[00:19] rockstar: around ? [00:19] https://bugs.edge.launchpad.net/launchpad-code/+bug/623102 needs qa [00:19] <_mup_> Bug #623102: AttributeError on +request-builds page

[00:19] its now the front of the queue in https://devpad.canonical.com/~lpqateam/qa_reports/deployment-stable.html [01:16] Fuck. [01:17] Does anyone know how far back we keep primary archive publisher logs? [01:24] good morning wgrant :-) [01:25] jelmer: Morning. [01:48] >>> BinaryPackagePublishingHistory.distroarchseries in [1] [01:48] True [01:48] That looks like a bit of a misfeature. [01:54] whu? How is that happening? [01:56] jelmer: I suspect it's because __eq__ returns a Storm expression. [01:56] Which evaluates to True. [02:00] (and causes a very nasty bug.......) [11:00] wgrant: nice branches [11:00] lifeless: I'm getting lots of timeouts on MPs tonight. [11:00] Hah, nice timing. [11:00] got some #'s [11:00] OOPS-1736EC482 [11:01] I suspected librarian issues, but then there were some reports of abnormal IBugTarget:+bugs timeouts in #launchpad. [11:03] 113003.01SQL-launchpad-main-masterSELECT "_6".branch, "_6".id, "_6".revision, "_6".sequence FROM BranchRevision AS "_6" LEFT JOIN BranchRevision AS "_7" ON "_7".revision = "_6".revision AND "_7".branch = %s WHERE "_6".branch = %s AND "_6".sequence IS NOT NULL AND "_7".id IS NULL ORDER BY "_6".sequence DESC LIMIT 10 [11:04] (13 seconds) [11:05] possibly fat indices on branchrevision [11:07] Or general DB slowness. [11:07] Given the other issues this evening. [11:22] were there oops ids for them? [11:22] we might be running on a different master I guess; I dunno where the migration got to. [11:23] OOPS-1736F673 [11:24] lifeless: That was one of them. [11:28] 115046.01SQL-launchpad-main-masterSELECT Bug.heat FROM Bug, Bugtask WHERE Bugtask.bug = Bug.id AND Bugtask.product = 8920 ORDER BY Bug.heat DESC LIMIT 1 [11:28] (15 seconds) [11:28] so yes, some issue [11:29] man the sodium crashes make our reporting so useless [11:29] I can't really tell if there is an oops spike or not. [11:51] jml: around ? [12:01] lifeless: am now. [12:03] jml: ah, lp:~lifeless/launchpad/bug-627701 [12:03] jml: It failed on ec2; I've pushed up a fix, but its a little deeper than previous [12:04] jml: I would like another eyeball over it [12:05] lifeless: sure [12:06] jml: I'm primarily worried about clarity for the next person eyeballing the code [12:06] pep8 grumble grumble [12:08] I will also admit that I sent it to ec2land when I thought you were out :> I will land a tweak if you have changes, and it gets through as-is. [12:09] + # DB tracers are not safe for reentrant use, so we nmust do this [12:09] typo [12:10] thanks [12:10] lifeless: other than that, looks about as clear as it's going to get. _get_request_timeout is hard to follow, but that's not new in your most recent change. [12:12] lifeless: I suspect anyone wanting to change this code is going to have to think hard & do some poking around (e.g. what's opstats? I kind of know, but I'd want to know more before I changed that code) [12:12] yeah [12:13] the opstats conditional there is fugly [12:24] jml: I hope my mail about qa didn't come across as grumpy or whatever; it wasn't meant to be [12:24] lifeless: no, it doesn't. [12:24] lifeless: I have some problems with the process that I think are fairly significant [12:24] https://devpad.canonical.com/~lpqateam/qa_reports/deployment-stable.html [12:25] jml: we should debug those [12:25] lifeless: but it's hard to be clear about them at the same time as being civil. :) [12:25] jml: in private mail you can be less civil if you like ;) [12:25] lifeless: ok :) [13:06] wgrant: I think there may be a db issue; theres a spike in OOPSes even though the data is crap. [13:07] losa ping - we're seeing normally fine queries ballooning to 15 seconds, multple tables; if someone can eyeball the db servers health etc that would be awesome [13:13] lifeless: sodium's still being crap? [13:19] wgrant: yes [15:56] hey guys [15:56] oh nigelb is here too [15:56] heh [15:57] Ursinha, there? [15:57] Ursinha, we need help [15:57] i registered a project but set the worng maintainer [15:57] and now cant change it to m4n1sh [15:58] beuno, jml ^^ [15:58] or lifeless ^ [16:17] seiflotfy, what happens when you try to change? [16:17] or you just can't? [16:20] I can't [16:20] (at least can [16:20] (at least can't find where the option is..) [16:20] maintainer == owner, no? [16:21] and you can only give ownership to teams you are a member of [16:24] maxb, Ursinha i cant do anything [16:24] i cant change it bac [16:24] althoug i creatd the project [16:24] there is no team (yeT) [16:25] as I said, you can only change it to teams you are a member of [16:25] i have zero access [16:26] you created a team and gave access to somone else wrongly? [16:26] s/team/project [16:27] nigelb, yes [16:28] In retrospect using a team would have been a better idea. I suppose somone here can change it for you. [16:32] yes please [18:54] any LP admins here? me and seiflotfy have made a huge mistake. jml lifeless Ursinha [19:32] m4n1sh: What exactly is the problem? Project has been given away to person who is not associated with it? [19:34] maxb: check https://launchpad.net/trophylicious [19:34] seiflotfy registsred it and tried to assign it to me (~manishsinha) [19:34] maxb: but did a mistake and assigned it to ~manish [19:34] maxb: now since he is just the driver, he cant revert it back [19:35] maxb: what is the solution now? [19:36] seiflotfy should file a question (https://answers.launchpad.net/launchpad/+addquestion) explaining what happend and asking for it to be reassigned [19:36] maxb: sure. Will tell him when he is around [19:37] thanks [19:37] sigh, I suggested that at the beginning. [19:39] nigelb: that was the only way even I could have thought of, but still [19:39] nigelb: seeing your after a long time. sup? [19:40] m4n1sh: all good, what about you? [19:40] I'm afraid I think you need a full admin to change project ownership, so this will almost certainly have to wait for Monday [19:40] nigelb: mine is also fine, too much work, too much code to handle :) [19:40] lol [19:41] maxb: full admin means? means the admin who handles LP Registry? [19:41] I mean the team which takes orphaned projects and re-assigns them? that team? [19:41] nigelb: yeah. Zeitgeist integration is very very time consuming [19:41] full admin rather than registry-admin [19:43] maxb: okay fine. We can wait till monday. Will tell seiflofty to file the question tomorrow, which can be taken care on monday [19:45] m4n1sh: It would be wise to come back here on Monday and ask nicely for an admin to look at the question [19:50] maxb: who all are the LP admin present on this channel? [19:50] I see gary online a lot of time, does he have admin rights? [20:32] moin [21:48] waaaaaaaaaaah [21:48] i need help [21:58] hi, whats up ?