15:01:34 #startmeeting XenAPI 15:01:35 Meeting started Wed Apr 9 15:01:34 2014 UTC and is due to finish in 60 minutes. The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:37 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:39 The meeting name has been set to 'xenapi' 15:01:53 Just me 15:01:54 Who is around for today's get together? 15:02:00 ^^ 15:02:11 hehe, you answered my question before I asked it 15:02:15 #topic CI 15:02:15 And I don't have much to say because I've been on vacation all this week! 15:02:31 :) 15:02:35 Well the CI was broken 15:02:37 and it's sort of fixed 15:02:43 but steadily going down hill... 15:02:49 because I have to keep disabling tests 15:02:52 which is frustrating 15:02:55 yeah, needs some attention 15:03:11 but I've got too many fires ATM which means the only way to get the CI back up and running is to fix it short term 15:03:13 so many new tests keep breaking when they are added? 15:03:27 right, gotcha 15:03:32 https://github.com/citrix-openstack/xenapi-os-testing/commits/master last few commits have just been to add more exclusions... 15:03:40 #help need more effort to maintain the XenServer CI 15:03:42 which is bad 15:03:45 agreed 15:03:50 It's not really maintaining the CI 15:03:53 just fixing bugs 15:04:03 I expect that if RS were running full tempest they would hit these too 15:04:10 and see "XenAPI doesn't work in XYZ scenario" 15:04:42 right, probably fixing the current failures would sort out the others 15:04:55 Of course, we don't even have bugs for most of them... which would be helpful itself... 15:05:06 anyways, seems like a good goal for Juno 15:05:20 try to get as much of full tempest going as possible, adding features and fixing bugs on the eay? 15:05:23 way^ 15:05:43 I suspect the easiest fix would be to decrease what "full tempest" is 15:05:49 by saying we don't support various extensions etc 15:05:58 in theory tempest should then skip tests rather than failing them 15:06:07 from what I understand 15:06:51 right, so thats also an API issue, would be good to discover what is actually supported 15:06:57 but anyways, I think I get your point 15:07:06 That might not be my full point ;) 15:07:08 but we'll see 15:07:08 have supported extensions, then let tempest use those 15:07:12 there is also a frustration 15:07:29 once we fix an issue - such as 1294069 - there is no easy way of verifying it's fixed for the CI 15:07:44 because the exclusion list is in https://github.com/citrix-openstack/xenapi-os-testing/ 15:07:49 which is not monitored by gerrit / CI 15:08:01 I wonder if there is a dumping ground we could put it 15:08:07 stackforge sounds like a dumping ground? 15:08:10 Is that possible? 15:08:46 I guess so 15:08:57 I wonder if we can move the whole of xenapi-os-testing to stackforge... 15:09:18 maybe, just I know they don't want everyone to use that 15:09:41 Do you know why / what can use it? 15:09:43 but would be handy to have gerrit etc 15:09:58 there was some eligibility, let me check 15:10:11 its mostly just a resources, number of people to fix stuff thing 15:10:15 well there is http://ci.openstack.org/stackforge.html 15:11:52 BobBall: thats the one 15:11:54 doh 15:12:03 maintain related unofficial projects 15:12:07 that sounds cool 15:12:20 Sounds like what we want 15:12:30 I'll talk to mate and the -infra guys about it 15:12:48 we've got a couple of repos that could go there; but I guess only xenapi-os-testing needs to be there for some kind of gerrit control 15:12:48 cool 15:13:04 would be nice if its all in the same place, but yeah 15:13:15 Well, that's the one that actually "does stuff" 15:13:21 the rest is just management infrastructure 15:13:50 sure, but makes fixing easier if its all in the same place 15:13:57 but anyways, sounds good 15:14:07 #topic Open Discussion 15:14:13 any thing else for today? 15:14:15 well hopefully the CI stuff isn't going to last forever 15:14:29 no, nothing from me. 15:15:36 not forever? 15:15:47 you mean get it into the upstream infra stuff? 15:15:55 indeed 15:16:15 it might still need xenapi-os-testing but wouldn't need any other projects AFAIR 15:16:41 gotcha, thats a good line to draw then 15:16:47 so, shall we call it a day? 15:17:22 well - any updates from you or anyone else at RAX? 15:17:45 nope, not really, nothing XenServer specific at the moment 15:18:11 OK 15:18:12 call it then 15:18:28 currently planning on focusing on performance and stability 15:18:33 personally 15:18:48 so tempest is certainly a good tool to help there 15:19:12 OK 15:19:41 anyways, lets catch up more on reducing the tempest exclusion list, just setting up an on-demand build of a working environment would be really handy 15:20:00 anyways, thanks BobBall catch you next week 15:20:02 I think the first step is moving xenapi-os-testing to stackforge if we can 15:20:03 #endmeeting