15:00:45 #startmeeting XenAPI 15:00:47 Meeting started Wed Apr 2 15:00:45 2014 UTC and is due to finish in 60 minutes. The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:48 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:50 The meeting name has been set to 'xenapi' 15:01:00 anyone got anything to raise this week? 15:01:09 no 15:01:12 short meeting! :D 15:01:16 quick CI update and quick chat about the summit maybe? 15:01:23 #topic CI 15:01:24 So the CI was horribly broken 15:01:29 for many reasons 15:01:34 mostly I blame trove ;) 15:01:40 lol 15:01:42 and now? 15:01:55 was added to devstack but using "localhost" as the service endpoint for nova/swift/smt_else 15:02:03 oh dear 15:02:12 but localhost didn't work in my RAX image due to some weird IPv6 breakage where it was resolving as IPv6 15:02:26 and trove only supports IPv4 through their use if AF_INET... 15:02:34 wow thats what it was, funky 15:02:35 So... there was a seriously dodgy message for a long time 15:02:51 But it only showed up as a couple of tempest failures 15:02:55 which was very annoying 15:02:59 hopefully that's fixed now 15:03:12 other issues were encountered such as the hypervisor extension being broken 15:03:17 OK, crazy, but sounds like that should be fixed with any luck 15:03:22 #link https://review.openstack.org/#/c/84138/ 15:03:32 but I've excluded the test for now, so the CI isn't blocked on it 15:03:45 I'm very frustrated with the way the CI is excluding tests 15:03:51 because it's independent of gerrit 15:04:05 which means there is no easy way of trialing re-inclusions 15:04:18 hmm, yes, thats anoying 15:04:25 also there is no way for anyone else in OpenStack to say "I want this test, but it's going to break XenAPI so we'll exclude it from that CI" 15:04:43 no - they just commit the test because it passes gate and then the XenAPI CI gets broken for _everyone_ because not all tests can pass 15:05:14 See https://review.openstack.org/#/c/81565/ where I proposed that this exclusion list should live in tempest 15:05:17 but that's not the right way forward 15:05:23 hmm, tricky 15:05:25 I just haven't had time to look at the right way forward due to firefighting 15:05:35 nothing should break XenAPI of course, or its not a valid test 15:05:48 Well, we might be doing things incorrectly 15:05:56 For example, we don't list which extensions we support 15:06:12 right, its something we should generally fix 15:06:15 if someone adds an extension to Nova and then the test in Tempest in theory that should be enough information to not run it for XenCI but to run it in gate 15:06:41 but as I said - I'm currently on fire patrol, so that's not even been looked at yet 15:06:53 right, needs some thought 15:07:16 no thought - just code :D 15:07:38 sureā€¦. 15:08:54 so... 15:08:57 #topic Summit 15:09:07 any major blueprints for XenAPI in Juno? 15:09:08 yay 15:09:12 looking forward to the summit 15:09:20 Uncertain at this point 15:09:33 do we need a XenAPI session? probably not I guess? 15:09:59 I'm not convinced there would be much value to having one 15:10:03 but I'm only one person 15:10:10 we usually over run in that session, but its often not that useful 15:10:13 I don't know whether others think that it's useful :) 15:10:21 the problem is it does reduce the visibility 15:10:30 and its good to get everyone in the same room 15:10:59 That would be unfortunate - but without a productive agenda I'm not sure that there would be benefit to doing that vs arranging a meetup over a pint 15:11:30 right, just need to make sure it happens, can I leave that to you to round people up? 15:11:40 I can help, just I will probably forget 15:12:18 Sure - why not. 15:12:23 cool 15:12:31 #topic Open Discussion 15:12:35 any more for any more? 15:12:44 I'm not here next week 15:12:44 :) 15:12:53 going anywhere nice? 15:13:05 Nah - just on vacation at home 15:13:14 ah well, that can be cool to 15:13:18 enjoy your week off 15:13:26 thanks bob 15:13:29 #endmeeting