[11:09] nigelb, mhall119, cjohnston http://www.micahcarrick.com/gedit-as-a-django-ide-for-linux.html [11:21] thanks daker :D [11:21] yw ツ === ryanakca_ is now known as ryanakca [11:56] daker: cool, thanks [15:23] james_w: HA, code deletion bug. My favorite type :) [15:23] :-) [17:07] someone added code to summit to delete meetings when the blueprints are deleted? [17:08] hmm, no, that's always been there [17:08] why is it breaking now? [17:08] Because summit is awesome. [17:08] :) [17:10] actually looks like it may be something to do with caching [17:10] james_w: corrupt launchpadlib cache? [17:10] seems that requesting +temp-meeting-export doesn't always give you the same thing [17:10] so it's creating meetings one run [17:10] james_w: were you ever able to get my new unit tests failing properly? [17:10] then deleting them the next run [17:10] mhall119, haven't tried yet, sorry [17:10] Oh god. [17:10] Why launchpad why. [17:10] james_w: was the delete cron being run regularly? [17:11] Is that launchpad caching? Maybe we're hitting multiple app servers [17:12] could be, but if that's the issues then there's a bug in LP [17:12] Well, to be fair on LP +temp-meeting-export is fairly heavy on the DB. [17:12] james_w: have you tested http://ec2-50-16-76-22.compute-1.amazonaws.com/ [17:13] nigelb: yeah, but it should be consistent [17:13] agreed. [19:22] nigelb, mhall119: https://bugs.launchpad.net/summit/+bug/849331 [19:22] that's the cause and fix for the cache issue [19:24] we're going to want a deploy to fix that quickly [19:24] so if I were to say I was going to do that tonight, how scared would you be? [19:24] as long as its checked in, minimal :) [19:25] oh, of course :-) [19:26] just wondered if you were scared about having the code in trunk in production [19:26] I know I need to to fix the autoscheduler for private rooms first [19:26] private sessions I mean [19:26] james_w: you can just merge to the stable branch [19:26] though the autoscheduler isn't yet running [19:26] oh, cool! [19:26] which is probably more stable. [19:26] what a great idea! [19:27] that makes my life easier, thanks :-) [19:27] heh [19:36] mhall119, cjohnston - tarmac set for stable branch as well. Should run. [19:36] (summit) [19:44] nigelb: great, but I'll be killing 1.x as soon as trunk is ready to be merged into production [19:45] mhall119: Dont' [19:45] I'm sure situations like these will pop up [19:45] james_w: explain that one to me? [19:46] wouldn't we get the same thing if we're hitting the cache, rather than different things? [19:46] james_w: yeah, check this into 1.x series [19:46] and it'll be easy to merge it into production and deploy [19:52] mhall119, there are two caches [19:53] and so it's 50% chance of getting each [20:00] yay, failing tests! [20:01] plus ViewDoesNotExist: Could not import summit.sponsor.views. Error was: No module named restfulclient.errors [20:01] which is likely a version mismatch [20:04] yep, that's fixed too [20:05] NoReverseMatch: Reverse for 'summit.schedule.decorators.inner' with arguments '('',)' and keyword arguments '{}' not found. [20:05] getting that on two tests [20:10] james_w: yeah, that's fixed in another MP [20:10] in trunk [20:10] cool [20:10] is the private scheduling one failing properly now? [20:10] * james_w pulls [20:10] yeah, looks like it [20:10] cool [20:10] just getting down to a "clean" test run first [20:11] mhall119, https://code.launchpad.net/~mhall119/summit/fix-etherpad-edit-links/+merge/75061 that one? [20:13] yeah [20:13] thanks [20:14] FAIL: test_no_available_private_room (summit.schedule.tests.PrivateSchedulingTestCase) [20:14] \o/ [20:15] yay [20:16] now this one is tricky, because we want to let the admins schedule a private meeting in a public room if necessary, but we don't want the auto-scheduler doing it [20:18] I don't know [20:18] http://paste.ubuntu.com/688601/ [20:18] passes tests [20:18] there's only one callsite - the autoscheduler [20:19] that'll work [20:19] we can look at adding the ability to auto-schedule private meetings into private rooms another time [20:20] make an MP and I'll approve it [20:20] target this one to trunk though [20:21] yep [20:44] mhall119, nigelb, I made a graph of the summit website model object structure, thought you may find it useful: http://people.canonical.com/~mars/summit-models.png [20:45] mars: did you use the django-extensions for that? [20:45] mars: Neat! [20:51] mhall119, yep. Pretty cool. I find pictures really helpful for exploring and discovery [20:52] mars: yeah [20:52] mhall119, nigelb, if you like it, feel free to pass it on to the other devs [20:52] I did that graphing thing for a project at my last job.....it was bad [20:52] hey mars [20:53] Hey james_w [20:53] james_w, how's the weather up by Owen Sound? Fall colours out yet? [20:54] mhall119: badly designed models? [20:54] nigelb: yeah, and lots of them [20:54] heh [20:55] mars, not quite [20:56] james_w, your team may find this useful as well: http://people.canonical.com/~mars/summit-models.png [20:56] yeah, thanks mars [20:57] very useful when you are getting to know a project [20:57] yeah [20:57] This explains why model_mommy was helpful for building things [20:58] james_w: were you going to make a change to the stable branch today? [20:58] mhall119, I want to [20:58] what's stopping you? [20:58] time [20:58] heh [20:58] https://code.launchpad.net/~james-w/summit/dont-autoschedule-private/+merge/75256 [20:59] ok, I'll check it [20:59] is it lp:summit/stable ? [20:59] I can't load the code page to get the list unfortunately [20:59] lp:summit/1.x [20:59] oh, I know [20:59] oh wait, yes, it is stable now [20:59] forgot we changed the series name [20:59] cool, thanks [21:00] going to work on https://bugs.launchpad.net/summit/+bug/849331 next [21:01] I'd like to deploy https://code.launchpad.net/~chrisjohnston/summit/linaro-link-fix/+merge/74877 soon too, but that one isn't breaking anything [21:02] after that then I want to look at importing from multiple Launchpad sprints for a single summit [21:02] james_w: that was was targetting trunk wasn't it? [21:02] trunk has several menu changes [21:02] Definitely need to remove the "summit" hilight. [21:02] james_w just pinged me with everyline. [21:02] the link fix? yeah [21:02] heh [21:02] summit [21:02] summit [21:02] summit [21:02] I love summit! [21:02] oh god [21:03] james_w: with your fix to the autoscheduler, I think trunk is ready to be merged into /production [21:03] ooh [21:03] exciting [21:03] yeah [21:04] james_w: I'll get you lp import fix landed in stable and pushed to production, then we'll work on getting trunk into production [21:04] ok [21:04] nigelb: what's taking tarmac so long? [21:05] when did you approve the MP? [21:05] there's 3 of them [21:05] maybe they don't have commit messages [21:05] * nigelb looks [21:05] ah, right. [21:06] james_w: heh, so for all the bugs cjohnston and mhall119 fixed, you buy them a drink per bug :P [21:06] mhall119: Overall status is Needs review [21:06] deal [21:06] nigelb: I thought an approved review message was enough for tarmac [21:06] No [21:06] ok [21:06] neither it is for PQM. [21:07] I'm fairly sure there's lots of inspiration from PQM for tarmac. [21:07] so I'd expect similar behavior. [21:07] nigelb: do they still need commit messages too? [21:07] yes [21:08] well, look who decided to show up [21:08] hush [21:08] id love to have 5 minutes to work on something [21:08] mhall119: he heard about the drink per bug deal :P [21:08] let alone take a breath [21:08] But.. I was filing RTs today from class.. so SHUT UP! [21:08] :-P [21:09] mhall119: let me know when you've fixed all the MPs [21:09] cjohnston: speaking of, was that resolved? [21:09] I'll run tarmac manually for ya [21:09] I pinged Ng in -sysadmin [21:09] mhall119: yes.. ty [21:09] nigelb: ready [21:09] cjohnston: cool [21:09] * nigelb runs [21:09] mhall119: has the % thingie been fixed [21:11] mhall119: james_w nigelb I believe we still need to make the conflict resolver NOT resolve private meetings, but move public meetings according to private meeting schedule [21:11] mhall119: did we ever try to get the cranberry memory error RT moved up? [21:11] Try - yes. Actually get it done - No. [21:11] cjohnston, the conflict resolver will not move a private meeting as far as I can see [21:11] mhall119: Done. [21:12] james_w: what about move public meetings if there is a reqired participant scheduled in a private meeting/ [21:12] cjohnston, also, hi :-) [21:12] hey bud [21:12] cjohnston, I think it will do that, but I would need to look to be sure [21:12] wanna fix some status bugs while your at it? ;-) [21:12] lol [21:13] nigelb: any chance you can do a quick fix to https://code.launchpad.net/~nigelbabu/summit/668528-last-refresh/+merge/63448 [21:13] jussi: who was it we needed to bug about getting the bug bot back? [21:13] cjohnston: I bugged already. [21:14] cjohnston: Not ATM. Working. [21:14] bugged about what [21:14] the bot. [21:14] cjohnston: I tried on that RT, jcastro said he doesn't have authority, and jono didn't respond [21:14] ok [21:29] bbl, kids' karate [21:30] gah, can't register the merge now [21:42] well, the fix is in https://code.launchpad.net/~james-w/summit/cache-control/ [21:42] doesn't seem like I'm going to be able to request a merge against stable [21:42] so either someone will have to bypass tarmac for me [21:42] or we can merge to trunk and go with deploying the whole thing [21:43] james_w: launchpad hates you. [21:43] I just proposed a merge [21:43] yes [21:43] it's doing it to a non-default branch that breaks it I think [21:43] I did it against stable [21:43] https://code.launchpad.net/~james-w/summit/cache-control/+merge/75265 [21:44] well, ok [21:44] thanks :-) [21:44] I think launchpad has inbuilt burn out prevention :P [21:44] what did you put in the box? [21:44] Nothing :p [21:44] description? [21:44] no, the target box [21:44] I clicked stable [21:44] ok [21:45] I didn't have that as an option [21:45] did you do it differently? AH. [21:45] thanks [21:45] well, we clearly don't want that print in there [21:45] * james_w fixes [22:12] argh, I can't get a migration going for a new model [22:22] I just want the new table so I can test that it works at all! [22:24] and what do you know, complaining made me stumble on the answer [22:24] missing Meta [22:25] that is so Meta :P [23:08] wahey, found another regression [23:09] james_w: in cache? [23:10] I just approved that one. [23:10] nope [23:10] in importing [23:10] Ah. [23:10] it now crashes if there is a "." in the meeting name [23:10] I thought I fixed that. [23:10] * nigelb clearly remembes fixing that. [23:10] you mean that you broke it? :-) [23:11] it was a side effect of your fix [23:11] Argh ;( [23:11] :( [23:11] confused me for a few minutes as I couldn't see how I had caused that error [23:11] let me guess, the reverse lookup doesn't work [23:11] easy to fix though, not a problem [23:11] phew [23:11] nope, in this case it crashes immediately on importint [23:11] sigh. [23:12] as you changed one place where it uses the name from launchpad, but not the other [23:12] ahhh! [23:12] so it can't find the thing that it just put in the database [23:12] Right! [23:12] Needs test cases. [23:12] we should be able to Mock Launchpad. [23:13] well, it just needs some xml [23:14] Hrm. That's right. [23:20] yay, my change works [23:21] \m/ [23:41] we need to figure out how to allow . in the url.py regex [23:48] why's that? [23:50] that's the reason we've been going around replacing . with _ or - everywhere, because it breaks the reverse url lookups [23:51] well, this is replacing it in the same place :-) [23:51] but if there are still remaining places then that would be good [23:51] though we should do it soon, otherwise it will cause dataloss