do ... stop; ... start rather than restart... maybe i should just rewrite the init script [15:47] sep_ (~sep@77.16.217.62.tmi.telenormobil.no) joined #fai. [16:19] sep_ (~sep@77.16.217.62.tmi.telenormobil.no) left irc: Remote host closed the connection [16:57] does anyone have experience with integrating a previously installed (without FAI) server into a FAI infrastructure, i.e. installing fai-client, editing /etc/fai/fai.conf and running fai softupdate? [16:59] mgoetze: hm. I did something like this recently [16:59] just without fai-client...that I did today. [16:59] did you encounter any stolpersteine? [16:59] mgoetze: was surprisingly easy [17:00] just write /var/log/fai/FAI_CLASSES, i guess? [17:00] the machine went into production just after one day... [17:00] and runs smooothly [17:00] I was REALLY surprised. [17:00] sounds good :) [17:01] mgoetze: I created one class GATEWAY, as the machine was out local gateway... [17:10] Mrfai (~lange@suenner.informatik.uni-koeln.de) left irc: Quit: leaving [17:25] Can I in $FAICONFIG/package_config/MYCLASS write a line like this after "PACKAGES aptitude" >> linux-headers-$(uname -r) << [17:34] i use spec. classes for those, because i use one kernel to install all machines, no matter of they shoud later be running -server -generic -rt or whatever. [17:35] right [17:35] edoreld (~Adium@p57A0987D.dip0.t-ipconnect.de) left irc: Quit: Leaving. [17:44] siggg (~jean@tra78-5-88-172-128-32.fbx.proxad.net) left irc: Remote host closed the connection [17:49] verwilst (~verwilst@router.begen1.office.netnoc.eu) left irc: Quit: Ex-Chat [18:39] elgskilt (~ro@p5B2ECCCA.dip.t-dialin.net) joined #fai. [19:04] Hi there! [19:05] Has anybody succeeded using FAI for Ubuntu 9.04 Jaunty? [19:52] elgskilt (~ro@p5B2ECCCA.dip.t-dialin.net) left irc: Quit: Leaving. [23:06] weasel00 (~weasel@66.201.51.66) joined #fai. [23:19] hello, [23:20] trying to use a fai-cd to boot client machines but they want to get their nfs shares and bootp info from a windows dhcp server, is there a way to make them look at the fai server for the install information? [fai-server on lenny] [23:21] i think the "rootserver" variable is the key but i dont where that definition is coming from or being created. [23:45] anyone.. how do i chnage the rootserver definition? [23:59] FAI_CONFIG_SRC is where they get there config from. [00:00] --- Fri Jun 5 2009