15:01:21 <johnthetubaguy> #startmeeting XenAPI
15:01:21 <openstack> Meeting started Wed Mar 19 15:01:21 2014 UTC and is due to finish in 60 minutes.  The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:23 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:24 <BobBall> o/
15:01:25 <openstack> The meeting name has been set to 'xenapi'
15:01:28 <johnthetubaguy> howdy people
15:01:30 <johnthetubaguy> hows things?
15:01:32 <BobBall> Just me today again I think!
15:01:41 <leifz> I'm lurking.
15:01:44 <BobBall> Well, I've got an orange and a can of diet coke
15:01:54 <BobBall> so that's got to be classified as "good" right?
15:02:03 <johnthetubaguy> yup
15:02:12 <johnthetubaguy> any specific topics?
15:02:22 <johnthetubaguy> its almost time to gather summit like thoughts
15:02:22 <BobBall> CI is broken :(
15:02:30 <johnthetubaguy> oh yeah
15:02:33 <johnthetubaguy> #topic CI
15:02:49 <johnthetubaguy> the CI died :( nasty new features
15:02:57 <BobBall> I'm going to blame Xavier Queralt for fixing a bug in tempest
15:03:02 <BobBall> https://review.openstack.org/#/c/78194/
15:03:07 <johnthetubaguy> so, should we look at getting my change in htere?
15:03:16 <BobBall> Broke XenAPI and we urgently need John's change and mine
15:03:26 <johnthetubaguy> can we please exclude that test for now?
15:03:31 <johnthetubaguy> while we get this fixed?
15:03:32 <BobBall> #link https://review.openstack.org/#/c/81497/1
15:03:38 <BobBall> I suppose we can now we know what's wrong
15:03:42 <BobBall> let me exclude it now... hang on...
15:04:02 <johnthetubaguy> you have tabs in there by the way
15:04:20 <BobBall> I just uploaded one without the tabs??
15:04:22 <BobBall> damn :P
15:05:15 <BobBall> excluded.
15:05:19 <BobBall> Next tempest runs won't try it
15:05:38 <johnthetubaguy> cool
15:05:41 <BobBall> I suppose I should re-run the failed jobs
15:05:45 <johnthetubaguy> yeah
15:05:50 <BobBall> What do you think we should do there?
15:05:59 <BobBall> re-running failed jobs will push them all on the queue and slow down results for new jobs
15:06:05 <BobBall> until the queue is drained
15:06:30 <johnthetubaguy> sweet
15:06:38 <johnthetubaguy> so… fixing it
15:06:39 <BobBall> so should I re-run them and delay results for new tests?
15:06:46 <johnthetubaguy> seems like its those two changes
15:07:10 <BobBall> yes
15:07:17 <BobBall> what's you change?
15:07:18 <johnthetubaguy> lets link them here
15:07:23 <BobBall> oh - I linked to patch #1 of mine
15:07:27 <BobBall> that's why you saw tabs :)
15:07:36 <BobBall> #link https://review.openstack.org/#/c/81497/5
15:07:50 <BobBall> #link https://review.openstack.org/#/c/81252/
15:08:01 <BobBall> johnthetubaguy: can you update yours to say partial-bug rather than closes-bug?
15:08:23 <johnthetubaguy> erm, well it closes the bug from a nova perspective, probably just need to reword the bug
15:08:39 <BobBall> I've added devstack to the list
15:08:44 <BobBall> in the bug
15:09:08 <johnthetubaguy> yeah, OK, so its closes-bug then? from a nova point of view?
15:09:54 <BobBall> yes, but closes-bug assumes that it's the only part of the bug
15:09:56 <BobBall> which it isn't
15:10:02 <BobBall> which is why both changes need partial-bug
15:10:29 <johnthetubaguy> erm… but it would be close form Nova, its independent for each project, so it leaves devstack open till it closes
15:10:42 <johnthetubaguy> there isn't another nova patch, as far as we know, is what I mean
15:10:55 <BobBall> Oh, ok.  I thought that closes-bug be the whole bug
15:11:12 <johnthetubaguy> only on a per project, seems independent
15:11:20 <BobBall> fine
15:11:24 <johnthetubaguy> cool
15:11:51 <BobBall> Then can you remove your -1 and poke Matt until he agrees so we can get it in?
15:12:27 <johnthetubaguy> I can try, will ask the other cores over here to take a look at least
15:12:38 <BobBall> I meant agrees to remove his -1
15:12:47 <johnthetubaguy> right
15:12:59 <johnthetubaguy> so, we should be good, assuming that all works
15:13:01 <johnthetubaguy> ?
15:13:24 <BobBall> Well I've already excluded the test
15:13:30 <BobBall> so that should mean we're ok from that perspective
15:13:57 <johnthetubaguy> true true
15:15:23 <BobBall> I've now re-queued all of the failed tests in the last 24 hours
15:15:45 <johnthetubaguy> nice, thank you
15:15:48 <johnthetubaguy> any more on CI?
15:16:17 <BobBall> nah guess not
15:16:34 <johnthetubaguy> #topic Juno Summit
15:17:22 <johnthetubaguy> anyone got things to raise this time?
15:17:31 <johnthetubaguy> we should start and etherpad again...
15:17:34 <BobBall> no - too early
15:17:50 <BobBall> too many other things going on ATM - can't think about Juno!
15:17:58 <johnthetubaguy> OK
15:18:15 <johnthetubaguy> #topic Open Discussion
15:18:18 <johnthetubaguy> Any other issues?
15:18:25 <johnthetubaguy> any rc1 looking bugs?
15:18:32 <BobBall> no, don't think so
15:18:38 <BobBall> I think we're generally good ATM
15:19:33 <johnthetubaguy> likewise from what I can see at the moment
15:19:57 <johnthetubaguy> are we all done then?
15:20:17 <BobBall> think so
15:20:28 <BobBall> oh
15:20:33 <BobBall> will probably have to rebuild CI soon
15:20:38 <johnthetubaguy> why is that?
15:20:42 <BobBall> I'd built it with the smallest flavor thinking "it's not doing much"
15:20:48 <johnthetubaguy> ah
15:20:49 <BobBall> but it's run out of the 512M ram ;)
15:20:57 <BobBall> heavily into swap now
15:21:02 <johnthetubaguy> is it standard flavor?
15:21:18 <BobBall> don't remember
15:21:28 <johnthetubaguy> snapshot then rebuild, or resize it if its standard
15:21:39 <johnthetubaguy> its standard if its 512, so just resize it
15:22:07 <BobBall> seems that nodepool is using 50% of the memory(!) pretty huge
15:22:14 <johnthetubaguy> wow
15:22:19 <BobBall> I've got a script that'll rebuild it
15:22:25 <BobBall> so the only thing we'd lose is queue state
15:22:32 <BobBall> unless I update it to copy the DB cross rebuilds
15:22:48 <BobBall> in which case we'd just lose gerrit stream watching for 30 minutes or so
15:22:48 <johnthetubaguy> yeah, but snapshot then rebuild and you don't loose the queue right?
15:22:51 <johnthetubaguy> or just resize
15:23:11 <johnthetubaguy> anyways, good to test your script!
15:23:25 <johnthetubaguy> assuming its in a git repo somewhere
15:23:25 <BobBall> dunno what I'll do - I'll figure it out.
15:23:34 <johnthetubaguy> cool, any more for any more?
15:23:51 <johnthetubaguy> (note resize can be quite slow)
15:24:18 <johnthetubaguy> thanks bob
15:24:22 <johnthetubaguy> #endmeeting