15:01:52 <johnthetubaguy> #startmeeting XenAPI
15:01:54 <openstack> Meeting started Wed Aug  6 15:01:52 2014 UTC and is due to finish in 60 minutes.  The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:55 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:58 <openstack> The meeting name has been set to 'xenapi'
15:02:11 <BobBall> o/~
15:02:13 <BobBall> hmmm
15:02:22 <BobBall> That looks better in some fonts than others
15:02:27 <BobBall> it's meant to be me waving :/
15:02:59 <johnthetubaguy> lol, hello
15:03:02 <johnthetubaguy> o/
15:03:36 <BobBall> yes yes - that's just holding your hand up
15:03:38 <BobBall> I can do that too
15:03:55 <BobBall> also I can do the \o/ <o> (o /o\
15:04:04 <BobBall> but that's probably beside the point
15:04:11 <BobBall> and should be saved for AOB at the end
15:04:15 <johnthetubaguy> #topic XenServer CI
15:04:27 <johnthetubaguy> so hows life in the CI world?
15:04:36 <BobBall> There seems to be a new race / timeout introduced
15:04:49 <johnthetubaguy> oh goody, xenserver specific one?
15:04:53 <BobBall> a whole bunch of the boto third party ec2 tests started failing the last few days
15:04:58 <johnthetubaguy> oh
15:05:08 <BobBall> I've not found references to it outside of xenapi - so it could be configuration specific
15:05:16 <johnthetubaguy> those have been dodgy in the past
15:05:17 <BobBall> but highly unlikely to be XenAPI specific
15:05:30 <BobBall> Anyway - they hammered the pass rate
15:05:31 <johnthetubaguy> yeah, I guess its possible, but seems unlikely
15:05:36 <johnthetubaguy> gotcha
15:05:38 <BobBall> so over the last few days I've disabled most of them
15:05:45 <BobBall> should be back to a sensible pass rate now
15:05:46 <johnthetubaguy> hows the skip list looking these days?
15:05:57 <johnthetubaguy> tempest test skip list that is
15:06:06 <BobBall> but the last set I only added today so it'll take a while to show up
15:06:19 <BobBall> #link http://git.openstack.org/cgit/stackforge/xenapi-os-testing/tree/tempest_exclusion_list
15:06:28 <BobBall> quite good in comparison to the original list
15:06:33 <johnthetubaguy> Ok, cools, so its not massive, yeah
15:06:47 <johnthetubaguy> thats quite a nice small set of tests to try and debug and improve
15:06:48 <BobBall> volumes stuff and ec2 compatability stuff
15:07:07 <BobBall> yeah - but the frustration is that I haven't been able to tickle them in the right way outside of the CI yet
15:07:36 <BobBall> and because everyone expects the third party CIs to be "better than jenkins" (my interpretation) then we can't rely on it
15:08:39 <BobBall> to qualify the "better than jenkins" expectation; if there is a random failure in a 3rd party CI my understanding is most people ignore the CI rather than think there is a race that needs to be tracked down; as such we have to pass significantly more of the time otherwise the CI would just fall into a permanently ignored state
15:08:45 * BobBall doesn't like disabling tests :/
15:09:37 <johnthetubaguy> well, to be fair, they expect it to be as good as jenkins, and jenkins runs like four tempest tests per run, etc, etc
15:10:12 <johnthetubaguy> need easier way to dev these kinds of test improvements I guess
15:10:24 <BobBall> That's perhaps a discussion for another time
15:10:44 <johnthetubaguy> yeah
15:10:55 <johnthetubaguy> any more on CI, looking like most things are stable-ish?
15:11:24 <johnthetubaguy> #link http://www.rcbops.com/gerrit/reports/nova-cireport.html is looking quite good still
15:11:41 <johnthetubaguy> not totally awesome, but thats the way it goes
15:11:51 <johnthetubaguy> #topic Open Discussion
15:12:00 <johnthetubaguy> so, we got anything else for today?
15:12:04 <BobBall> I haven't seen the report recently
15:12:08 <BobBall> ah yes
15:12:14 <BobBall> quite a few failures - many more than I want
15:12:27 <BobBall> we got to a failure rate of about 1 in 4 before I added the exclusions
15:12:49 <johnthetubaguy> yeah, I think that was the state at the last meeting
15:12:57 <BobBall> I'm pleased that we're still holding our own in terms of 3rd party CIs though
15:13:05 <johnthetubaguy> so glad that got stabalised, I am guessing the red bar is still from when that happened
15:13:11 <BobBall> yup
15:13:22 <johnthetubaguy> yeah, the other are much worse at this point, although that is causing its own issues
15:13:39 <johnthetubaguy> I have a bug fix up for some extra clean up with snapshots: https://review.openstack.org/#/c/100880
15:14:10 <BobBall> although I worry that those stats aren't being updated again
15:14:29 <johnthetubaguy> yeah, there was talk of moving it into infra at somepoint soon
15:14:42 <BobBall> for XenServer CI both the first and last 'Missed' votes are more than 30 days old
15:14:50 <BobBall> so I figure the stats are frozen at some point in the past
15:14:56 <johnthetubaguy> hmm, that does smell bad
15:14:57 <BobBall> (despite the 'updated' date
15:14:59 <BobBall> +)
15:15:54 <johnthetubaguy> anyways, yeah, that bug fix would be good to get some eyes on it, but other than that, not much to report
15:16:16 <johnthetubaguy> any more for today?
15:16:28 <BobBall> nah not from me
15:16:38 <johnthetubaguy> cool, non from me
15:16:38 <BobBall> oh sure - why not
15:16:43 <BobBall> fix for neutron has gone up
15:16:46 <johnthetubaguy> lets gain some time back for the rest of the day
15:16:55 <johnthetubaguy> ah, interesting, it was broken then
15:17:07 <BobBall> https://review.openstack.org/#/c/112018/
15:18:18 <johnthetubaguy> cools
15:18:27 <johnthetubaguy> thanks BobBall
15:18:30 <johnthetubaguy> #endmeeting