[00:02] cdrom-detect: cjwatson * r457 ubuntu/debian/ (changelog po/et.po): Update Ubuntu-specific translations from Launchpad. [00:03] cdrom-detect: cjwatson * r458 ubuntu/debian/changelog: releasing version 1.32ubuntu3 [00:08] console-setup: cjwatson * r144 ubuntu/debian/ (changelog po/cs.po): Update Ubuntu-specific translations from Launchpad. [00:10] console-setup: cjwatson * r145 ubuntu/ (Keyboard/KeyboardNames.pl debian/changelog): releasing version 1.34ubuntu14 [00:17] hw-detect: cjwatson * r140 ubuntu/debian/ (47 files in 2 dirs): Update Ubuntu-specific translations from Launchpad. [00:19] hw-detect: cjwatson * r141 ubuntu/debian/changelog: releasing version 1.73ubuntu3 [00:25] partman-auto-lvm: cjwatson * r230 ubuntu/debian/ (16 files in 2 dirs): Update Ubuntu-specific translations from Launchpad. [00:27] partman-auto-lvm: cjwatson * r231 ubuntu/debian/changelog: releasing version 33ubuntu4 [00:29] cjwatson: kees reported bug 563457 and media-info is not world readable it was from an alternate cd [00:29] Launchpad bug 563457 in xinput "xinput crashes on wacom properties list" [Undecided,New] https://launchpad.net/bugs/563457 [00:31] partman-auto: cjwatson * r314 ubuntu/debian/ (changelog po/cy.po po/et.po po/hi.po): Update Ubuntu-specific translations from Launchpad. [00:31] now you tell me, half an hour after I uploaded cdrom-detect ;-) [00:31] I'll have a look after I've finished this translation pass [00:31] great thanks! [00:33] partman-auto: cjwatson * r315 ubuntu/debian/changelog: releasing version 89ubuntu6 [00:43] partman-base: cjwatson * r205 ubuntu/debian/ (13 files in 2 dirs): Update Ubuntu-specific translations from Launchpad. [00:44] partman-base: cjwatson * r206 ubuntu/debian/changelog: releasing version 139ubuntu5 [00:49] partman-crypto: cjwatson * r698 ubuntu/debian/ (8 files in 2 dirs): Update Ubuntu-specific translations from Launchpad. [00:51] partman-crypto: cjwatson * r699 ubuntu/debian/changelog: releasing version 40ubuntu4 [00:53] partman-ext3: cjwatson * r765 ubuntu/debian/changelog: releasing version 58ubuntu3 [00:56] partman-target: cjwatson * r791 ubuntu/debian/ (changelog po/bs.po po/hi.po): Update Ubuntu-specific translations from Launchpad. [00:58] partman-target: cjwatson * r792 ubuntu/debian/changelog: releasing version 64ubuntu8 [01:02] pkgsel: cjwatson * r159 ubuntu/debian/ (11 files in 2 dirs): Update Ubuntu-specific translations from Launchpad. [01:03] pkgsel: cjwatson * r160 ubuntu/debian/changelog: releasing version 0.25ubuntu7 [01:08] cdrom-detect: cjwatson * r459 ubuntu/debian/ (cdrom-detect.postinst changelog): Make sure that /var/log/installer/media-info is world-readable. [01:09] user-setup: cjwatson * r218 ubuntu/debian/ (50 files in 2 dirs): Update Ubuntu-specific translations from Launchpad. [01:10] cdrom-detect change uncommitted, it needs to be done in installation-report instead [01:12] user-setup: cjwatson * r219 ubuntu/debian/changelog: releasing version 1.28ubuntu6 [01:13] installation-report: cjwatson * r73 ubuntu/ (debian/changelog finish-install.d/94save-logs): [01:13] installation-report: Make sure that /var/log/installer/media-info is world-readable if it [01:13] installation-report: exists. [01:16] installation-report: cjwatson * r74 ubuntu/debian/ (changelog installation-report.postinst): [01:16] installation-report: Make /var/log/installer/media-info world-readable on upgrades to this [01:16] installation-report: version. [01:17] installation-report: cjwatson * r75 ubuntu/debian/changelog: releasing version 2.39ubuntu4 [01:17] bdmurray: ^- fixed, thanks for the heads-up [01:18] cjwatson: no problem, thanks you [01:21] cjwatson: I got an opportunity to retest around bug 538536 last night: basic guided partitioning works great. guided LVM leaves the system unbootable. Do you want a new bug, or do you already know of some oddity with Apple-GPT+grub2+LVM? [01:21] Launchpad bug 538536 in partman-auto "automatic partitioning broken on Intel Macs" [High,Fix released] https://launchpad.net/bugs/538536 [01:22] (also, I don't really care if this gets fixed for release, as I have a path to a working install without LVM) [01:23] I think it would be best to file a separate bug as I've now applied several fixes for bug 538536 and a clean start might be less confusing [01:23] Launchpad bug 538536 in partman-auto "automatic partitioning broken on Intel Macs" [High,Fix released] https://launchpad.net/bugs/538536 [01:23] OK. Is it something you're likely to want to hack on for lucid, or more a maverick thing? [01:26] probably maverick at this point I'm afraid [01:28] I figured as much :) Just wanted to make sure it was safe to do a real install (working around the issues) after the bug filing install. I'll make sure I have some time to do a reinstall on this machine sometime for one of the earlier maverick milestones. [01:28] * persia reinstalls Mac OS X to ensure a sane start-point for the operation === ogra_ is now known as ogra [08:53] casper: superm1 * r809 casper/ (debian/changelog scripts/casper-bottom/24preseed): Unbreak early command from previous commit. [08:54] casper: superm1 * r810 casper/debian/changelog: releasing version 1.235 [09:25] cjwatson: see there's a new netboot/xen dir in i386, would it be possible to do the same in amd64? [09:46] ara: I'm getting a traceback when logging into testcases.qa.ubuntu.com. Would you mind adding the following to the historical bugs section of uoi-003 on my behalf? [09:46] "In the 10.04 cycle, prior to the release candidate, the installer visibly switched to a text console and stayed there for a noticeable amount of time while removing oem-config. Any case where the user is taken out of a graphical experience should be filed as a bug." [09:46] ev, sure, will do [09:47] very much appreciated [09:47] ev, and will investigate the login issue [09:47] cool, thanks [10:24] cjwatson, can you remind me what was the d-i equivalent to http://paste.ubuntu.com/414837/ ? [10:33] is anyone in particular working on RTL in KDE frontend? [10:37] ev, I saw you exported translations from Launchpad yesterday into ubiquity, thanks for that :) As the NonLanguagePackDeadline was today, there might be some teams who were translating yesterday after the export. Would it be possible to have a final export some time today or later? [10:38] dpm: definitely [10:38] cool, thanks! [10:41] dpm: anything I can do to help that lot. We had a number of late string changes to ubiquity, the slideshow, and usb-creator. [10:43] dmarkey: xen shouldn't need it? [10:43] dmarkey: err - amd64 shouldn't need it? [10:44] oh, huh, there's one in Debian [10:44] maybe it just has the extra config files or something [10:45] ev, thanks :) [10:45] hooray, I can reproduce the oem-config-kde loop issue. [10:46] ogra: that would correspond to changes in flash-kernel [10:47] cjwatson, indeed, i have them on the shelf here [10:47] ogra: for dove, we add rootfs='root=UUID=$uuid' to flash-kernel.conf - is that necessary for omap too? [10:47] there was nothing additionally in d-i ? [10:47] no [10:47] http://paste.ubuntu.com/414821/ [10:47] i'm not finished yet but essentially thats how it will look like [10:47] cjwatson: sorry i thought we'd have to break open netboot.tar.gz, but the vmlinuz/initrd is available in ubuntu-installer/amd64/ [10:49] ogra: you need http://paste.ubuntu.com/414852/ on top of that, and you should call check_subarch "omap" in flash-kernel to match the other cases [10:50] ogra: otherwise that should be all you need in d-i [10:50] great [10:50] cjwatson, I've been pointed out to some comments in installation-guide changelog (http://tinyurl.com/y7ef5bq) mentioning that only the English version is built. If that is the case, I think I'll disable the templates in Launchpad, what do you think? [10:50] ogra: you should mirror that uboot-envtools change in ubiquity, though [10:50] you mean make sure it gets pulled in additionally ? [10:50] dpm: I think disabling them would be reasonable. IMO enabling translations needs to involve a big sync with upstream [10:50] and committing the Ubuntu changes upstream as conditionals [10:51] hmm, k ... that makes the patch a bit bigger [10:51] but that's a lot of work [10:51] ogra: uboot-envtools should be added to ubiquity's Recommends on armel, and ubiquity should make sure that it's kept installed on the appropriate subarches [10:51] same as it does for uboot-mkimage now [10:51] right [10:52] cjwatson, ok, I'll disable them for now, thanks. I've got some other questions about debian-installer, but I'll come back to you later. [10:52] first i need it in main :) [10:52] this means things work reliably without network access [10:52] and the package needs some cleanup, asac wasnt happy with it as is [10:54] so thats a better one then i think http://paste.ubuntu.com/414853/ [10:54] dmarkey: so Debian does have a netboot-xen image for amd64, which consists of symlinks to the main kernel/initrd and an added xen configuration file [10:55] dmarkey: doing this for lucid would involve upgrading the build system a bit, and I'd rather avoid it if possible - do you think just externally pointing people to the xen configuration file in i386 is sufficient? [12:01] gah, crap, my CIA setup is borked in my ubiquity branch [12:02] ubiquity: ogra * r4081 ubiquity/ (debian/changelog debian/control scripts/install.py): add support for omap bootloader installation [12:12] cjwatson: it just means the script that downloads the vmlinuz/initrd will have to look in different places to download them between 386/amd64 [12:25] dmarkey: is that tolerable? [12:25] I was skating the edge of the freeze with this change as it was [12:26] will the ubuntu-installer/amd64/ [12:26] directory always exists on the mirror? [12:26] yes [12:26] well, any working mirror [12:27] lemme take a look at that script thats in the i386/xen directory [12:28] but yea, it def wont be a showstopper [12:28] any chance of xen support being in the release notes? :) [12:28] well not support, but "possible to install via Xen PV" [12:30] hmm kernelurl = installer + "/netboot/xen/vmlinuz" [12:39] dmarkey: if you file a bug on the ubuntu-release-notes project as a reminder, it will happen [12:42] cjwatson: which would you prefer, either we patch xm-debian.cfg, or replicate the dir structure in amd64 [12:42] oh, it's not possible to make it work without doing that? [12:42] sigh, guess I'd rather replicate it in amd64 [12:43] xm-debian.cfg: kernelurl = installer + "/netboot/xen/vmlinuz" [12:43] :( [12:43] ubiquity-slideshow-ubuntu: evand * r270 ubiquity-slideshow-ubuntu/ (2 files in 2 dirs): [12:43] ubiquity-slideshow-ubuntu: Add a higher resolution export of the personalize icon. Thanks [12:43] ubiquity-slideshow-ubuntu: Daniel Fore! [12:44] I don't know what the role of xm-debian.cfg is, nor how straightforward it is to override it [12:44] its really a helper script to automate a load of permanent steps, it downloads the vmlinuz/initrd, which you would otherwise have to do manually [12:45] err, small steps [12:52] only bother if you have time. I can document this in the wiki page or something [12:53] is everything working OK on i386? [12:54] it installed OK, i'm just going to test pygrub with it === ogra_ is now known as ogra [13:59] cjwatson: how much time did it take to write the first version of partman-auto-loop? [14:00] kusum1: I don't remember any more - maybe a few days [14:01] it was nearly three years ago, I don't keep notes that far back :) [14:01] just a rough timespan ? [14:01] it took a fair bit of polishing after the first version before it worked properly though [14:01] oh ok [14:02] if I had to do it from scratch again, I suppose I'd quote two to four weeks [14:02] abt 2-3 weeks ? [14:02] ok gr8 [14:02] thank you [14:02] remember that this was starting from a position of knowing our installer's partitioning code very well, though [14:02] about three years of experience with it [14:07] that's a serious point to note [15:15] ah ha, I think I've figured out the oem-config-kde bug. [15:18] cjwatson: hey again [15:18] strange question, but is there a read menuentry changed from 'Ubuntu' to 'Ubuntu' [15:19] read/reason [15:19] in grub.cfg [15:19] god, lemme try that again [15:20] is there a reason menuentry "Ubuntu, with Linux 2.6.32-21-generic-pae" changed to menuentry 'Ubuntu, with Linux 2.6.32-21-generic-pae' [15:30] yes, it was part of fixing quoting problems that caused the whole thing to fall over in a messy heap if a translator included a character special within quotes [15:30] grub2 accepts either double or single quotes there [15:30] see bug 552921 [15:30] Launchpad bug 552921 in grub2 "An apostrophe translation in the line "echo Loading Linux 2.6..." of Grub breaks the boot menu from this entry included in Grub (in Ubuntu 10.04)" [Undecided,Fix released] https://launchpad.net/bugs/552921 [15:31] shit [15:31] that has screwed pygrub [15:32] I really suggest getting pygrub fixed [15:32] I don't want to hold back sensible grub2 configuration fixes because they break pygrub [15:32] oh i'm writing a patch as we speak [15:32] cool, thanks [15:33] I committed that patch upstream as well so it won't just be Ubuntu [15:33] cjwatson, do you plan any other ubiquity uploads before full close down ? [15:33] ev: the loop, or the apt error? [15:33] * ogra committed the omap change but didnt upload [15:33] loop [15:33] ogra: I expect at least one more; please don't upload [15:33] cjwatson, right, thats what i thought [15:34] I'll tackle the apt error next if you're not already on it [15:34] just wanted to make sure its not lost :) [15:34] ev: I was just bringing up a test harness, but if you're already on it then I suggest I should go and see what the state of oem-config on server is instead [15:35] cjwatson: I'm not yet, actually. I'm in the process of verifying the loop change. [15:35] but if you want to fix oem-config on the server, that's a-okay by me :) [15:37] bah, that didn't quite work [15:40] ev: mostly want to not be doing it in release week this time [15:41] I'm missing the context on that slightly. Doing what in release week? Fixing oem-config on the server? [15:41] yes, that happened last cycle [15:42] cf. 2.1.0 changelog [15:45] I definitely remember 2.0.7 and 2.0.8 [16:22] cjwatson: is there a reason the memtest entries are still using double quotes [16:27] they aren't translated yet so haven't needed to switch [16:28] (plus, they're generated by a separate package) [16:47] ev, have you come across some weird padding where text is getting cut off on the sides or moved above the top at all on any of the GTK pages when you switch languages ever? if so, do you have some recommendations for what to do to make it stay in bounds? [16:48] superm1: yeah, it's due to the WrapLabel class and text alignment. There's a proposed branch that I haven't had a chance to review yet: https://code.edge.launchpad.net/~shlomister/ubiquity/bug-560114-ugly-hack [16:50] ev, oh so that's only setting it on a per widget basis.... so it would need to be applied to any widgets it's happening to in third party plugins then too [16:51] indeed [16:52] superm1: did you ever get the DEBCONF_DEBUG=developer log for bug 508173? [16:52] Launchpad bug 508173 in grub2 "postinst has errors with grub-probe that cause the system to stop booting" [Undecided,Incomplete] https://launchpad.net/bugs/508173 [16:52] ubiquity: evand * r4081 ubiquity/ (bin/ubiquity-dm debian/changelog): [16:52] ubiquity: Force garbage collection so we don't end up with stray X resources [16:52] ubiquity: when we kill the X server (LP: #556555). [16:52] Launchpad bug 556555 in ubiquity "oem-config loops indefinitely in Kubuntu OEM installations" [High,Confirmed] https://launchpad.net/bugs/556555 [16:52] and here I thought I understood python's scoping and garbage collection rules. [16:52] completely perplexed by the necessity of that [16:52] nasty [16:53] I think that deserves a tea [16:54] cjwatson: just to be clear, you're not currently working on the oem-config crash that presumably fell out of the progress changes, right? [16:54] that is, I'm free to work on it [16:54] cjwatson, no not yet. i've had to deprioritize for the moment with other things going on right now. i think right now it would be an SRU anyway once sorted, so i'll get it after i can fix these other higher priority things [16:54] it wouldn't necessarily be an SRU [16:55] if it's happening on upgrade, I would prefer to get it fixed for final [16:55] ev: correct [16:55] hadn't got round to watching the videos yet (what's wrong with log files?) [16:56] first video cuts off the critical information [16:57] everyone's obsessing about "no job control in this shell" which is not the operative error :) [16:57] looks like oem-config-wrapper / ubiquity-dm exited non-zero *after* removing ubiquity, for some reason [16:58] oh, wait, am I just an idiot [16:58] yep [16:59] ev: ok, never mind, I've got it [16:59] deserves another test round though, that will be a bit later [17:00] ubiquity: cjwatson * r4082 ubiquity/ (bin/oem-config-firstboot debian/changelog): [17:00] ubiquity: Break out of oem-config-firstboot's main loop if oem-config-wrapper [17:00] ubiquity: succeeds (LP: #558593). [17:00] Launchpad bug 558593 in omsk "During OEM-config removal, there is no graphical feedback" [High,Confirmed] https://launchpad.net/bugs/558593 [17:00] ah nice [17:00] I'll queue a local build and run that through its paces. [17:02] oh, ok, cool, if you could that would be great [17:03] ubiquity: cjwatson * r4083 ubiquity/bin/oem-config-remove: TODO comment about general oem-config-remove approach [17:22] wah, well I'm glad I tested this now [17:23] on server, oem-config and plymouth wind up on the same vt and the world implodes [17:36] ubiquity: cjwatson * r4084 ubiquity/ (3 files in 2 dirs): [17:36] ubiquity: Quit plymouth before starting either the emergency noninteractive [17:36] ubiquity: ubiquity frontend in automatic mode, or oem-config's debconf frontend. [17:39] still not entirely working - I think oem-config maybe didn't give itself a controlling terminal properly, or something [17:41] ... but enough for today [17:51] superm1: the event box you removed on the install window was used for a arguably slightly better visual impression (because we couldn't get rid of the grey border on the bottom without making the window look flat). But that's a-ok as I wasn't a fan of the change, and we desperately needed the vertical space anyway. [17:51] ev, oh i didn't realize it was a hack like that, seemed like an unimplemented oversight [17:51] that's too bad the grey border on the bottom can't be nuked [17:51] my fault for not documenting it better [17:51] well, it can, sort of [17:52] but then you get this: http://people.canonical.com/~evand/tmp/borderless-install-window.png [17:52] when we get client side decorations in metacity, all will be right in the universe [17:52] until then we're screwed :) [17:52] debian-installer: cjwatson * r1289 ubuntu/ (6 files in 5 dirs): Add amd64/netboot-xen configuration, to go with i386/netboot-xen. [17:53] dmarkey: ^- just for you ;-) [17:54] world imploding> yikes! [17:58] shtylman: do you have any time to look into bug 563309? [17:58] Launchpad bug 563309 in ubiquity "ubiquity crashes on manual disc setup" [Undecided,Incomplete] https://launchpad.net/bugs/563309 [17:58] I *think* it's KDE-specific [18:00] ev: I can check it out this weekend [18:01] shtylman: okay, I'll try to give you a hand with it if I'm not otherwise buried [18:01] kk :) [18:05] cjwatson: oem-config works [18:05] there's still a VT switch to the console briefly, but that's always been there [18:05] and I imagine it's not something we can fix for Lucid [18:06] cool - yeah, probably pretty hard to get gdm to reuse an existing X [18:06] as long as it's very brief [18:06] would still like to get rid of the EPIPE errors though ... [18:07] I should hit that with strace [18:08] * ev out - have a great evening guys [18:10] cjwatson: do you know of any more changes to grub.cfg that might be coming down the line before release off the top of your head [18:13] unlikely to be anything much [18:13] hopefully zero [18:14] make sure that your pygrub change handles strings like this [18:14] 'foo'\''bar' [18:14] should expand to [18:14] foo'bar [18:15] i.e. quoted foo, escaped ', quoted bar [18:16] i think it would just truncate the title. [18:16] i will test however [18:18] cjwatson: have you come to any conclusions about the directory structure on the mirrors? [18:22] all thats needed are symlinks to ../ubuntu-installer/amd64/{linux,initrd.gz} :) [18:31] that's what I did in the commit above [18:41] cjwatson: did you catch the decision that we won't show anything where we were getting the squashed ubuntu logo any more? [18:42] it's been a while since I've checked a daily so sorry if this seems late, but I know you had a look at the spec yesterday [18:48] nice dmarkey :) and an indirect thanks to you cjwatson [18:49] cjwatson: you're a star [18:54] michaelforrest: on the boot screen - so lose the circle from that (which should deal with most of it, circles are most visible when distorted)? No, I hadn't seen that. Has the doc team been informed, since that's one screen they're practically guaranteed to have screenshotted? [18:55] cjwatson: the latest decision was to show the background colour and nothing else until plymouth is ready [18:55] if we definitely can't get the graphics to be precisely rendered [18:55] I don't know how to communicate this to the doc team [18:55] I think we must have the icon at the bottom [18:55] e-mail [18:55] if you can't get it to look right, we can't put it in [18:55] it's too important from a design / branding perspective [18:56] sorry [18:56] the icon is not particularly susceptible distortion [18:56] and I'm sorry, I think it's vital [18:56] this is a sabdfl thing [18:56] we either need to make it right , or we don't put it in [18:56] I am happy to argue it with him [18:56] please do [18:56] during working hours [18:57] yea sorry - forgot you might not still be at work! [18:58] btw please don't assume I'll see spec changes - they're not a good way to communicate freeze-breaking changes [18:58] irc (like this) is fine, or e-mail [19:35] Does CASPER_GENERATE_UUID=1 not cause update-initramfs to generate uuid.conf in initrd anymore or did it move or something?