[12:09] bac benji frankban, my wife had to go to something this morning for our boys' school, so I'm left with the children. I'm trying to get something done, but the baby is being...loud. Would you like call with loud baby and echo in 2 minutes, or call without baby and echo in, say, 20 or 30 minutes? [12:09] T+30 [12:09] agreed [12:10] frankban, that ok with you? [12:10] yes gary_poster [12:10] if G+ hangouts can put mustaches on people can't they invent a baby scream filter? [12:10] :-) [12:10] cool, thanks all. I'll ping a few minutes before the call [12:33] bac benji frankban call in 7, at 20 till [12:33] k [12:40] joinng [12:40] ^i^ [13:06] benji: ready whenever you are [13:07] bac: the horde awaits [13:31] frankban, https://talkgadget.google.com/hangouts/_/extras/canonical.com/goldenhordeoneonone [13:31] ? [13:31] joining [13:32] frankban, also, I approved your branch with two very trivial thouhts [13:32] g [13:37] bac fell off the internet [13:43] bac: I lost you again [13:43] ok [14:29] bac: have you chosen a new card yet? [14:29] benji: no [14:30] bac: well, with my new card the active lane limit has been reached, so... you'll have to look elsewhere I guess :\ [14:31] benji: the one you picked looks familiar. you think it is the same handler problem? [14:32] it might be similar [14:56] bac: ha! I just applied the patch from 992692 to 992339 and it fixed it [15:00] Now that I've totally gummed up the active lane, I guess I'll go get a snack and work on a slack card. [15:18] benji: why not make 992339 a dupe of 992692 and clear the card from the lane? [15:18] bac: I suppose that would work. I'll do so. [15:21] benji: when the lane is clear i'm going to look at bug 993510 -- i think it is the same [15:21] <_mup_> Bug #993510: lp.services.job.tests.test_runner.TestJobRunner.test_runJobHandleErrors_oops_generated_user_notify_fails fails intermittently/rarely in parallel tests < https://launchpad.net/bugs/993510 > [15:22] yay [15:23] * gary_poster is updating ResultsLog page [15:23] and will be very curious to see the fix [15:25] benji: over here -- yay [15:29] benji: yep, dupe [15:30] bac: cool [15:30] (now in the right channel) [15:30] gary_poster: here is benji's fix that addresses four cards (so far): http://pastebin.ubuntu.com/964869/ [15:31] gary_poster: do you want to keep those dupe cards and move to done-done or delete them [15:31] cool bac, that's awesome. delete them I guess [15:32] I moved mine to done-done, so I'll delete it now. [15:34] We'll only be left with the (problematic) bug 992814 and bug 992184 then, I think... [15:34] <_mup_> Bug #992814: lib/lp/services/webservice/doc/launchpadlib.txt fails intermittently/rarely in parallel tests < https://launchpad.net/bugs/992814 > [15:34] <_mup_> Bug #992184: lib/lp/services/database/doc/textsearching.txt fails intermittently/rarely on parallel tests < https://launchpad.net/bugs/992184 > [15:35] bug 993467 too, probably easier [15:35] <_mup_> Bug #993467: lib/lp/bugs/browser/tests/distrosourcepackage-bug-views.txt fails when run alone (on lxc?) < https://launchpad.net/bugs/993467 > [15:39] * bac ->sashimi lunch special s/octopus/salmon/ [15:39] tmi? [15:48] :-) [15:48] nah, sounds good [15:59] gary_poster, Just thought you'd like to know that even using S3 and its massive pipes, doing the LP setup on the fly in our juju charm is just too slow for my use-case. I'm going to revert to the specific AMI model and blow a big raspberry at The Right Way of Doing Things. [16:03] gmb, well, glad you found out with enough time to do something about it :-) [16:03] gary_poster, As am I :). In other news, I've submitted my ec2 expenses for last month. I'd forgotten that it's expense deadline tomorrow... [16:03] gmb, and if you have the energy to cope with the assumed push back, it would be noble of you to report this to the juju team. I would cheer. From the sidelines. Quietly. [16:04] gary_poster, Agreed. I'm going to get it working and then write an email to the juju dev list. [16:04] great, thanks gmb. I'll go look at expenses right now. [16:07] done, gmb [16:10] frankban, I approved https://code.launchpad.net/~frankban/launchpad/bug-987904-intermittent-failure/+merge/104577 with a small comment [16:13] thanks gary_poster, also I submitted aws expenses [16:16] cool, frankban, approved. [16:17] benji, dragnob has now deleted your old incorrect vacation request, and I approved the newer, correct one. [16:17] gary_poster: cool, thanks [16:17] welcome [16:20] I'll have to be more careful in the future. Those mistakes live a long time. [16:29] lunch [18:30] I would have said to benji, [18:30] hey benji! [18:30] https://talkgadget.google.com/hangouts/_/extras/canonical.com/goldenhordeoneonone when you are ready! [18:31] ...but he is not here... [18:36] bac: I'm not sure if my last message made it through (slight network glitch here): our test run had 100% test success [18:36] benji: we can do bettuh [18:37] actually, benji, that's pretty great [18:37] :) [18:38] I'm working up the MP now. [18:38] benji hi. our call was 6 minutes ago. I told you about it on IRC, even though I knew you were not here. a tree-falls-in-a-forest-but-no-one-is-there-to-hear moment. Do you want to ping me when the MP is done? [18:38] gary_poster: nah, now is good [18:38] cool [18:38] https://talkgadget.google.com/hangouts/_/extras/canonical.com/goldenhordeoneonone [18:56] gary_poster: the connection has gone bad [19:29] bac, https://talkgadget.google.com/hangouts/_/extras/canonical.com/goldenhordeoneonone when you are ready. no rush [19:30] ok [19:30] just finishing up a mp [20:24] bac, this fixes bug 994158. Is it good enough, or is there a more elegant way to do this? http://pastebin.ubuntu.com/965478/ [20:24] <_mup_> Bug #994158: lp.registry.browser.tests.test_distroseriesdifference_views.DistroSeriesDifferenceTestCase.test_binary_summaries_for_source_pub fails intermittently/rarely in parallel tests with distroreleasepackagecache_pkey error < https://launchpad.net/bugs/994158 > [20:25] gary_poster: i have seen frankban and graham do similar fixes. [20:25] bac, ok. [20:31] bac, you ok with approving https://code.launchpad.net/~gary/launchpad/bug994158/+merge/104618 then? [20:31] done [20:32] gary_poster: in a non quid pro quo would you review https://code.launchpad.net/~bac/launchpad/bug-987499/+merge/104607 ? [20:33] bac, :-) sure [20:36] bac, so the change to lib/lp/app/browser/tests/test_stringformatter.py is no longer necessary (or even really used, since no one calls _setDeveloper(False) now), right? You thought it was still an improvement because it made the _setDeveloper call clearer in its meaning? [20:36] gary_poster: yep. [20:36] cool [20:38] approved bac, thank you [21:03] well, lookie there: [21:03] File "lib/lp/services/database/doc/textsearching.txt", line 579, in textsearching.txt [21:03] Failed example: [21:03] nl_term_candidates('firefox foo-bar give me trouble') [21:03] Differences (ndiff with -expected +actual): [21:03] - [u'firefox', u'foo', u'bar', u'foobar', u'give', u'troubl'] [21:03] ? - ^ --- ^^^ [21:03] + [u'ive', u'bl', u'bar', u'foobar', u'give', u'troubl'] [21:03] ? ^ ^^ [21:36] benji, how'd you do it? [21:37] gary_poster: I wrote a script to peg all the CPUs and do lots of memory access and then ran the test in a loop. [21:37] I then turned off the load script and the test still fails intermittently. :) [21:38] so -- at least on my box -- just running the script repeatedly will generate failures fairly frequently [21:39] s/script/test/ [21:39] great, benji!