[00:24] eartoast (~glawe@login.physik.fu-berlin.de) left irc: Remote host closed the connection [07:05] ebbie (~username@berlin046.server4free.de) joined #fai. [07:44] ErKa (~keryell@81.255.107.242) joined #fai. [08:03] eartoast (~glawe@login.physik.fu-berlin.de) joined #fai. [08:21] justleen (~leen@rev-131-102.virtu.nl) joined #fai. [08:31] \sh (~shermann@nat.de.netviewer.com) joined #fai. [08:50] Mrfai (~lange@malzmuehle.informatik.uni-koeln.de) joined #fai. [09:46] verwilst (~verwilst@router.begen1.office.netnoc.eu) joined #fai. [09:48] Mrfai (~lange@malzmuehle.informatik.uni-koeln.de) left irc: Quit: leaving [09:48] Mrfai (~lange@kueppers.informatik.uni-koeln.de) joined #fai. [10:00] pav5088 (~Mark@120.153.246.129) joined #fai. [10:11] Mrfai (~lange@kueppers.informatik.uni-koeln.de) left irc: Remote host closed the connection [10:12] Mrfai (~lange@kueppers.informatik.uni-koeln.de) joined #fai. [10:16] Mrfai (~lange@kueppers.informatik.uni-koeln.de) left irc: [10:16] Mrfai (~lange@malzmuehle.informatik.uni-koeln.de) joined #fai. [11:31] Mrfai: http://grml.org/patches/0001-Adjust-depends-recommends-regarding-tftp-server.patch - ack? [11:39] pav5088 (~Mark@120.153.246.129) left irc: Ping timeout: 480 seconds [11:39] pav5088 (~Mark@120.152.63.212) joined #fai. [11:41] mikap: nicht ganz. Zitat aus dem tftpd Paket: Tftpd is not suitable for use with the PXE bootloader; for that, use atftpd or tftpd-hpa. [11:41] ICh denke tftpd sollte ganz rausgenommen werden [11:41] also nur atftpd oder tftpd-hpa [11:41] Mrfai: oke [11:42] Mrfai: willst du den dann nach trunk? [11:42] jo, der kann auch nach trunk [11:43] oke [11:47] das kann ich auch machen. [11:48] mach ich schon [11:49] ok [11:49] Action: Mrfai SIGFOOD [11:52] fai: 3 mika committed revision 6199 to 7 fai: Adjust depends/recommends regarding tftp server. [11:52] fai: fai-server package recommended 'tftpd-hpa | tftpd', now we're [11:52] fai: using 'tftpd-hpa | atftpd' instead as tftpd is not suitable for use [11:52] fai: with the PXE bootloader at all. [11:54] LasKaR (~cari_butu@112.78.43.249) joined #fai. [11:54] Mrfai: http://git.grml.org/?p=fai-stable.git;a=shortlog;h=refs/heads/mika/3.4.5-rc1 -> is up2date [11:58] LasKaR (~cari_butu@112.78.43.249) left irc: Remote host closed the connection [12:16] http://piuparts.debian.org/sid/fail/fai-nfsroot_3.4.4.log - fai-nfsroot.postrm should probably use "-f" for "rm /etc/init.d/rcS" [12:16] or maybe that rm is fully unneeeded as dpkg-divert (next line) will take care? [12:17] h01ger: good question, Mrfai? [12:55] not sure if the rm is unneeded. But adding -f is a good idea. [12:56] BTW, installing fai-nfsroot without using make-fai-nfsroot makes no sense. And I never removed this packages, I always rebuild the nfsroot. [12:56] Mrfai: but removing should work anyway, i think that's holger's point [13:00] Mrfai, installing a package and removing it again very seldom makes sense ;) but still, it should work. i know fai-nfsroot is special (its one out of 4 out of 30000 packages i whitelisted yesterday in piuparts) but IMO it would still be nice if its not special anymore and one can remove it just like any other package... [13:02] brrr, [FAI] svn is soooooo slow :-/ [13:02] - rm /etc/init.d/rcS [13:02] + rm -f /etc/init.d/rcS [13:02] trunk/debian/fai-nfsroot.postrm - ack, Mrfai / h01ger ? [13:03] ack [13:03] ack [13:03] i just wonder if that whole rm can go (because of the next line) [13:03] mikap: it's so slow because it's a virtual machine with a load about 15 sometimes > 20 [13:03] +vielen dank [13:04] it was much faster in the past. [13:04] h01ger: let's keep it for now, breaking this now isn't worth the effort IMO [13:04] yeah [13:04] Mrfai: it's slow because it's svn :) and the load on the machine is an additional minus ;) [13:05] the machine load used to be less [13:08] fai: 3 mika committed revision 6200 to 7 fai: fai-nfsroot.postrm: use rm with -f to not fail if /etc/init.d/rcS does not exist [13:08] fai: Reported by Holger and Piuparts. [13:08] fai: Acked-by: Thomas Lange [13:08] fai: Acked-by: Holger Levsen [13:10] -> Committed revision 6200. [13:10] \o/ [13:11] queued (AKA cherry-picked) for 3.4.5 as well [13:12] <\sh> hmmm...what could trigger a port connect from an install client to the fai server on port 65535? [13:12] <\sh> during fai deployment [13:13] \sh: trojaner or torrent? :) [13:13] <\sh> mikap: lol... [13:14] <\sh> from :897 to :65535 [13:14] <\sh> where the port 897 differs [13:15] sniff traffic on port 65535 [13:15] nfs? [13:15] initramfs-tools puts this into the kernel command line [13:19] <\sh> h01ger: hmmm??? [13:19] <\sh> h01ger: any reference inside /usr/share/initramfs-tools ? [13:20] 192.168.1.1:/opt/ltsp/i386 on / type nfs (ro,relatime,vers=3,rsize=524288,wsize=524288,namlen=255,hard,nolock,proto=tcp,port=65535,timeo=7,retrans=10,sec=sys,m [13:20] ountport=65535,addr=192.168.1.1) [13:20] is what i see on a ltsp client here [13:20] (output of mount) [13:20] also a bit struggling to find where _all of_ these options come from. for some i know [13:21] <\sh> correct [13:21] <\sh> funny that there is no "port=65535" somehow in /usr/share/initramfs-tools/ somehow [13:24] <\sh> anyways, it works [13:25] <\sh> btw...fai + puppet == success [13:27] Mrfai: http://git.grml.org/?p=fai-stable.git;a=blobdiff;f=3.4/debian/changelog;h=b64f2e07b7a98bf1ffce304e2aa2876f01922582;hp=ce7ae558ad69dce9dc837b308e5ab95f23f2af70;hb=5ebaa2c2a97618e9fdd74b26bcb1bbb6652fb751;hpb=d59d35f45c6f8284e3dcbcb044cbf50014550e55 -> git clone git://git.grml.org/fai-stable.git && cd fai-stable && git checkout -b mika/3.4.5-rc2 remotes/origin/mika/3.4.5-rc2 [13:30] nice url [13:34] <\sh> btw..dc2 qooxdoo frontend does run on chrome as extension...very nice... [13:36] <\sh> fai + lucid works like a charm , that means I will have time next week to package fai 3.4.4 for ubuntu...phew [13:36] 3.4.5 ;) [13:36] <\sh> or 3.4.5 ;) [13:37] <\sh> now I just need to find a good way how to package all dc2 source foo [13:38] <\sh> then I could release (DC)² Version 0.1 ;) [13:38] cool [13:39] mikap: changelog looks good. I will build rc2 out of those sources and give it a short try. [13:47] Mrfai: excellent [13:48] i've named it 3.4.5 already in the changelog for the merge (if everything works as expected, if we've to adjust something i'll just do a non-fast-forward/rebase) [14:18] mikap: a simple test with squeeze and my config and 3.4.5 works. I'm also using rc1 since a couple of days without problems. [14:19] So, IMO you can release 3.4.5 [14:21] Mrfai: ok, will do in the evening then, would like to give other guys time for testing as well (and hopefully i can execute kantan in the evening too) [14:22] ok. Fine for me. [14:22] good [14:23] whats kantan? [14:23] pav5088 (~Mark@120.152.63.212) left irc: Ping timeout: 480 seconds [14:25] h01ger: http://grml.org/kantan/ - "simple test suite for autotesting software using Grml and KVM" (may autotest fai software) [14:25] ah. remember ow [14:25] +n [14:25] cool [14:26] Mrfai, which reminds me: does fai support logging via syslog? [14:31] FAI has no syslog support . IMO there's to much output from an isntallation. My installations have several thousands lines of output. Do you like to have them in syslog? Me not. [14:31] <\sh> what was the magic again to set a hostname on the kernel commandline? [14:32] hostname=foobar [14:33] Mrfai, i've configured remote syslog already and this way i dont have to give the clients access to the server [14:34] <\sh> h01ger: eventually you can write a simple syslog push script which pushes only special logstatements [14:35] and then pull the logs from the clients [14:35] yes, that could work [14:35] s/c/sh/ even :) [14:35] <\sh> well, another idea could be to mount another nfs share to your logfile server and copy the logs in the LAST class call to your logfile server [14:36] while i could use nfs in this setup i dont like to rely on it [14:36] so, no :) [14:37] \sh, do you have an example of a simple syslog push script? [14:37] Mrfai: but logging to a remote server would be cool [14:37] ah. logger [14:37] \sh, thanks! [14:37] <\sh> h01ger: hmm...no...but I could come up with a simple python script... [14:38] <\sh> oh yes...logger works too [14:38] Action: \sh is a bit more cool and writes logstatements during fai setup via xmlrpc to a couchdb ;) [14:38] <\sh> but this is really questionable [14:39] enterprise crap ;-p [14:39] <\sh> rotfl [14:39] logger really is what i looked for [14:39] thanks a lot! [14:39] <\sh> h01ger: you're welcome :) always :) [14:39] :) [14:47] siggg (~jean@tra78-3-82-244-85-196.fbx.proxad.net) joined #fai. [15:34] siggg (~jean@tra78-3-82-244-85-196.fbx.proxad.net) left irc: Quit: Leaving. [15:53] mt: I found a GRUB_PC/10-setup bug. Posted to the mailing list [15:54] <\sh> Mrfai: which one? [15:54] grub is always installed to hd0. But my boot disk is hd3 [15:56] <\sh> yeah... [15:57] I wonder that noone else had this bug yet. Is anyone using FAI? ;-) [15:57] Action: h01ger softupdates only :) [16:03] siggg (~jean@AVelizy-151-1-89-138.w86-205.abo.wanadoo.fr) joined #fai. [16:08] <\sh> Mrfai: my boot device is hd0 always ;) [16:10] <\sh> Mrfai: well, how do we convert $BOOT_DEVICE to $GRUB_DEVICE? [16:14] Action: h01ger thinks grub knows how to [16:16] for grub1 we used device2grub [16:17] siggg (~jean@AVelizy-151-1-89-138.w86-205.abo.wanadoo.fr) left irc: Ping timeout: 480 seconds [16:18] speaking of, does anyone know how i can configure default kernel parameters for grub1 on ubuntu 10.04? [16:19] <\sh> Mrfai: yes..but I don't find any comparable tool for grub2 [16:19] <\sh> oh grub-probe could be our help [16:20] it should be easy to fix device2grub for grub2 [16:24] $ROOTCMD grub-probe -tdrive -d $BOOT_DEVICE [16:24] justleen (~leen@rev-131-102.virtu.nl) left irc: Quit: Leaving. [16:24] this should work [16:27] <\sh> hmmm [16:28] <\sh> grub-probe -tdrive -d /dev/sda output: (/dev/sda) [16:28] squeeze[~]> grub-probe -tdrive -d /dev/sda [16:28] (hd0) [16:29] <\sh> ah only root works ;) [16:30] <\sh> yes...that's a good fix [16:31] mikap: you will get a bug report for 3.4.5~rc1 and a fix soon [16:32] siggg (~jean@tra78-3-82-244-85-196.fbx.proxad.net) joined #fai. [16:43] siggg (~jean@tra78-3-82-244-85-196.fbx.proxad.net) left irc: Ping timeout: 480 seconds [16:44] This bug is now #604938. Including a patch [16:46] fai: 3 lange committed revision 6201 to 7 fai: do not install grub always into (hd0) (closes: #604938) [17:27] Mrfai (~lange@malzmuehle.informatik.uni-koeln.de) left irc: Quit: leaving [17:33] ErKa (~keryell@81.255.107.242) left irc: Ping timeout: 480 seconds [17:54] siggg (~jean@tra78-5-88-172-128-32.fbx.proxad.net) joined #fai. [18:05] <\sh> hmmm...ca. 6 minutes from 0 to 100 with fai os deployment + puppet + a huge tomcat java app ... very nice [21:12] justleen (~leen@77.62.130.32) joined #fai. [21:20] justleen (~leen@77.62.130.32) left irc: Read error: Connection reset by peer [22:02] \sh: cool [22:33] siggg (~jean@tra78-5-88-172-128-32.fbx.proxad.net) left irc: Ping timeout: 480 seconds [22:44] \sh (~shermann@nat.de.netviewer.com) left irc: Remote host closed the connection [00:00] --- Fri Nov 26 2010