[08:48] <cjwatson> evand: looking at https://beta.launchpad.net/ubuntu/+source/ubiquity/+bug/93197 - shouldn't 'set_os_type $line' be 'set_os_type "$line"'? $line might well contain spaces and needs to be quoted
[08:49] <cjwatson> evand: also seems like it wouldn't hurt to have set_os_type return 1 if it's an unknown OS type and then do 'set_os_type "$line" || continue' or similar in calling code
[10:45] <CIA-15> oem-config: cjwatson * r275 oem-config/ (d-i/manifest debian/changelog): * Automatic update of included source packages: console-setup 1.13ubuntu9.
[10:52] <CIA-15> oem-config: cjwatson * r276 oem-config/debian/changelog: releasing version 1.12
[11:47] <CIA-15> ubiquity: cjwatson * r1967 ubiquity/ (configure configure.ac): bump to 1.4.3
[11:59] <saispo> hi
[12:00] <saispo> cjwatson: does ubuntu installer support pgp key with passphrase ?
[12:02] <cjwatson> should be irrelevant, the installer isn't signing anything, only verifying
[12:03] <cjwatson> verification doesn't need passphrase input, unless you're doing something truly weird
[12:03] <saispo> oki, thanks
[12:59] <evand> cjwatson: hrmm, I'll take a look at it after classes.  Thanks!
[01:51] <cr3> after installing in OEM mode, would there be a way to preseed questions asked to the enduser?
[01:51] <cr3> for example, if an OEM is selling machines in a country which only has one language, there would be little reason to prompt the user for language
[01:58] <cjwatson> cr3: not at present. Haven't we had this conversation before?
[02:00] <cjwatson> yes, we have, last Tuesday
[02:00] <cr3> cjwatson: wow, I need to keep better logs of my conversations
[02:02] <cr3> cjwatson: aha! I remember that the outcome of the conversation was that the OEM team should formalize their request to make it happen. I'll write them an email right away
[02:02] <cjwatson> right
[02:47] <CIA-15> ubiquity: cjwatson * r1968 ubiquity/ (debian/changelog scripts/install.py):
[02:47] <CIA-15> ubiquity: * Install dummy policy-rc.d and start-stop-daemon while doing anything
[02:47] <CIA-15> ubiquity:  that might install or remove packages in the target system, so that we
[02:47] <CIA-15> ubiquity:  don't try to run init scripts (LP: #89463).
[02:54] <CIA-15> ubiquity: cjwatson * r1965 feisty-beta/ (configure configure.ac debian/changelog scripts/install.py): merge r1967 and r1968 from mainline
[04:15] <thom> cjwatson: hey, is there any way to perform an action after the reboot into the new install?
[04:17] <cjwatson> write out an init script
[04:20] <thom> hrm. i only want to do this once; but yeah, that works
[04:36] <CIA-15> ubiquity: cjwatson * r1969 ubiquity/ (d-i/manifest debian/changelog): merge from feisty-beta branch
[04:51] <CIA-15> ubiquity: cjwatson * r1970 ubiquity/ (configure configure.ac): bump to 1.4.4
[04:51] <CIA-15> ubiquity: cjwatson * r1967 feisty-beta/debian/changelog: releasing version 1.4.3
[05:31] <evand> cjwatson: whoops, noted your comments on 94336 for future reference.
[05:32] <cjwatson> np
[08:13] <cr3> how can I debug the desktop installation? I'm attempting to install from nfs and the nfs mounted squashfs seems to start failing at "Starting GNOME Display Manager...", so I'd like to get some more details.
[08:16] <cjwatson> that's not the installer at that point so best not asked here
[08:16] <cjwatson> if squashfs is dying you would probably be best asking the kernel team for help
[08:16] <cjwatson> seeing as we employ the squashfs author and all that
[08:18] <cr3> cjwatson: the problem is most likely that the gnome display manager is affecting the network which where the squashfs is being nfsmounted. so, I made sure /etc/network/interfaces only contained lo information but that didn't seem to help.
[08:18] <cr3> I'll see if I can enable xtrace in the gnome display manager init script
[08:19] <cjwatson> that seems an odd inference and would surprise me if true
[08:19] <cjwatson> but at any rate, please not here
[08:26] <cr3> something installer related then :) I managed to get the nfs part of the installer working with a couple minor patches wihch I submitted to the right people, thought you might like to know :)
[08:27] <cjwatson> could you point me to those patches?
[08:27] <cjwatson> I would have expected it to need new components
[08:28] <cr3> cjwatson: they were sent directly to lifeless and tfheen, upon lifeless' request. I'll forward the message to you if you like
[08:28] <cjwatson> er - that's not the installer
[08:29] <cjwatson> casper is not the installer, it's the live CD bootstrap process
[08:29] <cjwatson> not bothered about it, I thought you meant d-i :)
[08:29] <cr3> ah, my mistake, thanks for pointing that out: booting != installer :)
[08:30] <cjwatson> the installer begins when you (a) boot d-i or (b) select the "Install" icon on the desktop. :)
[08:35] <CIA-15> ubiquity: cjwatson * r1971 ubiquity/ (debian/changelog ubiquity/frontend/kde-ui.py):
[08:35] <CIA-15> ubiquity: * KDE frontend: Disable subsidiary autopartitioning options according to
[08:35] <CIA-15> ubiquity:  the primary selection (thanks to Thorbjrn Tux for prompting me with
[08:35] <CIA-15> ubiquity:  suggestions; LP: #85980).
[11:07] <stephanecharette> Is this where I would ask about some installation problems?
[11:22] <cjwatson> ... yes