[00:16] are diffs in the merge proposals update themselves? [00:16] on the web that is [00:16] IIRC yes [00:26] geser: force reloaded page and now it updated ;-) === abentley1 is now known as abentley [01:27] hi. is launchpad bzr down atm? [01:38] hmm, seemso [01:46] hi folk, how i can fork a project from launchpad i see add branch button but it create and empty branch [01:47] well easier is to do it with bzr [01:47] bzr branch lp:project [01:48] bzr push lp:~my-id/project/my-fork [01:48] the tip of lp:project and my-fork will be the same [01:48] alternativly if you like your branch you create with web [01:49] just push to it [01:49] bzr push lp:~my-id/project/branch-name-created-online [01:49] and voila ovnicraft has a fork of a project [01:51] xnox, i was thinking in use lp for a personal project i am use bitbucket now, but with this i am going to forget it :) thx [01:51] why [01:51] you can push branches to launchpad without a project created [01:51] or if you want bugtracker and stuff than just create a project [01:52] ovnicraft: what do you actually want to achieve? [01:52] with bitbucket i just click in fork button and him does everything [01:52] hi. is launchpad bzr down atm? [01:55] ovnicraft: well launchpad is different for one it doesn't use hg as backend so such analogy is not possible. You can't fork *all* branches of a project [01:55] xnox, btw thx [01:55] ovnicraft: your welcome [04:12] The fork button in bitbucket is just a workaround for hg's lack of stacked branches, AIUI [04:14] * wgrant finds such a fork button to be a far worse UI than just pushing a new branch... [04:14] But GitHub does it too. [04:20] considerable timesaver to re-uploading a lot of history, though [04:23] maxb: Yes, but if you have to do that then your VCS or configuration sucks. [04:24] you can use bzr branch with two remote branches though, not sure if it gets everything locally and then re-upload to the new branch when it's the same server [04:52] stgraber: in the case of launchpad it will download and upload more than is strictly necessary, but *far* from the whole history. [04:53] stgraber: that is, there's no particular smarts to do the branch operation entirely server-side, but because of automatic branch stacking that doesn't really matter. [04:55] well, anyway, even if you do the branching server-side, you still need to get it on your machine at some point. Only real benefit from it would be using download bandwidth instead of upload bandwidth which for regular users would usually mean faster branching. [05:00] stgraber: right. [07:00] is soyuz processing uploads? [07:01] ..yes. Appears to be some lag. === sale_ is now known as sale [11:41] is the build dispatcher broken? as the build queues are filling but the buildds being idle (https://edge.launchpad.net/builders/) [11:45] and the ones that are claiming to be building something, are building packages which finished two weeks ago (wth?) (https://edge.launchpad.net/~pgquiles/+archive/ppa/+build/1401930) === sjid is now known as Cerveja [12:32] Hi, who is the current help contact [12:55] Yikes, the build farm is totally broken :-( [12:56] Chances of finding a LOSA on New Year's Day seem slim [12:58] For the record, all of these appear stuck: hassium iridium samarium bohrium osmium promethium mercury nobelium [13:00] Arrrrgh. [13:00] It's build #1226841 again. [13:00] I thought that was fixed. [13:00] At least twice. [13:01] is there something wrong with soyuz? [13:01] yes [13:01] aw, darn [13:01] got any specifics? [13:02] Not really, other than it seems to have got totally stuck and we need an admin to fix it. [13:02] does this build also block that other builds get dispatched? === wgrant changed the topic of #launchpad to: Package build dispatching currently broken | http://launchpad.net | Read https://help.launchpad.net for help | Help contact: - | Join https://launchpad.net/~launchpad-users | This channel is logged: http://irclogs.ubuntu.com/ | Launchpad is open source: see channel #launchpad-dev [13:03] geser: Yes. [13:03] There are at least a couple of insane builds building. [13:03] Probably more. [13:03] it's the duplicate build record thing [13:03] I've just run the SQL to clear it === elmo changed the topic of #launchpad to: http://launchpad.net | Read https://help.launchpad.net for help | Help contact: - | Join https://launchpad.net/~launchpad-users | This channel is logged: http://irclogs.ubuntu.com/ | Launchpad is open source: see channel #launchpad-dev [13:05] great! [13:05] Good thing most people will be holidaying not uploading for a bit... give the queues time to ease :-) [13:08] it's only been broken for 8 hours; with all the builders available, it shouldn't take long at all for that queue to clear [13:08] elmo: Should we be concerned that hassium has gone right back to building the problem build? [13:09] maxb: I don't think so no? what is the problem build from your point of view? [13:09] oh, sorry, browser cache fail [13:09] or perhaps launchpad cache fail [13:11] maxb: While I often see three different versions of /builders by repeatedly refreshing, I'm pretty sure hassium is really building that build now. [13:11] The seven other buildds that I mentioned above still look stuck, though [13:11] (bug #499095, that one is. there are several others) [13:11] Launchpad bug 499095 in soyuz "Builder stuck on reset build" [Medium,In progress] https://launchpad.net/bugs/499095 [13:12] blink [13:13] god damn it, I despise the build manager [13:13] elmo: What's it doing now? [13:13] (I think everyone hates it) [13:13] wgrant: https://launchpad.net/builders/mercury [13:13] (and the others maxb pointed out) [13:13] I can re-PPA-ize most of them, but some of these are permanent builders [13:14] Also, /builders/hassium claims to be idle whilst /ubuntu/+archive/test-rebuild-20090909/+build/1226841 claims to be currently building on hassium :-) [13:14] The data model changes seem to have basically caused everything to explode in ways that should really be rejected by the database constraints. [13:15] elmo: By 're-PPA-ize' you mean reset the VM? I'm not sure that'll help much, for some of them. [13:16] wgrant: it's a little more dramatic than that, it's a complete reinstall of the machine to a known good state. which is massive overkill, but it's trivial and invokes the right magic to (usually) make LP "forget" whatever the heck it thinks it was building [13:17] (basically I pretend to be Marc Tardif and claim I'm taking the machine to do enablement work on it - and then immediately give it back again) [13:18] I'll try it on hassium first in any event [13:18] elmo: Have you read the bug about that build? [13:19] oh, err, no [13:19] That one has been problematic a number of times already, and there is SQL there to fix it... although it seems to like unfixing itself. [13:20] Oh, hm, maybe it wasn't ever actually properly fixed after last time. [13:20] AAAA [13:20] there's 6 of them now [13:20] Of what? [13:21] if I run the SQL to detect them from the bug, I see 6 build record ids, not 1 [13:21] Also, that build OOPSes consistently for me now. It didn't 20 minutes ago. [13:21] elmo: Ew. [13:21] http://paste.ubuntu.com/349929/ [13:22] wgrant: what were the changes to the data model for? [13:22] is there a reason why the official lpia buildds sit idle with 9 jobs in the queue while the other official buildds get new jobs dispatched? [13:23] geser: Yes. [13:23] geser: Those are lucid builds that have yet to be destroyed. [13:24] They will not be dispatched, since the chroot has been removed. [13:24] They must have been waiting to build at the moment the chroot was removed, so they are stuck until somebody removes the DAS or kills them manually. [13:24] So there's nothing going wrong here. [13:24] ok [13:26] elmo: Sorry, missed the question. buildd generalisation. [13:26] gack [13:26] elmo: Parts of (currently Soyuz-only, but this will change in a week) buildqueue table were split out into the (previously Code-only) job table. [13:28] The interactions between them all now seem to be... imperfect. [13:31] hi [13:31] i'm getting the following error when i try to push to LP [13:31] Permission denied (publickey). [13:31] bzr: ERROR: Connection closed: Unexpected end of message. Please check connectiv [13:31] ity and permissions, and report a bug if problems persist. [13:31] no idea why this is occurring though :S [13:31] Noldorin: You do not have the appropriate SSH key set up. [13:32] wgrant: strange thing is, it always worked before i upgraded to bzr 2.0 [13:32] Noldorin: You need to have an SSH key on your machine associated with your Launchpad account. [13:32] any suggestions? i'm running windows 7 btw [13:32] wgrant: i have it configured through pageant [13:32] Ah. You might find more help in #bzr. This is not a Launchpad-specific problem. [13:32] and i've checked its the same key that's online [13:33] hmm ok [13:33] not having problems pushing anywhere else though, oddlyt [13:33] You're using bzr+ssh to other servers? [13:34] hi, happy new year everyone! [13:34] no... i see your point [13:34] any losa around? https://answers.edge.launchpad.net/launchpad/+question/93418 [13:35] wgrant: only reason i think this might be LP-related is that launchpad is meant to store my SSH key, but doesn't seem to be doing so correctly. [13:35] https://launchpad.net/~noldorin [13:35] * maxb chuckles at "Disabled for Enablement" [13:35] Noldorin: I see a key configured there. What suggests that it is not stored correctly? [13:36] wgrant: the fact that launchpad isn't recognising it? [13:36] of course, it could just as well be a problem with pageant [13:36] I suspect that you have not configured Pageant properly. [13:36] I do not know how SSH keys work on Windows, however. [13:37] wgrant: does it need any special config for bzr 2.0? [13:37] always worked for bzr 1.x hmm [13:37] so hassium is now building something else [13:38] I'm going to assume I can't make anything work by throwing the others back through the airlock process [13:38] err [13:38] anything WORSE [13:38] Noldorin: I do not know. [13:39] elmo: Worth a try. [13:39] wgrant: ok, thanks anyway [13:39] i'll try #bzr === menesis1 is now known as menesis [15:23] Looks like all but samarium and promethium are back in business now [15:23] yeah, I can't reset those through the webui (awesome), I'll do the SQL by hand in a bit [15:33] anyone to boost a quota? please :) [16:12] Happy new Year! [17:06] I can't find the button to report new bug on LP... what's wrong with me? [17:12] Report a bug. trailing vertical side [17:22] thanks [18:05] Hi, is there a help contact around [18:05] Can sonebody restart https://code.edge.launchpad.net/~vcs-imports/rainbow-olpc/mstone-trunk-new for me? === yofel_ is now known as yofel [18:09] translation licensing docs are confusing [18:11] Do answers get automatically solved if there are no further comments by the author in a certain time period? [18:12] It gets fiddly having to keep checking on an answer that I provided, marked as 'Answered', and there has been no further comments by the author, so surely it should be 'Solved'? [18:15] I can't file a bug [18:16] darthanubis: What bug [18:16] bug for mythbuntu [18:16] I keep getting the oops page [18:17] (Error ID: OOPS-1462C1577) [18:17] https://lp-oops.canonical.com/oops.py/?oopsid=1462C1577 [18:18] I've tried over the past 5hrs to no avail [18:23] darthanubis: Done. #502129 [18:23] thank you [18:24] darthanubis: Is that your lp username? [18:24] yes [18:25] darthanubis: I've subscribed you too that bug [18:25] ty [18:31] Does anyone know how often Karma gets updated on a user profile? [18:31] njn: should be daily again [18:31] micahg: Ah, ok. So just once a day? [18:32] njn: yep [18:32] micahg: Thanks === epsy is now known as \u03b5 [21:58] hmm [21:59] has there been a change in the cronned acceptance of source packages? [21:59] it used to be every five minutes [21:59] (this is to Ubuntu proper) [22:02] * xnox was under impression it's every 20mins [22:09] that would be a change from the previous [22:09] and it would have changed fairly recently in my recollection, i.e., within the last couple weeks [22:13] if my memory serves me right then the 20 min is the time between PPA publisher runs and the upload queue is still processed every 5 min [22:14] (of course if nothing broke, you know it's vacation time and services find it a good time to break) [22:16] I've managed to upload at :49 for one and at :54 for two source packages now, and each one hasn't been processed until :00 [22:16] hmm [22:17] builds are also quite delayed, it seems [22:18] the build dispatcher broke this night, but it got fixed already [22:23] crimsun: I've looked at my archived accepted mails for my uploads and the time between the timestamp in the signature and the date header in the accepted mail was max 5 min (sample size: 3) [22:24] all mails had a date header a few seconds past a mutiple of 5 min [22:37] I believe the upload-processor and publisher were running */5 last time I heard [22:37] That is indeed the case. [22:38] Well, the intention. [22:38] Come to think of it, is there any good reason why the upload processor isn't triggered by orderly shutdown of an FTP connection? [22:39] poppy sucks. === maxb changed the topic of #launchpad to: http://launchpad.net | Read https://help.launchpad.net for help | Help contact: - | Join https://launchpad.net/~launchpad-users | This channel is logged: http://irclogs.ubuntu.com/ | Launchpad is open source: see channel #launchpad-dev | samarium and promethium have stuck builds [22:41] poppy should eventually be replaced by some SFTP server that automatically processes the upload and immediately returns a success or failure message across the existing connection. [22:44] Does SFTP allow that sort of feedback? [22:49] hello [22:49] does launchpad have git-serve installed currently? [22:49] maxb: We should be able to just send it over stderr. [22:49] Noldorin: No. [22:49] wgrant: rubbish. so ma i out of luck when it comes to mirroring on github? [22:50] am i* [22:50] maxb: There might be some better method of sending arbitrary text, but I haven't read the spec enough to find out. [22:50] Noldorin: Launchpad can mirror from a git repository, but it only exposes bzr branches. [22:51] wgrant: so no luck mirroring a bzr branch that exists on Launchpad on github? [22:54] Noldorin: I don't know what GitHub supports. [22:54] That would seem to be more of a GitHub issue. [22:54] wgrant: fair enough. i'm asking the git guys too.. [22:54] thanks anyway [22:55] wgrant: hmm. would i have any better luck with SVN? i mean, is bzr-svn installed ont the LP servers? [22:55] Noldorin: Again, from svn, but not to svn. [22:56] heh i see... [22:56] Pushing to svn would require that LP have credentials for every user, which seems ridiculous. [22:56] wgrant: can bzr/launchpad mirror *to* anythiung then? [22:56] that's true... [22:56] guess i haven't thought it through thoroughly [22:56] No, it cannot. [22:58] Noldorin: Well, you can use bzr to push to svn or git, but Launchpad will not do it for you. [23:01] wgrant: oh, i see. that might be the least evil solution then. [23:02] wgrant: so i could in theory do something like "bzr push-git ...." (i have no idea of the actual syntax) after each commit/push to LP to keep my branches in sync? [23:04] Noldorin: 'bzr dpush' for git, I think, but I'm not entirely sure. [23:04] You can work just like normal on a bound svn branch. I'm not sure about git. [23:04] excellent - well, i'll take a look into it myself now [23:04] right [23:05] i need the bzr-git plugin, or does that come with the default windows distribution? [23:05] You'll probably need to install the plugin, but I do not know what the Windows distribution includes. [23:05] right ho [23:05] wgrant: thanks for the help :) will report back if there are any major probs.