[09:54] xnox: http://paste.ubuntu.com/1274646/ is what I meant in my MP comments the other day - does that work for you? [09:55] cjwatson: yeap, I did understand you =) and it should work. Didn't test it yet. [09:56] cjwatson: are you prepping a release? [09:56] Not immediately [09:56] (well, upload) [09:56] Thought I'd do a translation sync while I remember, but doesn't seem worth respinning for as yet [10:12] xnox: Anything I can do to help with bug 1051306, or have you managed to reproduce it? [10:12] Launchpad bug 1051306 in os-prober "Windows not found unless partition is mounted" [Critical,Confirmed] https://launchpad.net/bugs/1051306 [10:13] cjwatson: got that one assigned late at night yesterday. [10:13] cjwatson: didn't do much with it. any pointers would be nice. [10:13] * xnox doesn't have windows machines handy by the way. [10:14] I can possibly resurrect an install in the corner [10:14] Will take a bit [10:14] Plus considerably more waking up [10:15] =) i used to have a windows xp VM somewhere.... on one of these external drives. [10:15] * xnox looks at a box in a corner [10:31] cjwatson: that paste from you http://paste.ubuntu.com/1274646/ looks very nice, just tested. " no visible jitter" === maxb_ is now known as Guest72551 [12:06] Right, restoring this Windows install now. May take a while [12:58] cjwatson: the debug is confusing. the syslog shows that it did mount it. [12:59] cjwatson: could there be a difference between ntfs & GRUB ntfs drivers? [12:59] with the later one not showing some files e.g. bootmgr. [12:59] It's certainly *possible* [13:00] It would be extraordinarily tedious [13:01] I can't reproduce it here :( [13:02] which windows do you have? [13:02] Device Boot Start End Blocks Id System [13:02] /dev/sda1 63 80324 40131 de Dell Utility [13:02] /dev/sda2 * 81920 501379071 250648576 7 HPFS/NTFS/exFAT [13:02] is in the bug report with windows 7 (loader) [13:02] Windows 7 [13:03] I'll ask for a bit more output [13:03] sure. [13:03] I have a separate loader partition, apparently, so not quite the same [13:23] cjwatson: hmm... os-prober uses unshare and hence gets a different mount context.... any way to hack into it? as the OP in bug 1051306 cannot list the files grub-mount is showing. [13:23] Launchpad bug 1051306 in os-prober "Windows not found unless partition is mounted" [Critical,Confirmed] https://launchpad.net/bugs/1051306 [13:24] Err [13:24] I didn't ask him for anything involving os-prober [13:24] (their own implementation of unshare but whatever) [13:24] true. [13:24] * xnox is confused sorry. [13:24] So that shouldn't be remotely relevant [13:24] cjwatson: ok. but then why can he not access the files? [13:25] He hasn't bothered to read what I wrote and follow it exactly [13:25] sudo grub-mount results in something that's only root-readable by default [13:25] If he'd followed my directions this wouldn't have been a problem [13:26] But he editorialised [13:26] "sudo ls -la /mnt doesnt give the correct result either." [13:26] hmm... [13:26] And he didn't tell me what [13:26] * cjwatson has little sympathy with people failing to follow precise directions and thinking they know better [13:34] xnox: Any luck with bug 1065034? [13:34] Launchpad bug 1065034 in ubiquity "'ubuntu ubiquity: umount: /tmp/tmp.h3NCLhoxSh: not mounted' during a Reinstall attempt on a previously manually partioned vm installation" [High,Confirmed] https://launchpad.net/bugs/1065034 [13:35] cjwatson: nope. still need to debug which (u)mount this is coming from. [13:36] you want candidates, I get it =) [13:38] It's listed as a respin trigger so I need to make sure we're on top of it. [13:38] Can you reproduce it? [13:39] not in the past. will try the current image. [13:45] xnox: ooh - well, I may not have been able to reproduce on my test machine, but I sure can on my laptop, albeit not with NTFS [13:46] grub-mount $my_root_partition /mnt -> empty /mnt [13:46] xnox: Mind if I reassign this bug to grub2 and steal assignee? [13:46] cjwatson: knock yourself out =) [13:47] cjwatson: well, maybe os-prober should check that after mounting /mnt has some files. & try remounting without grub-mount. that would be a workaround to the problem. [13:48] * xnox off to the reinstall attempt then. [13:48] But it would reintroduce the very problems that grub-mount was introduced to avoid, and worse, would only do so sometimes depending on whether grub-mount happened to work or not. [13:48] We should fix the underlying bug instead. [13:49] I just pushed a fix for visible formatting strings in kde frontend bug 1065989 [13:49] Launchpad bug 1065989 in ubiquity "[kde] formatting for strings visible in string" [Undecided,Confirmed] https://launchpad.net/bugs/1065989 [13:49] I'm not keen on stacks of software working around other bits of software. [13:49] and it's perfectly valid to have empty file systems ;-) [13:50] Riddell: Er, those strings had already been translated. [13:50] Can we not strip the tags in code instead? [13:51] If not, you need to go through all the translations and fix them. [13:51] cjwatson: looking at ubiquity.template from today's image there's no translations? [13:51] Look at bzr. [13:51] I did a translation update from Launchpad earlier today. [13:51] aah [13:51] I strongly recommend stripping in code if at all possible, because otherwise confusion will ensue ... [13:52] Riddell: translations are in the .po files. And note my comment on #ubuntu-release. that you didn't "fix" all existing translations. [13:52] yep will do [13:52] But even then, fixing the .po files may well create confusion at the next translation sync [13:53] It could easily take an upload or two to get things synced up again, and we don't have time. [13:53] They should be fixed at some point, mind, unless you really want those strings to contain markup forever... [13:53] Sure. But for now, better to work around in code. [13:53] (But maybe after release, sure) [13:53] cjwatson: wouldn't the upload import the translations into launchpad? [13:53] xnox: Yes, but consider translations between the point I downloaded and the point we upload. [13:53] xnox: translation imports aren't all that quick, plus yeah, collisions. [13:53] *sigh* I see. [13:53] Given the number of translations we've had already, this isn't an implausible problem. [13:54] * infinity notes that he gets email about errors in the dpkg translations usually a day or two after his uploads, so it can't be quick... [13:54] The queue backs up quite a bit sometimes. [13:54] Which is why I always forget to fix those two errors, because I've context switched away by then. [14:02] cjwatson: i cannot reproduce grub-mount not showing any files locally. *confused* [14:05] ok nicer fix pushed for formatting strings in kde frontend [14:09] Riddell: Thanks, should be better for now [14:11] Riddell: true. we can fix the translations post-q release. [14:56] cjwatson: Are we still hoping for more ubiquity bugfixes that can be rolled up with Riddell's? [14:57] infinity: i can reproduce bug 1065034 locally and working on it. [14:57] Launchpad bug 1065034 in ubiquity "'ubuntu ubiquity: umount: /tmp/tmp.h3NCLhoxSh: not mounted' during a Reinstall attempt on a previously manually partioned vm installation" [High,Confirmed] https://launchpad.net/bugs/1065034 [14:57] infinity: but it's not ready yet. [14:59] xnox: That's fine, the current images are still building anyway, maybe we can give people a small chance to test them before we invalidate them all. :P [15:00] infinity: you are cruel. =) have you read the Hunger Games? You remind me of gamesmakers that set off a wall of fire against contestant lol =) [15:00] infinity: yeah, that one. also I'm going to have a grub2 fix as soon as I work out what it is.