15:00:41 #startmeeting XenAPI 15:00:42 Meeting started Wed Nov 12 15:00:41 2014 UTC and is due to finish in 60 minutes. The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:43 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:46 The meeting name has been set to 'xenapi' 15:00:50 hello all 15:01:08 for people who are around, do you want to keep the same agenda? 15:01:48 matel: do you know if BobBall can make it today? 15:02:49 OK, not sure we have people around for the meeting today 15:02:59 I can restart it later if people turn up 15:03:04 hellp 15:03:17 matel: ? 15:03:23 yep, here. 15:03:52 ah, hello, that makes sense 15:04:03 Howdy! 15:04:09 Sorry for being late 15:04:11 What did I miss? 15:04:15 nothing 15:04:20 #topic XenServer CI 15:04:23 Oh that's good. 15:04:25 johnthetubaguy: agenda of the meeting is this, right..http://eavesdrop.openstack.org/meetings/docteam/2014/docteam.2014-11-12-14.01.txt 15:04:32 so, hows things in CI world? 15:04:40 so, no nova/cell discussion.. 15:04:45 VineetMenon: this is the XenAPI meeing 15:04:46 #link http://f0ab4dc4366795c303a0-8fd069087bab3f263c7f9ddd524fce42.r22.cf1.rackcdn.com/ci_status/recent_finished.txt 15:04:51 oh ..k 15:04:52 sorry 15:04:58 VineetMenon: no worries 15:05:17 Credits to Mate. He's been keeping the CI going wonderfully lately after fixing it 15:05:20 VineetMenon: maybe the time changes got you 15:05:31 matel: good stuff 15:05:45 how are we doing from memory these days? 15:05:46 That was BobBall 15:05:49 did that get fixed 15:05:59 hmmm 15:06:00 memory 15:06:01 good question 15:06:05 let me log in to a host 15:06:25 Did we have issues with memory recently? 15:06:34 Sorry, I might have missed it. 15:06:45 it was slowing things down 15:06:47 KiB Mem: 6153288 total, 5261644 used, 891644 free, 189560 buffers 15:06:50 KiB Swap: 4193276 total, 0 used, 4193276 free. 2044800 cached Mem 15:06:51 Was it domzero or domU 15:06:55 excessive paging 15:06:57 DomU 15:06:58 domU 15:07:07 Happy happy joy joy memory usage there 15:07:12 cools 15:07:12 I'm happy with 0 swap 15:07:16 We have 800MB free 15:07:47 funny, was super bad before, I guess we dropped some services somewhere 15:07:59 yes 15:08:05 we increased ram to 6GB 15:08:17 6G consumed by OS - that's lovely. 15:08:20 and as a followup to my devstack change to limit the number of API nodes 15:08:26 the default is now set to 2 by devstack 15:08:35 because infra were seeing similar issues 15:08:53 ah, cool 15:08:55 that would do it 15:08:58 there were lots of apis 15:09:06 Indeed 15:09:20 so, skipped tests 15:09:29 I guess we can't reduce that list again? 15:09:34 Probably can 15:09:37 but it's quite small ATM 15:10:05 http://git.openstack.org/cgit/stackforge/xenapi-os-testing/tree/tempest_exclusion_list 15:10:25 would be good to get those back on if we can 15:10:35 assuming they are reliable now 15:11:04 It's an interresting assumption 15:11:09 What's that assumption based on? :) 15:11:19 well that gate is not totally hosed 15:11:28 sure, not many things running 15:11:41 would like to know if its a XenAPI specific thing, or a general thing, I guess 15:11:49 Ah, yes, that's a good point 15:11:58 happy to do some work on making volume more stable if it shows issues, or least I should be able to find someone 15:11:58 I believe the memory issue was affecting volume tests with timeouts 15:12:19 oh, that would make total sense 15:12:26 anyways, good to push for an empty list 15:12:36 but better that its actually running well! 15:12:43 so moving on... 15:12:48 #topic Open Discussion 15:12:53 anything else 15:13:07 I don't think much was raised at the summit that would surprise us 15:13:22 No real surprises, no 15:13:31 I kinda hope to get people working on cells + XenServer tests, but I need to find the people from somewhere 15:13:34 https://review.openstack.org/#/c/133975/ 15:13:36 FYI# 15:13:38 -# 15:13:56 BobBall: cool 15:14:06 Not quite sure what to do though 15:14:23 just recheck 10 times ;) 15:14:33 replace 10 with some number that makes you feel happy 15:14:43 anyways... 15:14:53 Perhaps... 15:14:59 I would rather run it on a separate CI. 15:15:00 any more for any more? 15:15:09 Seperate CI matel? 15:15:10 hmmm 15:15:20 I wonder if that'd be a good test for your seperate nodepool install 15:15:25 matel: cells would sure be a separate CI setup, if thats what you mean 15:15:36 BobBall - yep 15:15:38 matel's been working on using upstream nodepool 15:15:46 cool 15:15:51 There are various bugfixes upstream that we need to take 15:16:03 johnthetubaguy - I was talking about the null-exclusion list 15:16:05 BTW, johnthetubaguy, did you get a creedence t-shirt at the summit?? 15:16:31 BobBall: didn't get any t-shirts, didn't stay in that bit long enough, and I don't really like queues 15:16:54 While the t-shirts did go very quickly we weren't mean enough to make people queue up for them. 15:17:13 I'll see if we can find a spare one for you. Medium? 15:17:22 so I did go to the stand, on the first day, before the keynote, but there was no one there 15:17:30 Large if possible, I suspect they went 15:17:46 You want me to send you one over there in Hungary Mate? 15:17:52 BobBall: would be nice 15:17:54 L - will do. 15:18:07 L is fine for me. 15:18:08 BobBall: I can collect mine at some point anyways 15:18:19 Yes, we weren't there before the keynotes - I actually didn't arrive until after Monday's keynotes :/ 15:18:22 Stupid easyjet 15:18:23 BobBall: I can send you some wine for that 15:18:29 BobBall: doh 15:18:38 anyways 15:18:40 Wine in exchange for a t-shirt? 15:18:41 I guess we are done 15:18:44 thanks all 15:18:48 nice to see a green CI 15:18:48 Let me send you a whole case of t-shirts then matel. 15:18:57 BobBall: cool 15:19:04 case of wine is OK if you are out of t-shirts 15:19:14 I'm sure I'll find a t-shirt john. 15:19:18 Nice try. 15:19:20 :) 15:19:22 But yes, think we're done. 15:19:25 #endmeeting