[04:27] <evand> good evening
[04:30] <LaserJock> evening
[10:14] <cjwatson> morning
[10:14] <LaserJock> hi
[10:14] <zakame> hi cjwatson , just read your mail on d-i
[10:15] <cjwatson> hm, not immensely useful for me to be told of new subscribers to ubuntu-installer; /me turns that off
[01:06] <heno> cjwatson: when someone reports a bug in 6.06 and then tries it on 6.06.1 and finds it fixed do we simply mark is as 'Fixed' without fully understanding the bug? (no logs forthcoming) eg bug 48117
[01:30] <cjwatson> heno: might as well mark that fix-released; freezes are typically kernel bugs or bad CDs anyway, and we won't be able to diagnose it any further
[01:32] <heno> ok, thx
[02:03] <allmanj> cjwatson: hey - still having difficulty running a command after users are created on a dapper install cd. I created a file in /usr/lib/finish-install.d but it doesn't appear to be run...
[02:05] <cjwatson> allmanj: in dapper, you need to use /usr/lib/prebaseconfig.d
[02:05] <cjwatson> /usr/lib/finish-install.d was introduced in edgy
[02:06] <neutrinomass> Not sure if this is the right place but I just want to double check that I didn't wrongly reject this bug : https://bugs.launchpad.net/distros/ubuntu/+source/ubiquity/+bug/61598
[02:06] <allmanj> cjwatson, thanks! hard to find that...
[02:07] <cjwatson> neutrinomass: this is the right place. Your rejection seems fair to me.
[02:07] <neutrinomass> cjwatson: Ok, thanks :-)
[02:08] <cjwatson> neutrinomass: it may not *necessarily* be a dead hard disk, though; the ubiquity/install/copying_error/hd_fault template in debian/ubiquity.templates in the ubiquity source package is roughly what I usually use for rejection messages in that case
[02:08] <cjwatson> there might be other reasons other than temperature as well; if you encounter any, do let me know ...
[02:09] <cjwatson> hmm, I wonder if there's any way that I could arrange for bzr pushes of installer branches to the supermirror to have their commit messages logged here
[02:10] <cjwatson> https://launchpad.net/products/bzr-cia looks interesting
[02:11] <neutrinomass> ahh, I didn't know that ubiquity printed such errors nowadays ... neat :)
[02:11] <cjwatson> yeah, I did that in edgy to try to reduce the number of bugs we got due to broken disks of various kinds
[02:12] <cjwatson> and in general it seems like a good idea to try to print saner crash messages where possible, if the crash is unavoidable ...
[02:13] <neutrinomass> I take it that nothing similar can be done for faulty RAM, right ? (I rejected such a bug today thinking it's impossible)
[02:13] <cjwatson> it's pretty hard since the bug could show up just about anywhere
[02:13] <cjwatson> which bug was that?
[02:15] <neutrinomass> just a sec, I lost it
[02:18] <neutrinomass> https://launchpad.net/bugs/70441
[02:22] <cjwatson> no, I think you're right there
[02:22] <cjwatson> ok, hopefully we'll have the CIA bot here soonish; I've asked for it to track ubiquity and oem-config commits for now
[02:24] <Riddell> crivvens, a new channel for me
[02:25] <cjwatson> hello
[02:25] <Riddell> cjwatson: a kubuntu live session should be enough (it just needs python-qt4) but there's a bug in python-qt4 where the user interface compiler only works with python2.5 so it may need python2.5 installed until I fix the package (top of my todo list)
[02:25] <cjwatson> I can do that
[02:26] <cjwatson> if my test machine hadn't decided that DISPLAYING ANYTHING was for wimps
[02:26] <cjwatson> ah, there it is
[02:27] <Riddell> cjwatson: I didn't look at the xembed stuff yet, so the qtparted window pops up externally for now
[02:27] <cjwatson> ok, hopefully I'll manage to make that go away anyway
[02:27] <cjwatson> booting kubuntu now, will take a while to build stuff etc.
[02:52] <cjwatson> http://cia.navi.cx/stats/author/cjwatson <- excellent, ubiquity commits
[02:53] <allmanj> i've created a script in there with a late_command and the permissions are good (i intentionally put in a crappy command at the end so the system wouldn't reboot)
[02:53] <allmanj> the script creates a directory for a user created earlier and copies a file from the cd into it
[02:53] <allmanj> but it doesn't appear to be run...
[02:53] <cjwatson> what's the script name?
[02:54] <allmanj> 01ssh_keys?
[02:54] <cjwatson> should be ok. is there an "info: Running /usr/lib/prebaseconfig.d/01ssh_keys" or "error: Unable to execute /usr/lib/prebaseconfig.d/01ssh_keys" message in syslog?
[02:55] <cjwatson> oh, hang on, you're creating the script with a late_command?
[02:55] <allmanj> yes?
[02:55] <allmanj> i need it to be run after the user has been created...
[02:55] <cjwatson> allmanj: that won't work, you need to create it with an early_command; by the time late_command is run, prebaseconfig has already decided what scripts in /usr/lib/prebaseconfig.d to execute
[02:55] <cjwatson> late_command being run from a prebaseconfig script itself
[02:56] <allmanj> ... oh. d'oh!
[02:56] <allmanj> thanks
[02:56] <cjwatson> allmanj: you'll need to set the number appropriately
[02:56] <cjwatson> allmanj: I suggest 07 rather than 01
[02:56] <cjwatson> user-setup is run at 06
[02:57] <allmanj> ah groovy - much thanks. That definitely would have got me!
[02:57] <cjwatson> sometimes I really wish I could update the dapper installer in non-trivial ways :) gets very confusing
[02:57] <cjwatson> but then I consider the pain that would cause, in turn ...
[02:58] <allmanj> aye. it's just a shame that just about all of the documentation is for edgy
[03:00] <cjwatson> there's the installation-guide package in dapper?
[03:00] <cjwatson> installation-guide-i386
[03:00] <cjwatson> sorry if I've mentioned it to you before
[03:00] <cjwatson> it does have a few errors though - mentions base-config and such
[03:02] <cjwatson> Riddell: "pyuic: File generated with too recent version of Qt Designer (4.0 vs. 3.3.7)"
[03:11] <allmanj> that did the job there anyway - thanks a mil
[03:12] <cjwatson> good
[03:15] <Riddell> cjwatson: what says that?
[03:15] <Riddell> it may help to do:  cp -r /usr/lib/python2.5/site-packages/PyQt4/uic/ /usr/lib/python2.4/site-packages/PyQt4/uic
[03:15] <Riddell> cjwatson: you don't need to copmile the .ui file any more, it does that at run time
[03:15] <Riddell> compile
[03:23] <cjwatson> Riddell: kdepyuic; you left that call in debian/rules
[03:40] <jerom1> Hi all
[03:41] <jerom1> When i add theses packages in my kickstart linux-headers-server, linux-image-server, linux-source my initrd is failed after reboot. Have you an idea ?
[03:41] <cjwatson> could you please give more detail, e.g. exact error messages?
[03:42] <jerom1> unmoutable device ... kernel panic after reboot
[03:42] <jerom1> and my grub conf is ok
[03:42] <cjwatson> check that the initramfs is being generated for the correct kernel flavour
[03:42] <cjwatson> #ubuntu-kernel will probably be better at this
[03:42] <jerom1> oki thanks
[03:43] <cjwatson> mind you, you shouldn't be installing linux-image-server in the %packages section in kickstart
[03:44] <cjwatson> jerom1: to use the server kernel, 'preseed base-installer/kernel/override-image string linux-image-server' (or linux-server to include restricted modules as well) is better
[03:44] <cjwatson> if you installed it in pkgsel it's entirely possible that the initramfs just wasn't generated
[03:45] <Riddell> cjwatson: I've not been making .debs out of it, only running it manually
[03:47] <cjwatson> my normal testing method is to use debuild to build a package; makes sure it keeps working ...
[05:14] <Riddell> packaging updated
[05:14] <cjwatson> note I just did some packaging work on my side too
[05:14] <cjwatson> to make desktop, pixmaps, po subdirs be built even if UBIQUITY_NO_GTK=1
[05:33] <jerom1> how to install kernel image server with kickstart please ?
[05:34] <jerom1> i install generic, but i cannot install server, how to install this please ?
[05:48] <cjwatson> 14:44 < cjwatson> jerom1: to use the server kernel, 'preseed base-installer/kernel/override-image string linux-image-server' (or linux-server to include restricted modules as well) is better
[05:48] <cjwatson> requires edgy
[05:48] <cjwatson> oh, no, I tell a lie
[05:48] <cjwatson> I implemented base-installer/kernel/override-image in dapper
[05:49] <cjwatson> so just put the above in the main section of your kickstart file
[05:57] <cjwatson> Riddell: do we still need to build-dep on python-kde3-dev?
[05:57] <cjwatson> since kdepyuic isn't called in the build process any more
[05:59] <Riddell> cjwatson: no
[06:00] <cjwatson> excellent
[06:14] <jerom1> cjwatson> thanks i test tomorrow, bye
[08:44] <stgraber> hi