=== Spads_ is now known as Spads === bodie__ is now known as bodie_ === nottrobin_ is now known as nottrobin === cppforlife__ is now known as cppforlife_ [21:28] what is the godeps flag to look at the current deps? [21:31] nm [21:31] * thumper is running all the tests [21:32] exorcise complete in master [21:32] exorcism [21:36] 34 files, +370 −691 [21:36] not a bad change [21:36] http://reviews.vapour.ws/r/5103/ [21:45] hmm [21:45] I spotted a few things myself in that review [21:45] * thumper goes to tweak... [21:55] * thumper runs up to physio [21:55] bbs [22:37] wallyworld: ping [22:44] davecheney: review above removes fslock from master [22:44] * thumper starts on 1.25 backport [23:08] thumper: hey [23:11] thumper: this adds access to the minion reports to the migrationmaster apiserver facade: http://reviews.vapour.ws/r/5105/ [23:15] wallyworld: you are reviewer, and I have a review :) [23:15] so i see [23:18] * thumper sighs... [23:35] thumper: ta [23:36] thumper: in RunTestSuite, why pass in a wall clock instead of a mock? is the thinking that the delay is "only" 250ms? why not use testing.ShortDelay? [23:36] wallyworld: I was trying to reduce the work to get the code in [23:36] really it isn't going to matter because the lock isn't held [23:37] so the delay never occurs [23:37] could change to testing.ShortDelay [23:37] should have zero impact [23:37] yes please [23:37] for consistency [23:37] with out tests around the place [23:37] *other [23:37] add comments [23:37] I'll go through them [23:38] if lock is never held, isn't that a gap in the tetst? [23:38] there are some places where we test for held locks [23:39] in the important places [23:39] well, most places that actually use the lock [23:40] ok [23:46] thumper: looks pretty close to me [23:57] wallyworld: while you finish looking, I'll head out to lunch and address both davecheney's comments and yours after lunch [23:57] ok