14:00:31 #startmeeting tripleo 14:00:31 Meeting started Tue Jun 20 14:00:31 2017 UTC and is due to finish in 60 minutes. The chair is EmilienM. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:33 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:35 The meeting name has been set to 'tripleo' 14:00:41 #topic agenda 14:00:46 * review past action items 14:00:48 * one off agenda items 14:00:50 * bugs 14:00:52 * Projects releases or stable backports 14:00:54 * CI 14:00:54 o/ 14:00:56 * Specs 14:00:58 * open discussion 14:01:00 Anyone can use the #link, #action and #info commands, not just the moderatorǃ 14:01:02 hey folks, who is around today? 14:01:07 o/ 14:01:11 o/ 14:01:13 /o 14:01:17 o/ 14:01:29 o\ 14:01:40 hi 14:01:45 \o (first timer) 14:02:21 hjensas: if it's your first time, you have to fight. 14:02:22 let's start 14:02:28 #topic review past action items 14:02:36 workflow team to review https://review.openstack.org/#/q/topic:bp/stop-using-mistral-env 14:02:38 o/ 14:02:42 o/ 14:02:49 sounds like everything merged, so DONE 14:02:57 follow-up next meeting about OVB check gates moving to third party 14:03:16 o/ 14:03:19 we'll probably talk about that during CI topic 14:03:27 follow-up next meeting about tripleo periodic jobs will be moving to rdo infra 14:03:28 ditto 14:03:32 team to give feedback on http://lists.openstack.org/pipermail/openstack-dev/2017-June/118157.html 14:03:51 o/ 14:03:54 o/ 14:03:54 mwhahaha: ^ did you have enough feedback on this thread? 14:04:10 0/ 14:04:12 EmilienM to send email about how to use RSS to monitor Tempest results in TripleO CI: POSTPONED 14:04:16 i haven't seen any comments on my last patches 14:04:17 o/ 14:04:17 #action EmilienM to send email about how to use RSS to monitor Tempest results in TripleO CI 14:04:21 about moving the bits to tripleo-common 14:04:35 other than that i think i'm good 14:04:41 o/ 14:04:46 o/ 14:05:11 mwhahaha: ok, moving forward then 14:05:13 #topic one off agenda items 14:05:25 #link https://etherpad.openstack.org/p/tripleo-meeting-items 14:05:38 weshay: sir, floor is now yours 14:06:22 EmilienM, ah.. well good news. I believe the issue is resolved by an update to lon's email tool :) 14:06:27 * weshay crosses it out 14:06:42 mhh ok 14:06:51 hjensas: please go ahead! 14:07:35 sounds like you want to add coverage to avoid having https://bugs.launchpad.net/tripleo/+bug/1699026 next time 14:07:36 Launchpad bug 1699026 in tripleo "Composable roles in tht master - 'dict object' has no attribute 'lower' error" [High,In progress] - Assigned to Harald Jensås (harald-jensas) 14:07:36 EmilienM: thanks, well not much to discuss. But looks like the roles jinja is'nt tested in CI. 14:08:14 hjensas: do you mean custom roles? 14:08:17 yes 14:08:26 it is tested, by 3-nodes jobs 14:08:34 which has been broken for some days, maybe it's the reason 14:09:00 but gate-tripleo-ci-centos-7-3nodes-multinode-nv is already covering custom roles 14:09:08 it deploys 2 controllers 14:09:16 and one of them is a custom controller with some services 14:09:38 hm, the commit I belive cause the issue was from april. 14:09:55 one role is Controller and another, custom is ControllerApi 14:10:10 hjensas: FYI gate-tripleo-ci-centos-7-3nodes-multinode-nv isn't voting 14:10:21 so it's highly possible that this patch in April broke the job 14:10:57 hjensas: it's weird, it passed CI: https://review.openstack.org/#/c/456739/ 14:11:18 hjensas: /me doing recheck on https://review.openstack.org/#/c/475750/ - we had git failures this morning 14:11:30 hjensas: anything we need to discuss in this meeting? 14:11:59 no, feel free to move on. 14:12:17 ok 14:12:21 #topic bugs 14:12:25 #link https://launchpad.net/tripleo/+milestone/pike-3 14:12:40 do we have any bug to discuss this week? 14:13:23 i've proposed on the mailing-list to make some triage for https://bugs.launchpad.net/tripleo/+bugs?field.tag=deployment-time 14:13:35 we have a Performance sprint, starting tomorrow and we need some inputs 14:13:50 if you're aware about bugs in tripleo that affects the time of deployment, please tag the bug as deployment-time 14:14:19 if there is no outstanding bug to discuss, let's move on 14:14:30 #topic projects releases or stable backports 14:14:49 no updates on release management on my side - do we have any question or feedback? 14:15:27 #topic CI 14:15:37 ok so we have 2 topics at least, regarding CI 14:16:02 OVB check gates moving to third party 14:16:15 pabelanger, sshnaidm, adarazs, weshay ^: what is the progress here? 14:16:48 maybe just a quick status or summary for the team, it doesn't need to be long (we already have the tripleo ci squad meeting) 14:16:55 EmilienM, so the infrastructure is not ready yet for a move, dmsimard is point for details afaik 14:17:20 for moving tripleo periodic jobs to rdo-infra ? 14:17:25 * dmsimard missing context 14:17:25 yes 14:17:38 periodic and also gate OVB 14:17:48 * adarazs is not directly involved in setting stuff up at this point. 14:18:00 Hoping to be able to build upstream images with nodepool elements out of project-config from review.rdo sometime today or tomorrow 14:18:10 dmsimard, apevec said it's not ready until you say its ready :) 14:18:31 I meant to ask and was redirected to weshay about quota -- we don't currently carry enough quota in our current tenants to accomodate for tripleo based jobs 14:18:33 ok, sounds like we'll have more updates for our CI meeting on Thursday, where anyone is welcome to join btw 14:18:45 dmsimard, similiar but diff topic 14:19:06 weshay: ok let's sync off-meeting if you want but I need to know where those jobs will end up running 14:19:19 dmsimard, sure thing.. and thanks 14:19:45 anything else about CI? 14:19:51 just BAH 14:19:54 lol 14:20:01 sshnaidm: I saw multinode jobs are back 14:20:06 woot 14:20:06 the git sync issue is fixed I guess 14:20:11 I wanted to ask if anyone has bandwidth to help convert the 3nodes job to oooq 14:20:11 see http://tripleo.org/cistatus.html 14:20:31 I started looking at it, but I'm not clear how we're supposed to populate the inventory so the bootstrap happens 14:20:49 if any quickstart experts have time to help, I'd appreciate it 14:20:50 panda: was it something in the transition backlog? If yes, what was the priority? ^ 14:21:01 context is, we want a containers 3node HA/updates job 14:21:24 EmilienM: it was in the backlogs, but the priority is not high AFAIK 14:21:24 so first step is make quickstart do undercloud + 3nodes vs undercloud +1 14:21:41 but lending a hand is possible I think 14:21:49 EmilienM, we also should be running the undercloud containers oooq job at this point, /me getting on the to-do list 14:21:53 Master promotion is lagging 4 days behind right now, is that under control ? 14:22:15 dmsimard, it promoted on saturday I think 14:22:28 Yeah, 4 days ago :) https://dashboards.rdoproject.org/rdo-dev 14:22:32 I would say it's under control 14:23:02 yeah i think we had some tripleo-ci patches recently, from sshnaidm to fix some issues 14:23:06 hopefully it's okay. yesterday ovb-ha-oooq failed, today it passed, today ovb-updates failed and ovb-ha-oooq passed... so probably we're not in too bad shape. 14:23:09 that were blocking promotion I guess 14:23:22 dmsimard: https://review.openstack.org/#/c/475715/ 14:23:27 https://review.openstack.org/#/c/475805/ 14:24:07 EmilienM, yeah, the last were submitted today for stable branches, we used there "current" instead of "consistent" 14:24:33 ok 14:25:27 shardy, panda: maybe we could try to work in pair to make this 3-nodes thing, to transfer a bit of knowledge 14:25:40 shardy: how's the priority for the 3-nodes job to be transitioned for you ? The person that developed the multinode parts left, so it will take some time for anyone to help 14:25:56 oops :( 14:26:22 panda: well we're about to land the final patches for containers+HA, so we really need CI coverage ASAP 14:26:30 panda: I can try to have a look this week 14:26:40 I'll probably ask for some review / help 14:26:44 hence I started looking at it, but I'm struggling due to lack of quickstart knowledge 14:26:46 but I volunteer to start the investigation 14:26:59 the alternative is I just write the test in the old scripts, because I know them pretty well 14:27:09 EmilienM, panda is pretty slamed atm w/ getting the containers check gate to work 14:27:09 shardy: let's sync on this thing later and maybe share progress / ask questions on the ML if support is needed 14:27:15 then we switch both 3nodes jobs to quickstart later 14:27:24 weshay: yeah no worries. I'll start looking 14:27:48 #action EmilienM to start looking how to convert 3-nodes jobs in oooq 14:28:07 * EmilienM has no idea in what he's jumping in 14:28:07 shardy, EmilienM I can try to help there as well 14:28:19 cool thanks 14:28:25 weshay, EmilienM: ack thanks, lets figure out how to work together on it, thanks! 14:28:28 do we have anything else CI related? 14:28:35 just BAH! 14:28:36 EmilienM: piece of cake! 14:28:40 I understand how to get the containers pieces running, and can propose a new featureset for that 14:28:54 it's the bootstrap multiple nodes for deployed-server I'm unsure about atm 14:28:56 shardy: yeah, i'll look the multinode part 14:29:15 shardy: let's ketchup later on topic 14:29:24 +1 14:29:32 #topic specs 14:29:35 #link https://review.openstack.org/#/q/project:openstack/tripleo-specs+status:open 14:30:05 do we have any spec to discuss this week? 14:30:32 I still some some specs for Pike 14:30:38 it's great to see that people are confident :D 14:30:56 we should probably postpone the specs to Queens at this stage. 14:31:29 #action EmilienM to postpone open specs to Queens, unless code has already been started during Pike and made good progress 14:31:32 any feedback on this? 14:32:01 #topic open discussion 14:32:18 reminder: tomorrow we have a 2-days sprint to improve TripleO deployment time 14:32:41 https://etherpad.openstack.org/p/tripleo-deploy-time-hack 14:32:57 I haven't seen much interest to be very honest 14:33:47 if you're interested to participate, please write your name in the etherpad 14:33:48 maybe we're too close to release ? 14:34:06 panda: pike-3 is in 1 month 14:34:16 we're in the middle of the cycle (a bit after) 14:34:18 People might not be sure they have enough skills/knowledge to contribute maybe... (<- me). Will be following along though :) 14:34:31 people finish starting and start finishing ... 14:34:37 jpich: I was expecting the people familiar with it, trying to help a little 14:34:48 anyway 14:34:51 Fair enough! 14:34:54 fyi, the etherpad: https://etherpad.openstack.org/p/tripleo-deploy-time-hack 14:35:02 is there anything else for this week? 14:35:21 thanks everyone 14:35:26 have a great week and have fun 14:35:28 #endmeeting