[06:24] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [06:32] Nick change: \sh_away -> \sh [07:17] Nick change: MT -> Guest41 [07:17] Guest41 (~MT@dove.informatik.tu-muenchen.de) left irc: Read error: Connection reset by peer [07:17] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [07:41] ErKa (keryell@m5.wifi.enstb.org) joined #fai. [08:10] Mrfai (~lange@kueppers.informatik.uni-koeln.de) joined #fai. [08:50] moin [09:29] meandtheshell (~markus@85.127.114.140) joined #fai. [09:45] meandtheshell (~markus@85.127.114.140) left irc: Quit: Leaving. [09:58] ErKa (keryell@m5.wifi.enstb.org) left irc: Ping timeout: 480 seconds [10:10] Mrfai (~lange@kueppers.informatik.uni-koeln.de) left irc: Quit: leaving [10:11] Mrfai (~lange@suenner.informatik.uni-koeln.de) joined #fai. [10:51] mikap, do you have virtualbox 1.5 running with usb-support? [10:51] i read your blog and used strace, but i can find "usb" in the strace... [10:56] mikap, in the opensource editition or the propietary [10:56] ? [10:58] ErKa (keryell@m5.wifi.enstb.org) joined #fai. [11:09] h01ger: I'm using the version from virtualbox.org (so not the Debian packages from the debian pool) [11:09] h01ger: and yes, IIRC usb worked (though not booting from USB IIRC) [11:10] though would have to check that again to verify it ;) don't use the usb stuff inside virtualbox because what I would need is booting from usb [11:10] Action: h01ger too [11:11] if i run "vboxmanage list usbhost" i get errors... [11:20] Nick change: MT -> Guest48 [11:20] Guest48 (~MT@dove.informatik.tu-muenchen.de) left irc: Read error: Connection reset by peer [11:20] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [11:38] h01ger: ah ok; I'll check it if I have access to my system again [11:43] i-R2D2 (~deed@tor-irc.dnsbl.oftc.net) left irc: Remote host closed the connection [12:03] i-R2D2 (~deed@tor-irc.dnsbl.oftc.net) joined #fai. [12:10] Mrfai: I've answered you nfs-common question on bts. [12:11] to fai debian users: is the chroot size in debian 330 MB for a unmodifed fai-* installaiton as given in the doc? Du -sh in gutsy gives 383 MB [12:12] Action: h01ger blames python [12:12] u.a. [12:12] Action: allee kicks python [12:13] Action: allee ponders if it's worth fixing the size in the docs for (k)ubuntu [12:14] no it's not worth it. I may change the guide and say this is the size for a Debian nfsroot [12:15] Mrfai: right, maybe a README.ubuntu listing the small diff debian<->ubuntu is the way to go (if I'll find more items) [12:16] indeed [12:17] Mrfai: ah, one more: License use often old FSF address '...Temple Place...'. They moved to Franklin Street [12:19] Hits for Franklin 1, for Temple 25 [12:20] hi siretart [12:20] hi allee [12:21] btw. I've now a package that boots demohosts (still errors). (== Head in uploaded trunk-ubuntu) [12:21] allee: please submit bug reports for whose little changes, otherwise I will forget to change those [12:21] Action: h01ger points Mrfai to linda+lintian. the ones in unstable [12:22] Action: siretart looks at http://codebrowse.launchpad.net/~fai/fai/trunk-ubuntu/changes [12:23] ahh. that's how I expected it to be. Mrfai should be able to easily merge patches like http://codebrowse.launchpad.net/~fai/fai/trunk-ubuntu/bundle/allee%40kubuntu.org-20070925221030-fi57h7dy4yr6816w/bundle.txt :) [12:26] allee: http://codebrowse.launchpad.net/~fai/fai/trunk-ubuntu/revision/allee%40kubuntu.org-20070926000234-tmemfln2qjs2h66a?start_revid=allee%40kubuntu.org-20070926002108-fa766hanm0gac36j#bin/fai is a very good catch! I wondered why your expr didn't work for me [12:48] Nick change: \sh -> \sh_away [12:58] siretart: :) [12:58] Action: allee fires up reportbug -bts debian [12:59] allee: If you don't have more changes in your pipe, feel free to upload your package to our PPA. do you know how to do it? [13:02] siretart: I once read about it. and wasn't there a ppaput or so? I'll find it again I hope ;) [13:03] allee: ppaput is for reviewing purposes. you don't need that for this [13:03] allee: http://paste.debian.net/38068 <- add this to your ~/.dput.cf [13:04] when uploading, double check the Distribution: line of your changes file [13:04] in my last uploads, I batch uploaded all backports in the same upload. They mustn't contain the orig.tar.gz anyway. [13:05] well, not in the same upload, but without waiting for the ACCEPTED mail from the first upload [13:05] siretart: oh, no .orig.tar.gz. So like in debian archive? only once for new upstream tarball, then only diffs? [13:06] allee: like in the ubuntu archive. the orig.tar.gz must be contained in the first upload only [13:06] they take tarball from the distro archive? Cool (and dialup friendly) [13:07] but I already uploaded the fai-3.2.1 tarball [13:08] Action: allee feeds debian bts, then cares about the ppa pkg [13:34] Nick change: \sh_away -> \sh [13:37] rcy (~rcy@DSL-209.90.181.211.Vancouver.primus.ca) left irc: Ping timeout: 480 seconds [13:38] Lin (~igor@200.179.57.57) left irc: Quit: Ex-Chat [13:47] Mrfai: the nolocking bug is that the $romountopts or what the name is, isn't exportet from fai. [13:48] because the get-config-dir-nfs script is a script and not a function it doesn't get the $romountopt variable. [13:48] romountopt is the name [13:49] line 35 of fai [13:49] and that one should have a -n in it because mtab is a symlink to /proc/mounts [13:51] glance: yep, romountopt will be exported in the next fai version. I will also add -n [13:52] I just added thos changes to my fai version, but did not commit it yet. [13:55] check [15:11] meandtheshell (~markus@85.127.114.140) joined #fai. [15:39] Nick change: \sh -> \sh_away [15:40] meandtheshell (~markus@85.127.114.140) left irc: Quit: Leaving. [15:53] Mrfai (~lange@suenner.informatik.uni-koeln.de) left irc: Quit: leaving [16:11] ErKa (keryell@m5.wifi.enstb.org) left irc: Ping timeout: 480 seconds [16:34] ErKa (keryell@keryell.pck.nerim.net) joined #fai. [16:46] Mrfai (~lange@xdsl-213-196-242-194.netcologne.de) joined #fai. [17:12] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds [17:31] i-R2D2 (~deed@tor-irc.dnsbl.oftc.net) left irc: Remote host closed the connection [17:32] i-R2D2 (~deed@tor-irc.dnsbl.oftc.net) joined #fai. [17:38] meandtheshel1 (~markus@85.127.114.140) joined #fai. [17:58] MT (~MT@ppp-82-135-93-79.dynamic.mnet-online.de) joined #fai. [18:01] meandtheshel1 (~markus@85.127.114.140) left irc: Quit: Leaving. [18:28] ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds [18:33] Lin (~igor@200.179.57.57) joined #fai. [18:33] Lin (~igor@200.179.57.57) left irc: Read error: Connection reset by peer [18:37] mhm, grub is installed after memtest86 and memtest (update-grub in postinst) complains /boot/grub not found. Same on debian? [18:44] MT (~MT@ppp-82-135-93-79.dynamic.mnet-online.de) left irc: Quit: MT [18:55] allee: paste you log file to paste.debian.net [19:05] Bluemoon (~mj@www.bfr.bund.de) left irc: Read error: Connection reset by peer [19:05] Bluemoon (~mj@www.bfr.bund.de) joined #fai. [19:08] Mrfai: until my fai install finished again ... [19:08] grub/postinst: use /var/log/fai/disk_var.sh but when I reboot it's in /var/lib/fai/ [19:09] and why $target if it's designed to work only with softinst ;) [19:14] Bluemoon (~mj@www.bfr.bund.de) left irc: Read error: Connection reset by peer [19:18] Bluemoon (~mj@www.bfr.bund.de) joined #fai. [19:27] Mrfai: software.log http://paste.debian.net/38082 Searching for memtest8|grub shows the ordering problem [19:31] ok, this is really a memtest86+ problem. If the ubuntu version of memtest86 need grub, it must have a depends on it. [19:33] maybe related to #129614 in LP [19:33] ubuntu changes https://edge.launchpad.net/ubuntu/+source/memtest86+ [19:35] siretart: I fail to see the related change/fix :( [19:37] Mrfai: yeap that's the bug [19:38] rcy (~rcy@209.90.184.166) joined #fai. [19:38] In debian etch in memtest86+.postinst update-grub is not called. This is done in the ubuntu version. [19:39] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [19:39] siretart: your opinion where I should keep a todo/notes lists. FAIwiki. TODO.ubunt in source package somewere in lauchpad? [19:40] how about lp bugs? [19:40] so we can make comments via email [19:41] I think the faiwiki is a good place to keep ubuntu related todo's and notes [19:41] *nod* [19:43] for really small notes/todo lists etc this is perhaps the easiest [19:44] Mrfai: any preference how to name ubuntu centric pages? UbuntuWhatever? or ... [19:45] FAIonUbuntu? [19:45] FAI ist natuerlich ueberfluessig im faiwiki [19:46] RunningOnUbuntu? [19:46] UbuntuTeam/Todo [19:46] UbuntuTeam/Minutes [19:46] ... [19:47] ah, I like UbuntuTeam prefix [19:47] good idea [19:47] for now I will simple remove memtest86 and add to UbuntuTeam/Todo [19:55] memtest86+ on debian has a longer test in postin before calling update-grub: [19:55] if test -e /boot/grub/grub.cfg && which update-grub [19:57] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds [19:57] seems that ubuntu is just too old. Debian bug #425221 was fixed in 1.70-3. So ubuntu should update to debians version [20:04] Mrfai: interesting. Only system installed from kubuntu lifecd there is no /boot/grub/grub.cfg and my sys uses for sure grub [20:07] seems that grub.cfg is new. I also didn't knew it before. [20:07] so grub too old too :( And I had the impression using ubuntu mean living on the edge ;) [20:19] Mrfai: uhm, problem with wiki confirmation e-mails? received the mail. clicked the link -> invalid confirmation code (valid until 3-Okt was in email) [20:20] what is too old? grub? [20:21] grub | 0.97-29 | unstable | source, amd64, hurd-i386, i386 [20:21] grub | 0.97-29ubuntu4 | gutsy | source, amd64, i386 [20:22] siretart: ah, you know by accident why there no grub.cfg in gutsy? debian test for it in memtest86+ postinst. Nevertheless I've non here. [20:23] err, what's grub.cfg? where is it supposed to be? [20:23] [21:56] if test -e /boot/grub/grub.cfg && which update-grub [20:24] I don't have a /boot/grub/grub.cfg either [20:24] is that an alternative name for /boot/grub/menu.lst? [20:25] siretart: is completely new for me too [20:26] http://patches.ubuntu.com/by-release/atomic/debian/m/memtest86+/memtest86+_1.70-3.patch [20:26] hmmm [20:26] maybe used in grub2? I don't know. [20:26] But the newer memtest package in debian seems to fix the bug [20:27] wll, since thats no new upstream version, lets just fixup that merge here: http://merges.ubuntu.com/m/memtest86+/ [20:28] siretart: how to fixup. As gutsy has no grub.cfg what's the right test? [20:30] Lin (~igor@200.179.57.57) joined #fai. [20:31] there seem to be some bashism reported in LP [20:32] hmm [20:32] okay. /boot/grub in not in grub package or any other. Postinst of grub and memtest use simple update-grub and update-grub abort when there's no boot/grub dir [20:33] eh, /boot/grup dir should not appear from nowhere [20:34] mmh, tired now [20:34] allee: if you want to finish the merge today, feel free. If not, I'll finish it tomorrow [20:34] good night, folks! [20:39] Lin (~igor@200.179.57.57) left irc: Quit: Ex-Chat [20:40] siretart: I stick with testing fai for the rest of the night [20:40] siretart: sleep well [20:46] siretart: what's the best way to install gutsy (on real hardware, for now)? [20:47] are there cd's yet? [20:47] hmm, too late, I guess... [20:48] ah, cdimage.ubuntu.cim is the answer... [20:54] ErKa (keryell@keryell.pck.nerim.net) joined #fai. [21:56] MT (~MT@dove.informatik.tu-muenchen.de) joined #fai. [22:00] mikap? [22:01] mikap, I'll respond to your request on FAI_CONFIG_SRC via the BTS... [22:04] I think we need to work on the FAI developers workshop on catching error codes during make-fai-nfsroot and dirinstall and other parts [22:06] Mrfai, I was talking about #444213 [22:06] fai-client: support for softupdate without need for FAI_CONFIG_SRC [22:07] IMHO this report is simply invalid [22:11] Action: lazyb0y just replied to this [22:11] does CONFIG_SRC already understand file:// urls? [22:12] yep get-config-dir-file exists [22:13] jap [22:13] though I'm not sure whether it's documented... [22:14] hmm, at least not in the fai-guide [22:14] but the comment in fai.conf contains it [22:15] # how to access the fai config space [22:15] # default if undefined here: nfs://`hostname`/$FAI_CONFIGDIR [22:15] # supported URL-types: nfs, file, cvs, cvs+ssh, svn+file, svn+http, [22:15] # git, git+http, [22:15] #FAI_CONFIG_SRC=nfs://yourservername$FAI_CONFIGDIR [22:22] MT: ok, feel free :) [22:22] feel free ... to do what? [22:23] if it's about the comment to your bug report - I already commented on it, you should have received an email already [22:23] MT: regarding "I'll respond to your request on FAI_CONFIG_SRC via the BTS..." I meant, sorry - read your mail right now [22:24] if FAI_CONFIG_SRC is unset: where is the config stuff of fai dirinstall by default then? [22:25] h01ger: VBoxManage list usbhost > vbox_usb.txt => http://grml.org/tmp/vbox_usb.txt [22:26] I think FAI_CONFIG_SRC must be set in any case [22:26] h01ger: you know the http://michael-prokop.at/blog/2007/07/11/virtualbox-usb/ problem, right? [22:26] MT: I definitely haven't set it :) [22:26] but what's so wrong about setting it to a sane value? [22:26] MT: that's why I never understood why I should have to set it at all ;) [22:26] mikap, yes + thx. thats with the non-ose version, right? [22:26] MT: cause I don't know what it should be for ;) [22:27] Action: h01ger made himself a customized d-i cd, because usb-booting doesnt work anyway :/ [22:27] h01ger: yepp, 1.4.0 [22:27] a "sane value" cannot be guessed easily [22:27] normally FAI_CONFIG_SRC will be appended to /srv/fai/nfsroot/etc/fai/fai.conf which is then read by the install clients. [22:27] _you_ will definitely know a sane value [22:28] MT: it would be OK for me to set it to a "sane value" if I don't have to deal with it any further; though I don't know yet what would be the benefit (instead of just not setting it at all) and what's the easiest solution [22:28] % grep -r FAI_CONFIG_SRC /etc/grml/fai [22:28] /etc/grml/fai/fai.conf:#FAI_CONFIG_SRC=nfs://yourservername$FAI_CONFIGDIR [22:29] and I'm invoking 'fai -C/etc/grml/fai ...' [22:30] well, then copy your config space from the NFS server to some directory [22:30] MT: NFS-server? I don't have one [22:30] that line is not activated (see '^#...') [22:30] it's just the default from simple-example of FAI, never activated it [22:30] well, but somehow you got your chroot dir-installed, didn't you? [22:31] MT: sure, running fai dirinstall ;) [22:31] well, using which config? [22:31] which config do you mean? :) [22:31] sorry, I really don't know what FAI_CONFIG_SRC is for, never used that for dirinstall ;) [22:31] running fai dirinstall also requires a config space [22:32] I definitely didn't define a config space [22:32] hmm, Mrfai, was it the the simple example he used? [22:32] what's supposed to be stored inside this config space? [22:32] mikap: the default is used - /srv/fai/config [22:32] lazyb0y: I don't have such a directory [22:33] hmm [22:33] strange, then [22:33] but somewhere, you must have it [22:33] exaclty that's why I'm so such confused :) softupdate told me to use a config space which I never needed so far ;) [22:33] where do you define your package lists, etc? [22:34] FAI_CONFIG_SRC is for the isntall client. They must know how/where to access the config space. [22:34] ls /etc/grml/fai/config [22:34] class/ debconf/ hooks/ package_config/ scripts/ [22:34] ah! [22:34] config space is not the same a config dir that is specified with fai -C [22:34] well, there it is ... [22:34] /etc/grml/fai/fai.conf:FAI_CONFIGDIR=/etc/grml/fai/config [22:34] => but that's not FAI_CONFIG_SRC?! [22:35] FAI_CONFIG_SRC="file:///etc/grml/fai/config" [22:35] mikap: Mrfai: it is the same from the contents [22:35] this is obviously a bit confusing [22:35] so I should just give it a try with FAI_CONFIG_SRC="file:///etc/grml/fai/config"? [22:35] yep [22:35] yep [22:35] great :) [22:35] I will try that tomorrow, bed is calling ;) [22:35] mikap: if you copied that into the chroot, yes [22:36] lazyb0y: not manually at least ;) [22:36] mikap, please change your bugtitle [22:36] to [22:36] otherwise, as I wrote in the bts, bind.mount it into the chroot [22:36] documentation of FAI_CONFIG_SRC is incomplete and not clear enough [22:36] MT: feel free to adjust it to whatever is more appropriate [22:36] well, it's your bug report :-) [22:36] mikap: I really like to see you at the fai developers workhop. Will you try to attend? [22:36] mikap: I'd guess dirinstall doesn't copy the configspace into the chroot, but i might be wrong [22:37] as soon as I checked out whether it does what it need and if I understand the code I will provide a patch for the docs [22:37] thx!! [22:37] Mrfai: that's the one on 8/9. november, right? [22:37] hmm, but it copies /etc/fai, and as you "accidentally" have the configspace in /etc/fai, it might be there [22:37] meandtheshell (~markus@85.127.114.140) joined #fai. [22:37] _normally_ it is not automatically there, e.g. when it's at the default location /srv/fai/config [22:38] guys, I'll leave for bed ... [22:38] Action: MT says good night! [22:38] good idea :) [22:39] good n8! :) [22:39] mikap: 9-11 november [22:40] Action: Mrfai bed is calling too [22:41] Mrfai: right, a pretty cheap flight would be available for Graz=>Köln and back, but IIRC just on 8th and 12th :-/ so I would need a place where I could stay in that time (assuming that I manage it to have time) [22:42] mikap: don't stay at Mrfai's place - he has dangerous cats! [22:42] ok, first let's see if we get enough money to held the workshop. [22:42] lazyb0y: *gggg* [22:42] Mrfai: ok [22:42] Action: mikap is back tomorrow, n8 :) [22:43] Mrfai (~lange@xdsl-213-196-242-194.netcologne.de) left irc: Quit: leaving [22:46] MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds [22:51] ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds [23:12] Action: lazyb0y just installed gutsy and tried fai-setup... error log follows :( [23:33] lazyb0y: pastebin. I've almost ready to upload gutsy ppa package, with contains several fixes: https://code.launchpad.net/~fai/fai/trunk-ubuntu [23:34] lazyb0y: im rebuiding nfsroot, then a pxe demohost install. and with luck then an upload to ppa [00:00] --- Thu Sep 27 2007