[21:24] <davecheney> thumper: here is an example http://paste.ubuntu.com/11716589/
[21:24]  * thumper looks
[21:43] <thumper> davecheney: I have been noticing many intermittent failures with cmd/jujud/agent
[21:44] <thumper> davecheney: running that package tests with the race detector has indicated a number of data races
[21:44] <thumper> davecheney: just wanted to check you weren't looking at that package now
[21:49]  * thumper headdesks
[22:09] <thumper> wallyworld: love a chat when you turn up
[22:20] <wallyworld> ok, will ping you after another coffee
[22:34] <wallyworld> thumper: free now in 1:1 if you want
[22:35] <thumper> coming
[23:07] <thumper> wallyworld: http://reviews.vapour.ws/r/1925/
[23:07] <thumper> wallyworld: wasn't able to fix the intermittent failure with this
[23:07] <thumper> but it does fix the races
[23:08] <thumper> menn0: if you are curious later, waitForUpgradeToFinish in cmd/jujud/agent/upgrade_test fails intermittently for some weird reason
[23:09] <menn0> thumper: do you have any details?
[23:10] <thumper> menn0: nope... only that I had it fail here, then pass 8 times in a row
[23:10] <thumper> menn0: this test -> UpgradeSuite.TestUpgradeStepsHostMachine
[23:10] <thumper> menn0: just fails with false is not true
[23:10] <wallyworld> ok, will look aftr standup
[23:11] <thumper> wallyworld: cheers
[23:11] <menn0> thumper: I believe there's a ticket for that one already
[23:11] <thumper> yeah
[23:11] <menn0> thumper: that test actually predates me but I've worked a lot with that area so i'll take a look
[23:14] <menn0> thumper: bug 1444576 is the closest match
[23:14] <mup> Bug #1444576: Skipped TestUpgradeSteps* in cmd/jujud/agent/upgrade_test.go <skipped-test> <test-failure> <juju-core:Triaged by menno.smits> <juju-core 1.24:Triaged by menno.smits> <https://launchpad.net/bugs/1444576>
[23:15] <menn0> thumper: i've just grabbed it although it'll be a friday afternoon one
[23:15] <thumper> kk
[23:44] <thumper> menn0: re: http://reviews.vapour.ws/r/1924/diff/# you were mentioning that you don't actually have to close the iterators, is that right?
[23:46] <menn0> thumper: yeah it's not critical, the servers kills iterators after 10mins of inactivity anywayu
[23:46] <thumper> but the code is fine, right?
[23:46] <menn0> thumper: the code you reviewed is fine
[23:46]  * thumper -> gym
[23:47] <menn0> thumper: closing the session to force the blocked next call to unblock is not
[23:47] <menn0> thumper: depending where mgo is at, it might panci
[23:51] <davecheney> thumper: menn0 i'll look at jujud/agent next when I'm done with this PR