[01:54] Yippie, build fixed! [01:54] Project db-devel build (331): FIXED in 6 hr 22 min: https://hudson.wedontsleep.org/job/db-devel/331/ [01:54] Launchpad Patch Queue Manager: [rs=buildbot-poller] automatic merge from stable. Revisions: 12284 [01:54] included. [02:02] === Top 10 Time Out Counts by Page ID === [02:02] Hard / Soft Page ID [02:02] 182 / 299 BugTask:+index [02:02] 13 / 12 Archive:+copy-packages [02:02] 12 / 3 Cve:+index [02:02] 10 / 11 Branch:+index [02:02] 8 / 0 BinaryPackageBuild:+retry [02:02] 7 / 27 DistroSeriesLanguage:+index [02:02] 7 / 8 Person:+uploaded-packages [02:14] Why is the top 10 only 7? [02:17] Indeed, 2-4 are omitted. [02:17] Bad lifeless. [02:34] that would be freenode fail [02:34] their new software throttles far too aggressively [19:24] Monday morning, and all quiet [19:25] heh [19:25] can fix that [19:26] thumper: btw maxb reopened your stuck builds fix [19:26] yeah [19:26] saw that [19:35] hmm [19:35] in linking series to branch [19:36] would it make sense to offer a likely list? e.g. code imports + branches called 'trunk', 'MAIN', 'HEAD' etc [19:43] lifeless: file a bug :) [19:44] ok [19:56] are the windmill tests disabled? [19:56] I'm not sure [20:18] g'morning nz [20:57] woo [20:57] no new bugs [20:57] no confirmed bugs [20:58] hi jelmer [21:01] morning [21:02] wallyworld: morning [21:02] wallyworld: thanks for taking the CHR slot [21:02] thumper: np === wallyworld changed the topic of #launchpad-dev to: Topic for #launchpad-dev: https://dev.launchpad.net/ | PQM is open | firefighting: - | On call reviewer: wallyworld | https://code.launchpad.net/launchpad-project/+activereviews [21:06] and woo every undecided importance bug is incomplete. [21:08] ah, this is why i have so much email, perhaps [21:08] not that i get launchpad bug mail any more [21:11] james_w: bug 645640 [21:12] yes? [21:12] james_w: you commented that its blocked, but not by what [21:13] ah bah [21:13] wrong bug [21:13] hang a sec while I try to find the right one again [21:13] well, why is that one incomplete? [21:14] it was incomplete pending a reply, which you have given [21:15] james_w: though I do have another question for you on that bug :) [21:18] james_w: bug 395200 === wallyworld changed the topic of #launchpad-dev to: Topic for #launchpad-dev: https://dev.launchpad.net/ | PQM is open | firefighting: - | On call reviewer: | https://code.launchpad.net/launchpad-project/+activereviews === Ursinha-afk is now known as Ursinha [21:46] poolie: call? [22:18] Anyone up for reviewing https://code.launchpad.net/~wgrant/launchpad/bug-710360-gina-wheezy/+merge/47942? [22:19] wgrant: ugh [22:19] Yes. [22:19] wgrant: are you a reviewer yet? [22:19] No. [22:22] I need a mentor. [22:24] ok, I'll mentor you [22:27] lifeless: Sure? [22:28] yes, I think I can handle it [22:30] Thanks! [22:33] thumper: you free yet? [22:33] wallyworld: yep [22:33] sorry [22:34] mumble? [22:34] ok [22:34] wgrant: can you let bac know, get yourself on the roster etc etc [22:38] StevenK: ping [22:38] StevenK: mumble ping [22:38] thumper: Hai [22:40] thumper: I can [22:41] thumper: You think? [22:46] thumper: I need a bit more time to clean up some SQL [22:46] thumper: Yes [22:46] Same thing [22:46] It's all IPerson.merge [22:47] Speaking of which, can I object to the delete-recipes-on-merge thing? :) [22:47] wgrant: thumper just did [22:48] Yay! [22:49] thumper: So, I don't think the queue-recipes will take very long [22:49] thumper: Then I'll rip team-merge apart [22:49] thumper: branch merging is all done with SQL [22:50] thumper: def _mergeBranches( [22:55] wallyworld: https://launchpad.net/oops-tools [23:00] wallyworld: http://pastebin.ubuntu.com/560402/ [23:02] lifeless, mwhudson: didn't someone organise to have the doctests rendered as ReST and available somewhere on a site? [23:03] thumper: BjornT_ was working on that [23:03] thumper: benji presented a plan [23:03] thumper: ages ago though [23:03] thumper: in july [23:03] so not yet then... [23:03] not yet [23:04] searching my mail archive turns up http://people.canonical.com/~bjorn/lp-architecture/ [23:05] but i don't know if that ever got past proof-of-concept [23:05] it was a redo [23:05] new docs [23:05] new subtree [23:05] existing stuff untouched [23:05] ah right [23:05] I decided it wasn't worth pursuing in that form [23:08] wow RootObject:+daily-builds is sick [23:08] lifeless: how sick? [23:08] https://devpad.canonical.com/~lpqateam/ppr/lpnet/latest-daily-timeout-candidates.html [23:09] never renders in < 9 seconds [23:09] heh [23:10] 273 queries won't help [23:10] should aim for < 20 [23:10] yup [23:12] I was looking for pages whose 99% completing fell between 14 and 15 seconds [23:12] to see the impact of dropping the timeout by a second [23:12] only two [23:13] that, and PillarNameSet:EntryResource:search [23:13] both < 50 hits a day [23:13] lifeless: probably a day or so to fix that one [23:17] StevenK: the morning standups just aren't the same if you can't talk [23:17] StevenK: plz fix :) [23:18] Here I was thinking you'd say they're better if I can't talk [23:26] lifeless: You know that armel PPAs are coming RSN, right? [23:27] fsov rsn [23:27] I believe that's included in the definition of RSN. [23:27] wgrant: I thought we already had armel PPAs? [23:28] jelmer: well sort of, but they're not virtualized [23:28] jelmer: We have restricted fake virtualized builders. [23:28] That are not virtualized. [23:28] right, so we have armel PPAs :-) [23:44] lifeless: Does http://pastebin.ubuntu.com/560425/ look sane to you? I don't think Storm is bracketting the two date_created conditions correctly [23:46] StevenK: why do you have the two comparisons with date created ? [23:47] lifeless: Because if there is a build that's been created within the last 24 hours, we don't want to schedule another [23:47] that only requires a < [23:48] not < is equivalent to >= [23:49] Er, won't that return builds that have been scheduled in the last 24 hours? [23:49] StevenK: I think your math is broken :) [23:50] StevenK: date created is a point in time; we are interested when the build was scheduled before now-23h50m [23:50] StevenK: if we had an interval, we would query on interval > 23h50 [23:51] StevenK: if we have a position, we query on things before that position [23:52] lifeless: Keep in mind a recipe can have more than BuildFarmJob [23:52] More than one [23:52] ah yes [23:52] I really dislike this data model [23:53] So I'd like to make sure if there is one BFJ, it was created more than one day ago, and if there are more than one, that they all were created more than one day ago [23:54] lifeless: Patches welcome [23:54] lifeless: But keep in mind that BFJs are used by more than just recipes [23:55] sure [23:55] the basic problem is that you want to work on stuff that is 'current' but the schema is an archival schema [23:56] better to hav a current schema and an archive schema, if you want archiving. [23:58] lifeless: What does current mean? [23:58] wgrant: live, active [23:58] In this case we are looking for records that might have built a day ago. [23:58] Are they archived? [23:59] wgrant: in a different model you wouldn't be looking for anything in the archive schema [23:59] wgrant: so that question is noddy [23:59] Oh? [23:59] How would you model it? [23:59] StevenK: so, we want SPR with either (no builds) or (no builds in the last 24 hours) right ?