15:00:32 #startmeeting third-party 15:00:33 Meeting started Mon Sep 7 15:00:32 2015 UTC and is due to finish in 60 minutes. The chair is anteaya. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:34 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:36 The meeting name has been set to 'third_party' 15:00:40 hello 15:01:54 anteaya: hi 15:02:51 hey lennyb 15:02:55 how are you? 15:03:35 anteaya: thank you, just came back from vacation. How are you? 15:03:52 how lovely 15:03:56 I am well thank you 15:04:02 Hello! 15:04:02 so far it is just you and I 15:04:06 and eantyshev 15:04:07 hello 15:04:15 what shall we discuss today? 15:04:32 Well.. I have nothing interesting today to ask or tell 15:04:47 okay thanks for saying so lennyb 15:04:47 git.o.o had problems today, as I noticed 15:04:56 did it? 15:05:36 we had failures merging patches, cloning repositories etc. 15:05:58 all due to connection failures to git.o.o 15:06:01 have you a paste? 15:07:29 http://paste.openstack.org/show/448824/ 15:08:29 and it had been happening from 9:30 to 14:07 UTC 15:08:50 #link http://cacti.openstack.org/cacti/graph_view.php?action=tree&tree_id=2 15:09:02 that seems to match the load spike on the graph 15:09:41 did you mention it in the -infra channel? 15:10:31 no, I saw nobody there and didn't try 15:10:48 there seems to be twice daily spikes: http://cacti.openstack.org/cacti/graph_image.php?action=view&local_graph_id=2566&rra_id=2 15:11:11 I'm going to guess that is in line with image creation 15:11:48 and that some third party cis offset a bit from infra's image creation but don't offset that much 15:11:58 that is my guess 15:12:46 I beleive that 15:13:19 so the solution would be to get ci operators co-ordinated so that everyone picked a different time for image creation 15:13:38 so they are spaced out from each other, not all lumped together 15:14:09 I've modified mine, to match my timezone 15:14:14 awesome 15:14:31 I wish every ci operator paid as much attention to their system as you do 15:14:36 thank you eantyshev 15:15:11 perhaps, I've customized it even further than I should 15:15:21 you never know until you try 15:15:40 you can post your findings to the -infra channel so we know you are having git failures 15:15:55 even if noone is in the channel you can post your findings and your paste 15:16:09 then they have it as a data point 15:16:29 all the git servers were upgraded to centos7 form centos6 last week 15:16:29 Does it make sense to do now? 15:16:40 so we do want to hear your experience 15:16:43 yes 15:16:57 post the failure message and the time frame 15:17:08 clarkb will want to hear 15:17:17 he will read the logs when is online tomorrow 15:17:20 then I'll do it after our meeting 15:17:25 thank you 15:17:40 other than git.o.o not behaving as expected today for you 15:17:51 do we have anything else we would like to discuss? 15:18:19 I have a small request, for somebody from core reviewers to look at https://review.openstack.org/#/c/215029/ 15:18:33 that change which we discussed a week ago 15:18:47 to support third-party OSes in devstack-gate 15:19:07 right, I'm not a devstack-gate reviewer 15:19:17 so I can't help much in that regard 15:19:22 two suggestions 15:19:25 1) patience 15:19:37 2) add it to the infra meeting agenda for discussion 15:20:58 okay, I haven't tried 2) 15:21:18 #link https://review.openstack.org/#/c/215029/ 15:21:26 you can if you wish 15:21:40 that is the only suggestion I have on this patch 15:22:41 Thank you, that's all I wanted to discuss 15:22:49 great thank you eantyshev 15:23:03 unless we have anything else, I'll close the meeting for today 15:23:08 ... 15:23:22 stop me if I should wait for anything 15:23:37 * anteaya waits to see if anyone wants to stop her 15:23:49 thanks for your attendance and kind participation today 15:24:11 I hope to see you next week, and for some at 0800 utc Tuesday 15:24:14 thank you 15:24:17 #endmeeting