[05:57] ubiquity: superm1 * r3140 mythbuntu-ubiquity/ (debian/changelog ubiquity/frontend/mythbuntu_ui.py): [05:57] ubiquity: * Mythbuntu: [05:57] ubiquity: - Don't prevent going forward in the UI if the connection test fails. [05:57] ubiquity: It's no longer necessary for 9.04 to actually work. (LP: #343432) [11:02] a friend of mine just test the 9.04 beta on a system with window already installed and he got: http://people.ubuntu.com/~mvo/tmp/Bildschirmfoto-1.png in the partitioning step. is that intentional that it defaults to "use entire disk" in this case? [11:08] mvo: I think that's a bug - "side by side" should be the default if offered [11:14] cjwatson: do you want me to file it into LP? I guess I should wait until I can make him run it with debug in ubqiuity [11:14] yes please [11:14] (file it; obviously debug output would be good too) [11:14] I suspect it is easily reproducible though [11:58] debian-installer: cjwatson * r1074 ubuntu/debian/changelog: last commit closes LP: #345534 [12:29] <_ruben> man .. why is grub causing troubles on a large number of recent deployments .. getting fairly annoying .. installing onto pata flashdisks and also sata disks for storage [12:29] <_ruben> both hardy and intrepid [12:37] debian-installer: cjwatson * r1075 ubuntu/debian/changelog: releasing version 20081029ubuntu29 [12:37] I'm just testing a usb stick produced with usb-creator. when I try to install wine via apt-get it tells me "no space left on device" is that intentional, i.e. a known limitation ? I(using the default options in usb-creator) [12:38] mvo: Evan's off this week and I'm not sure I know [12:39] might it depend on the amount of space you allocate for persistent data in usb-creator? [12:39] I think we selected 128mb (I have a intern doing some testing for me, I need to confirm to be 100% sure) [12:39] I will file a bug and just ask evan to close it if its intentional [13:01] ubiquity: cjwatson * r3141 ubiquity/ (debian/changelog debian/rules scripts/install.py): [13:01] ubiquity: Depend on grub | grub-pc, and don't remove grub-pc until after [13:01] ubiquity: grub-installer has had the chance to decide whether it wants to use it [13:01] ubiquity: (LP: #349835). [13:29] ubiquity: cjwatson * r3142 ubiquity/ (4 files in 2 dirs): [13:29] ubiquity: Add a compatibility wrapper for update-dev to ensure that it never [13:29] ubiquity: attempts to call 'udevadm trigger', which isn't necessary in ubiquity [13:29] ubiquity: and can cause problems (LP: #349937). [13:45] cjwatson, btw, ppp-udeb is in main, not universe. [13:53] cody-somerville: so it is, but it's Priority: optional and therefore not used by default. If it's being pulled in for you then you must be doing something wrong [13:53] nothing depends on it [13:54] ubiquity: cjwatson * r3143 ubiquity/ (debian/changelog ubiquity/validation.py): [13:54] ubiquity: Permit dmraid-style /dev/mapper/* device names in advanced boot loader [13:54] ubiquity: selector, as well as disk and partition numbers over 9 (LP: #342354). [13:54] cjwatson: what is the right place in ubiquity to remove stuff that is on the livefs but should not be on the installed system? /var/lib/apt-xapian-index should be removed on install so that it can be re-created freshly with the full sources.list of the installed system [13:59] scripts/install.py - is that a directory? [13:59] yes [13:59] I don't have it on my system [14:01] mvo: will it be reindexed automatically or does ubiquity need to run something? [14:01] cjwatson: it will be re-indexed automatically when the next apt.cron.daily runs (I'm adding this now) or if synaptic detects its missing [14:02] http://paste.ubuntu.com/140714/ [14:02] oh, I already had a different patch written [14:02] I put it in configure_apt [14:02] heh :) [14:02] ok [14:03] where did you put it? just out of curiosity [14:03] though I guess it's better in configure_apt than in configure_hardware where I put it [14:04] ubiquity: cjwatson * r3144 ubiquity/ (debian/changelog scripts/install.py): [14:04] ubiquity: Remove /target/var/lib/apt-xapian-index per Michael Vogt, since it needs [14:04] ubiquity: to be rebuilt based on the installed system's sources.list, and apt's [14:04] ubiquity: cron.daily script and/or synaptic will do this. [14:04] thanks [14:05] * mvo hugs cjwatson [14:05] np [14:07] thanks for the apt merge [14:16] np [14:17] I will upload new apt today, I was just merging some small apt.cron.daily fixes in [14:22] ubiquity: cjwatson * r3145 ubiquity/ (debian/changelog ubiquity/components/partman.py): [14:22] ubiquity: Preserve ordering of automatic partitioning choices when replacing them [14:22] ubiquity: with our customised strings (LP: #351547). [14:35] ubiquity: cjwatson * r3146 ubiquity/debian/changelog: typo [15:16] ubiquity: cjwatson * r3147 ubiquity/ (70 files in 2 dirs): Update translations from Launchpad. [15:27] ubiquity: cjwatson * r3148 ubiquity/ (d-i/manifest debian/changelog): [15:27] ubiquity: Automatic update of included source packages: apt-setup 1:0.37ubuntu11, [15:27] ubiquity: base-installer 1.98ubuntu4, console-setup 1.28ubuntu7, grub-installer [15:27] ubiquity: 1.36ubuntu5, hw-detect 1.71ubuntu6, partman-base 129ubuntu4, [15:27] ubiquity: partman-target 58ubuntu6. [16:22] superm1: your revision 2984 broke the indication of language support being missing for any language not on the CD. bug 337748 [16:22] Launchpad bug 337748 in ubiquity "[jaunty] ”Incomplete Language Support” message not shown anymore after installation" [Undecided,Confirmed] https://launchpad.net/bugs/337748 [16:23] superm1: I think language-support-* should be treated specially instead [16:24] actually, it's more complicated than that, since language-pack-foo might be on the CD while language-support-foo might not be but it might still exist [16:44] ubiquity: cjwatson * r3149 ubiquity/debian/changelog: releasing version 1.12.0 [18:08] cjwatson, oh man. considering it's more complicated than it first seems, feel free to revert it then and i'll put a workaround back in the factory process for it until a better solution is come up with [18:09] I was wondering if there might be something straightforward like extracting the full list of packages that actually exist from the live filesystem's apt cache [18:10] I say "straightforward" [18:10] (actually pain in arse because scripts/install.py configures apt up-front to only use /target) [18:10] oh, but the blacklisting code runs before that ... hmm [18:11] maybe then select_language_packs should check whether each package exists and only add it if so [18:13] superm1: something like http://paste.ubuntu.com/140842/ maybe === mcasadevall is now known as NCommander [22:45] cjwatson, ill give that a shot, but i'd worry it's going to increase install time by iterating a large list again [22:45] oem-config: superm1 * r640 oem-config/ (debian/changelog gui/glade/step_language.glade): [22:45] oem-config: Change the number of columns in the language icon view to 5. This [22:45] oem-config: should make it fit in 1024x600 or at least very close. (LP: #347350) [22:48] superm1: wouldn't expect so, checking for existence in the cache should be really quick [22:48] oem-config> thanks, that's appreciated [22:48] cjwatson, yup np. it was really bothering me and looked obv to fix [22:49] superm1: ideally I'd like to figure out how to make the padding less unnecessarily large [22:49] superm1: but I stared at GtkIconView for ages and couldn't see how to do that [22:49] cjwatson, yeah i was thinking the same thing, I figured there should be a lever for it's padding but it wasn't standing out [23:00] cjwatson, actually you know what. that iv_size_req thing: if it's taken out, everything fits in 800x576 perfectly for me [23:01] cjwatson, evan said it was a hack to not show scrollbars in the code, but they're not showing anyway for me with it commented out [23:01] that's where the extra padding is coming from i think [23:01] huh, odd. Is that producing *internal* padding? [23:02] it's producing padding on the iconview [23:02] i.e. around each individual label? [23:02] ah that; no. [23:03] I think if that makes it fit in 800x576, that's good cause to take it out for now [23:04] we have revision control if we need to restore it later [23:04] ok i'll commit that too then. [23:05] cool, thanks [23:05] I was planning an oem-config upload tomorrow [23:06] in the process of going thru this, i caught a small crasher that happened going back and forth between pages too. might want to double check for that before you upload if you can [23:06] submitted an apport on it [23:06] ok [23:13] oem-config: superm1 * r641 oem-config/ (debian/changelog lib/frontend/gtk_ui.py): [23:13] oem-config: * Remove old hack in gtk frontend to only show scrollbars on language [23:13] oem-config: selector when it would be larger than the desktop. With the hack taken [23:13] oem-config: out, the scrollbars aren't showing up on 800x576, which is likely the [23:13] oem-config: lowest supportable resolution. [23:34] cjwatson: I think that brltty and no keyboard bug is due to no udev loaded, so no USB keyboard modules loaded. I wonder whether that setup script should be moved to be after udev, assuming that this is possible before usplash loads... [23:34] * TheMuso has a look [23:35] Looks like not. [23:43] cjwatson, that ubiquity diff didn't appear to work properly in a test run, i'll think about it more, but feel free to revert the changeset that caused it to come back up if nothing comes up solution wise