[00:08] <CIA-3> ubiquity: superm1 * r3834 ubiquity/ (d-i/manifest debian/changelog):
[00:08] <CIA-3> ubiquity: Automatic update of included source packages: base-installer
[00:08] <CIA-3> ubiquity: 1.103ubuntu4.
[00:08] <CIA-3> ubiquity: superm1 * r3835 ubiquity/debian/changelog: releasing version 2.1.25
[07:46] <ara> ev, https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/526581
[07:46] <ara> ev, I assigned temporarily to cjwatson, but you might be a better person to have a look to it
[07:46] <ara> ev, morning, by the way :)
[09:05] <ev> ara: ah, looking into it now.  Thanks, and good morning :)
[09:36] <ev> superm1: thanks for the get_mod_bool fix.  That was my intention, but ended up as a stupid typo on my part.
[10:12] <cjwatson> ev: so, what did I do wrong in migrationassistant? I declared that it was supposed to run after partman, but that bug looks for all the world as though parted_server isn't running ...
[10:14] <cjwatson> (and there's a minor issue with the PartedServer instance going out of scope and having its destructor run *after* dropping privileges, but I'm not sure I care
[10:14] <cjwatson> )
[10:14] <cjwatson> I see a slideshow parse error in that debug log too, which is a bit worrying
[10:25] <ev> indeed, it's after partman in the actual order, so order_plugins isn't broken (though it looks like it's getting placed after the summary).
[10:38] <cjwatson> ew, that would explain it
[10:38] <cjwatson> so we need something to force summary to go last except for install?
[10:38] <cjwatson> or something
[10:42] <thomas> ev, cjwatson: what copyright license is ubiquity released under? there is no COPYING file.
[10:43] <persia> thomas: debian/copyright has the interesting bits, with d-i sources having their own debian/copyrights
[10:45] <thomas> persia: does that apply to both code and images?
[10:46] <ogra> images ?
[10:46] <persia> thomas: It is intended to be comprehensive, but I'm not authoritative on the matter.
[10:47] <ogra> thomas, what kind of images do you refer to ? slideshow graphics ?
[10:49] <cjwatson> debian/copyright is authoritative
[10:49] <ev> cjwatson: I'd have to dig in the code to be sure, but surely BEFORE = ['summary'] in migrationassistant would do the right thing, no?
[10:49] <cjwatson> ev: oh, yeah
[10:49] <ev> after all, summary sources data from m-a
[10:50]  * ev tries
[11:46] <CIA-3> ubiquity: evand * r3836 ubiquity/ (2 files in 2 dirs):
[11:46] <CIA-3> ubiquity: Specify migrationassistant's order as simply coming before summary,
[11:46] <CIA-3> ubiquity: as setting it as coming after user-setup and partman was causing it
[11:46] <CIA-3> ubiquity: to come after summary (LP: #526581).
[11:46] <ev> ^ I'd like to fix this properly by rewriting order_plugins as a topological sort, so that we can specify both BEFORE and AFTER and ubiquity will do the right thing.
[11:52] <davmor2> ev, cjwatson: still get the bug with live desktop saying there is a volume with software on it
[11:52] <cjwatson> bleh
[11:53] <cjwatson> I thought I'd nailed that to the floor
[11:58] <ara> cjwatson, do you know if  Spanish lang packs are in the CD?
[11:59] <cjwatson> which CD?
[11:59] <ara> cjwatson, sorry, alternate
[12:00] <cjwatson> http://cdimage.ubuntu.com/daily/current/lucid-alternate-i386.list says no
[12:01] <ara> cjwatson, OK, it makes sense now, thanks
[12:01] <cjwatson> (for desktop CDs, you need to look at both .manifest and .list, BTW)
[12:01] <cjwatson> davmor2: is there a cdrom entry in /etc/apt/sources.list?
[12:03] <davmor2> cjwatson: no all http://
[12:04] <cjwatson> could I see /var/log/casper.log?
[12:04] <cjwatson> FWIW I'm not seeing this behaviour here, in kvm
[12:04] <cjwatson> so I did at least manage to successfully fix it for myself :)
[12:05] <cjwatson> it might depend on the CD kernel driver in use or something
[12:07] <davmor2> cjwatson: yeap just juggling a few things for a second and then I'll get back to you
[12:18] <davmor2> cjwatson: http://www.davmor2.co.uk/casper.log
[12:25] <davmor2> cjwatson: is that all you need?
[12:37] <CIA-3> ubiquity: jriddell * r3837 trunk/ (debian/changelog ubiquity/frontend/kde_ui.py):
[12:37] <CIA-3> ubiquity: * kde_ui.py: don't show integrated progress bar, it seems to cause a crash
[12:37] <CIA-3> ubiquity:  https://launchpad.net/bugs/526534
[12:37] <CIA-3> ubiquity: * kde_ui.py: Show progressDialog for install, still needed (LP: #526454)
[13:12] <CIA-3> ubiquity: jriddell * r3838 trunk/debian/changelog: releasing version 2.1.26
[14:22] <CIA-3> ubiquity: cjwatson * r3839 ubiquity/ (debian/changelog ubiquity/components/ubi-partman.py):
[14:22] <CIA-3> ubiquity: Check whether choose_partition should be thawed when going back from
[14:22] <CIA-3> ubiquity: scanning a non-resizable partition (LP: #527057).
[15:08] <CIA-3> ubiquity: cjwatson * r3840 ubiquity/ (bin/ubiquity-wrapper debian/changelog): devkit-disks was renamed to udisks; try to inhibit that first.
[15:13] <CIA-3> ubiquity: cjwatson * r3841 ubiquity/debian/changelog: releasing version 2.1.27
[15:27] <mpt> ogra, thomas was interested in the timezone map in particular.
[15:27] <ogra> ah
[15:27] <mpt> It would be nice if that same map was used in Time & Date Settings (and in the clock applet).
[15:28] <ogra> i thought the slideshow, which is in a different package and surely has its own copyright file
[15:39] <ara> cjwatson, the iSCSI test cases are now in the server images under "optional testcases": http://iso.qa.ubuntu.com/qatracker/test/3666
[15:40] <ara> cjwatson, also, I wrote an example using KVM & Virt-manager
[15:40] <ara> cjwatson, you can check it at http://testcases.qa.ubuntu.com/Install/ServeriSCSIRoot/KVMExample
[15:40] <ara> cjwatson, to see if the steps are OK, or this is supposed to be easier for users
[15:50] <ev> michaelforrest: sorry to be so late in reply; I was at Pycon from Thursday to yesterday.
[15:51] <ev> michaelforrest: re installing extras> how do you want to handle the case whereby we don't have a network connection?
[15:51] <ev> (figured this would be quicker than email)
[15:52] <michaelforrest> ev: i figured it would be deferred
[15:53] <ev> so, not show the page at all, or somehow queue the install of the packages up for when a connection is present?
[15:53] <michaelforrest> ideally queue the install ready for when a connection is present.
[15:53] <michaelforrest> but if this is too hard for now (should really have a framework for that!) then possibly don't show the page
[15:54] <ev> okay
[15:54] <michaelforrest> although if we're numbering installer steps this is a proble!
[15:54] <michaelforrest> *problem! unless you calculate based on whether there's a network connection
[15:55] <ev> but the numbers are going away next cycle, right?  So I'm not too concerned about that
[15:56] <ev> I am worried about how we can figure out if we can hit archive.ubuntu.com, and how happy the powers at be are with us running /usr/share/doc/libdvdread4/install-css.sh
[15:58] <ev> while not breaking on captive portals and not downloading a massive amount of updates (so no to just running apt-get update first, which would checksum the download :) )
[15:58] <ev> I guess we could do the http://www.apple.com/library/test/success.html thing
[15:58] <ev> (sorry for thinking out loud)
[16:06] <cjwatson> ara: thanks, much appreciated, the last bits of fix-iscsi-root are about second on my to-do list at the moment
[16:20] <superm1> ev, it looks like that function for copying oem-config keys in install.py doesnt work still since the db is locked
[16:21] <superm1> (i haven't been able to test all the way up to that part until now for other extenuating circumstances)
[16:21] <ev> line?
[16:22] <superm1> 2147/install.py
[16:27] <cjwatson> yeah, that can't work, why not do it in bin/ubiquity with the other debconf-copydb stuff?
[16:30] <ev> works for me
[16:35] <ev> superm1: http://paste.ubuntu.com/383106/ - look okay to you?
[16:36] <superm1> ev, yeah that's probably sane
[16:38] <ev> thanks
[16:39] <CIA-3> ubiquity: evand * r3842 ubiquity/ (bin/ubiquity debian/changelog scripts/install.py):
[16:39] <CIA-3> ubiquity: Copy oem-config debconf questions only after we're done using the
[16:39] <CIA-3> ubiquity: database.
[16:45] <cjwatson> maybe we should copy oem-config/* only in oem-config mode?
[16:47] <kirkland> cjwatson: i was recovering a broken system yesterday with a Lucid server ISO, and I noticed that /sbin/fsck.ext4 was not in the rescue image
[16:47] <kirkland> cjwatson: is this something we could rectify easily, seeing as we're defaulting to ext4 on new installs?
[16:48] <ara> cjwatson, <ara> davmor2, can you test something quickly on real HW?
 go on
 davmor2, kubuntu desktop, clicking on "Install Kubuntu" (without passing by the live env)
[16:48] <ara>  davmor2, does it start correctly for you?
 might take a minute or 2
 ara: running now
 davmor2, at it starts ubiquity correctly?
 ara: nope blank screen after the plymouth logo disappears
 davmor2, same thing...
[16:50] <cjwatson> kirkland: probably, for now you can just symlink it to e2fsck
[16:50] <ara> cjwatson, ok, known issue
[16:50] <cjwatson> ara: Riddell would know more
[16:50] <kirkland> cjwatson: ah, okay, so e2fsck is capability of checking/repairing ext4 as well
[16:52] <cjwatson> it's the same binary - 'ls -li /sbin/fsck.ext*' and you'll see they're hardlinks on a normal system :)
[16:54] <cjwatson> kirkland: pending upload, remind me after a3 just in case?
[16:54] <cjwatson> actually never mind I'll put it in the *short* to-do list
[16:55] <kirkland> cjwatson: doh, cool, thanks;  will remind you later
[17:37] <cjwatson> kirkland: http://bugs.debian.org/571247
[17:37] <cjwatson> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=571247
[17:37] <cjwatson> silly bot
[17:40] <kirkland> cjwatson: cool, thanks
[17:40] <kirkland> cjwatson: just add one for ext5 too, for fun :-P
[17:41] <cjwatson> ... no :)
[20:21] <CIA-3> ubiquity: superm1 * r3843 ubiquity/ (debian/changelog ubiquity/components/ubi-usersetup.py):
[20:21] <CIA-3> ubiquity: Don't mistakingly always show ubiquity running in debug mode warning
[20:21] <CIA-3> ubiquity: in oem-config mode.
[23:09] <jcole> hello everyone, ive remastered an ubuntu install cd and noticed the sources.list mirrors after the install are not the same as on my livecd... does the ubuntu installer get the values from debconf?
[23:11] <cjwatson> no, apt-setup generates them.  in this case, see bug 39156
[23:12] <jcole> cjwatson: perfect, exactly what i was looking for
[23:12] <jcole> cjwatson: thank you