[00:44] <ev> okay, so it's clearly failing inside update.d/99*ubiquity, which is a bit odd given how simple the code is, so I'm wondering if debconf is getting wedged earlier and this is just surfacing the error
[00:45] <ev> (|| logger -t ... the update.d scripts in update_partition clearly points the finger at 99ubiquity)
[01:53] <ev> GOT IT.
[01:53] <ev> finally
[02:00] <CarlFK> ev: does that mean you can take a peek at bug 728710 ? :)
[02:00] <ubot2> Launchpad bug 728710 in debian-installer "d-i apt-setup/local0/repository not working" [Undecided,New] https://launchpad.net/bugs/728710
[02:01] <ev> heh, not at 2am.  Remind me tomorrow?
[02:01] <CarlFK> works for me
[03:02] <CIA-31> partman-auto: evand * r601 partman-auto.ubuntu/ (automatically_partition/reuse/do_option debian/changelog):
[03:02] <CIA-31> partman-auto: * Fix a bug whereby reuse/do_option was filling debconf with garbage by not
[03:02] <CIA-31> partman-auto:  using the correct delimeter for read_paragraph.
[03:02] <CIA-31> partman-auto: * Refactor reuse/do_option so that it doesn't call blkid in an O(n^2)
[03:02] <CIA-31> partman-auto:  fashion.
[03:06] <CIA-31> partman-auto: evand * r602 ubuntu/debian/changelog: releasing version 93ubuntu11
[09:25] <cjwatson> ev: partman-auto> oh, awesome, does that fix psusi's bug 725408?  I got stuck on that yesterday ...
[09:25] <ubot2> Launchpad bug 725408 in partman-auto "installer hangs detecting existing partitions" [High,Triaged] https://launchpad.net/bugs/725408
[09:27] <cjwatson> ev: ... ignore that, that isn't touching the relevant code
=== ogra is now known as Guest67326
=== Guest67326 is now known as ogra_
[11:22] <ev> yikes, that one is quite perplexing
[11:54]  * ev stares harder
[11:57] <davmor2> ev: staring won't fix it honest hit it with a hammer and see what breaks
[11:57] <ev> haha
[12:03] <davmor2> ev: works for me :D
[12:05] <ev> I'm going to take it you mean the installer, and not smashing things with a hammer
[12:05] <ev> in which case my job here is done unity O
[12:05] <ev> until*
[12:09] <davmor2> Virtual hammer.  I only pound on things till they break.  You guys fix them :)
[12:23] <ev> lol
=== ogra_ is now known as ogra
[13:47] <CIA-31> ubiquity: cjwatson * r4588 trunk/debian/ (changelog rules):
[13:47] <CIA-31> ubiquity: Preprocess our variant of keyboard-configuration.postinst, so that it
[13:47] <CIA-31> ubiquity: includes the keyboard_present function.
=== JanC_ is now known as JanC
[15:14] <cjwatson> ev: have you seen bug 738366?
[15:14] <ubot2> Launchpad bug 738366 in ubiquity ""Erase Natty" wording DANGEROUSLY misleading" [High,Confirmed] https://launchpad.net/bugs/738366
[15:15] <cjwatson> looks like a bug rather than merely misleading wording ...
[15:16] <CIA-31> ubiquity: cjwatson * r4589 trunk/debian/ubiquity.templates: consistently end sentences with full stops
[15:16] <ev> if Ubuntu is the only operating system it finds on a disk, then it wipes the disk for the "Erase Natty and install 11.04"
[15:19] <cjwatson> the report says that Windows was installed
[15:20] <ev> indeed, I suspect os-prober isn't picking up windows (perhaps it's mounted), or the code is broken somewhere
[15:20] <CIA-31> ubiquity: cjwatson * r4590 trunk/debian/changelog: changes in automatic partitioning text fix LP: #736959
[15:20] <ev> the bigger problem is that I need to get the introductory string in that the specification mentions
[15:20] <cjwatson> being mounted shouldn't matter any more ...
[15:20] <ev> oh, right
[15:21] <ev> intro string> it wouldn't mention Windows, which would hopefully alert users to something being up
[15:23] <cjwatson> mm, I agree that is a problem although I'm not convinced it's a bigger problem :-)
[15:24] <ev> heh, perhaps that was poor phrasing on my part
[15:24] <ev> I've followed up on the bug
[16:40] <cjwatson> bah, the oem-config-debconf changes broke operation from the server CD
[16:40] <cjwatson> and I think probably generally just broke oem-config-debconf
[16:48] <CIA-31> ubiquity: cjwatson * r4591 trunk/debian/ (changelog oem-config-debconf.oem-config-debconf.upstart):
[16:48] <CIA-31> ubiquity: Start oem-config-debconf on stopping rc RUNLEVEL=[2345], rather than on
[16:48] <CIA-31> ubiquity: (filesystem and stopped udevtrigger); this guarantees that we will never
[16:48] <CIA-31> ubiquity: end up without a controlling terminal because tty1 has already started.
[16:50] <cjwatson> ev: IIRC you were scratching your head over bug 739891 or a similar bug; if it helps, it looks to me as though something, probably a maintainer script, has either written to stdout or read from stdin in an uncontrolled manner, causing the debconf protocol to get out of sync
[16:50] <ubot2> Launchpad bug 739891 in ubiquity "Ubiquity fails with: DebconfError: (10, "oem-config/enable doesn't exist")" [Medium,Confirmed] https://launchpad.net/bugs/739891
[16:51] <cjwatson> that might make sense of the weird man-db failure too
[16:53] <ev> hmm
[16:54] <ev> thanks
[16:54] <cjwatson> I suspect the nvidia packaging
[16:55] <cjwatson> hmm, that's silly, never mind
[16:57] <cjwatson> I've asked for debconf debugging output, which I suspect will make it clear
[17:00] <CIA-31> ubiquity: cjwatson * r4592 trunk/debian/ (changelog oem-config.oem-config.upstart):
[17:00] <CIA-31> ubiquity: Don't run the oem-config job in recovery mode (thanks, Shih-Yuan Lee;
[17:00] <CIA-31> ubiquity: LP: #734695).
[17:17] <CIA-31> ubiquity: cjwatson * r4593 trunk/ (debian/changelog ubiquity/frontend/debconf_ui.py): Run plugininstall in debconf frontend (LP: #740183).
[17:28] <CarlFK> ev:  bug 728710 - I am around for a few hours, happy to try things
[17:28] <ubot2> Launchpad bug 728710 in debian-installer "d-i apt-setup/local0/repository not working" [Undecided,New] https://launchpad.net/bugs/728710
[19:26] <cjwatson> ev: yay, wubi booted successfully this time
[19:26] <cjwatson> confirming that it gets all the way through an installation ...
[20:54] <bdmurray> ev: I'd like to have a bug day for ubiquity this week and specifically focus on recreating bugs in the New state and help to move them to Confirmed.  Does that seem alright with you?
[20:56] <cjwatson> FWIW, the easiest bugs for at least me to fix are those that come with a recipe for recreating them in kvm - I don't know about ev
[20:56] <cjwatson> not that others are invalid, but ...
[20:58] <bdmurray> At one point in time I tried to create a multi-partition disk image for kvm to recreate a bug but didn't get very far.  Do you have any tips for that?
[20:58] <cjwatson> what went wrong?
[20:58] <cjwatson> that sort of thing generally just works for me
[20:59] <cjwatson> but I have some set habits that perhaps are helping
[21:01] <cjwatson> (got to go out, back later)
[21:03] <bdmurray> I believe I was looking at bug 659106 at the time
[21:03] <ubot2> Launchpad bug 659106 in ubiquity "Maverick installer lost Windows partitions (dup-of: 655950)" [Critical,Confirmed] https://launchpad.net/bugs/659106
[21:03] <ubot2> Launchpad bug 655950 in ubiquity "ubiquity confusion w/ auto-resize option results in OS and data loss" [High,Fix released] https://launchpad.net/bugs/655950
[21:03] <bdmurray> so was trying to setup windows on a parition in the image