15:02:11 #startmeeting puppet-openstack 15:02:11 Meeting started Tue Nov 29 15:02:11 2016 UTC and is due to finish in 60 minutes. The chair is mwhahaha. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:02:12 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:02:14 The meeting name has been set to 'puppet_openstack' 15:02:24 #link https://etherpad.openstack.org/p/puppet-openstack-weekly-meeting-20161129 15:02:36 looks like a fairly empty ajenda today 15:02:38 who's here? 15:02:40 o/ 15:02:44 o/ 15:02:46 hi 15:02:54 o/ 15:03:59 cool. let's go 15:04:11 #topic Review past action items 15:04:23 EmilienM to collect openstack deprecations and file launchpad bugs 15:04:54 * mwhahaha pokes EmilienM 15:05:04 probably postpone that until next time again 15:05:19 EmilienM sync with RDO about puppet upgrade for packstack/tripleo 15:05:36 i think it's still blocked in packaging, i'll try and dig up the BZ today and check on the status 15:05:44 EmilienM to investigate scenario001/002 timeouts 15:06:25 I think that was worked on a bit by dmsimard, and we've reduced the amount of tests we actually run 15:06:32 hello 15:07:02 mwhahaha: afik puppet was updated in RDO 15:07:10 it's in buildlogs.centos.org/centos/7/cloud/x86_64/openstack-ocata/ 15:07:12 so we've got puppet4 no? 15:07:17 yes 15:07:18 now 15:07:20 cool 15:07:23 awesome 15:07:23 but I need to check CI jobs 15:07:24 =D 15:07:27 if it pick it up 15:07:44 sounds good 15:07:52 mwhahaha, I did some scans of neutron logs a week or so ago and didn't come up with anything. I'll keep at it though as I'll be exercising many variants of neutron deployments over the next couple of weeks 15:08:05 beagles: cool thanks 15:08:16 beagles: we had a warning in neutron, what was it again? 15:09:16 EmilienM, lots of oslo warnings... I don't recall something specific to neutron. I'll go back through in case I'm missing it somehow 15:09:29 mwhahaha to propose a virtual midcycle for next year - so i'm going to pick some days in January for next year and send a mail out today 15:09:30 beagles: oh right 15:10:11 mwhahaha to sync with openstack release team about metadata.json updates - postponed as I didn't get a chance to do this last week with the holiday 15:10:16 anyway that's it on past items 15:10:25 #topic Open Discussion, Bug and Review triage 15:10:32 anyone have anything they want to talk about today? 15:10:41 yes 15:10:53 I would like to propose that we convert run_tests.sh into ansible 15:10:56 for 2 reasons 15:11:13 1) zuul v3 is coming soon, and having it in ansible will make our transition to zuul v3 easy 15:11:30 2) we could use ARA https://github.com/dmsimard/ara 15:11:46 which would help us to understand and troubleshoot our CI 15:12:16 I don't have time to work on it now, but I wanted to propose it 15:12:30 and I think dmsimard liked the idea and proposed to start 15:12:37 I think it might be beneficial 15:12:52 probably can reduce some of the complexities we have as well 15:13:20 o/ 15:13:28 dmsimard: we were talking about ara 15:13:36 and also translating run_tests.sh in ansible 15:14:04 dmsimard: do you think we can prototype something sometimes soon? 15:14:28 Yeah. 15:14:33 This week is good enough ? 15:14:37 lol 15:14:40 tonight ! 15:14:52 joking, yes of course, it's when you can 15:14:56 I have other things on my plate right now, like making scenario004 gate RDO promotion that still hasn't landed? :P 15:15:01 you can bootstrap it and we can help you of course 15:15:30 Yeah, sure. I can iterate -- like translate copy-logs.sh first and so on 15:15:53 The question mark for me 15:15:54 up to you, I would start by run_tests.sh maybe, and split in tasks 15:16:01 Is still the long catalog runs 15:16:06 Where ara loses a bit of value 15:16:25 and I don't think we want to run each manifest (project) individually to have that level of granularity 15:16:36 no we don't do that 15:16:42 we run twice 15:16:52 and if we could use the puppet ansible module it would be awesome 15:17:05 I already contributed and used it, it's pretty good 15:17:12 yeah, the puppet module is fine 15:17:15 what is your concern with puppet catalog? 15:17:34 but what I meant is that this is still one long running task with lots of output -- if you see it failed, you still need to sift through the longer amount of output 15:17:55 I have part of the solution for that, I still need to think about how to implement it properly 15:18:14 the idea is not to split the puppet catalog but the tasks in run_tests.sh 15:18:21 I know :) 15:18:22 a task would be "first puppet run" 15:18:35 and if it fails, people will still have to dig in the catalog logs to find out why it fails 15:18:42 right 15:18:48 but I want to make that easier 15:18:52 baby steps 15:18:53 that's the whole point 15:18:55 but yeah 15:18:58 give me a week 15:19:05 you get 2 days :D 15:19:13 15:19:16 I'll try and get an ara release out, the new version will have a feature I want 15:19:19 for now, just bootstrap the tasks, to have feature parity 15:19:33 we'll see puppet catalog later 15:19:38 sure 15:19:42 mwhahaha: i'm done 15:19:49 cool 15:19:55 anyone else have anything they want to talk about? 15:20:04 still snowing here :( 15:20:31 * mwhahaha mails EmilienM a shovel 15:20:43 snow, psh 15:20:45 I have icy rain 15:20:48 * beagles grumbles about having nothing but rain 15:20:53 yeah *icy* rain 15:21:00 * beagles grumbles more 15:21:09 can't ski on that 15:22:03 alright, if no one has anything else we'll call it. 15:22:10 Thanks everyone. see you next week 15:22:12 #endmeeting