14:02:35 #startmeeting tripleo 14:02:35 #topic agenda 14:02:35 * Review past action items 14:02:35 * One off agenda items 14:02:35 * Squad status 14:02:36 * Bugs & Blueprints 14:02:37 Meeting started Tue Feb 4 14:02:35 2020 UTC and is due to finish in 60 minutes. The chair is weshay|ruck. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:02:38 * Projects releases or stable backports 14:02:38 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:02:40 * Specs 14:02:40 The meeting name has been set to 'tripleo' 14:02:42 * open discussion 14:02:44 Anyone can use the #link, #action and #info commands, not just the moderatorǃ 14:02:46 Hey folks! who's around? 14:02:47 «o/ 14:02:47 o/ 14:02:48 o/ 14:02:53 o/ 14:02:55 \o 14:02:58 Tengu: too slow 14:02:58 o/ 14:03:04 o/ 14:03:06 o// 14:03:06 hi 14:03:07 EmilienM: I was first according my screen 14:03:15 o/ 14:03:18 \o/ 14:03:25 EmilienM: #laggingHard :) 14:03:25 o/ 14:03:26 o/ 14:03:27 -o- 14:03:45 o/ 14:04:02 o/ 14:04:24 o/ 14:04:30 hey hey.. big crowd!! 14:04:36 nice tripleo!! 14:05:30 reminder.. agenda is here https://etherpad.openstack.org/p/tripleo-meeting-items 14:05:46 * Tengu hides 14:06:05 last week we discussed pinning ansible and how to test it prior to a bump.. discussion is on-going 14:06:28 #topic one off agenda items 14:06:29 #link https://etherpad.openstack.org/p/tripleo-meeting-items 14:06:39 ok.. Tengu you are up 14:06:49 o/ 14:06:54 Emilien Macchi proposed openstack/tripleo-heat-templates master: Add support to run Container Puppet tasks without Paunch https://review.opendev.org/702323 14:06:56 sooo yeah. CI breakage, validations and all. 14:07:25 there's an ongoing discussions about the best place|way to use validations within CI. Something was introduced last year, and it hit hard last week in CI. 14:07:43 to be fair.. it was not just validations 14:08:01 so the current thoughts are more: "use validations within promotion pipeline", and then get a dedicated job for the validation framework itself. non-voting, of course. 14:08:12 weshay|ruck: right - but it didn't help solving the issue :) 14:08:19 it's always something.. but protecting the stability of the gate jobs is important.. 14:08:24 aye 14:08:37 there's a proposal here 14:08:41 #link https://hackmd.io/3mNnVaVHS-SE793vRtmiAA?view 14:08:41 https://hackmd.io/3mNnVaVHS-SE793vRtmiAA 14:08:57 and it would be nice to get some feedback and discussions around all of that. 14:09:18 We're pretty sure validations have their place in the CI, at some point, and might point to actual issues, especially once we get more validations :). 14:09:42 Tengu, the bit that's important to raise here is.. https://hackmd.io/3mNnVaVHS-SE793vRtmiAA#Define-a-single-scenario-and-standalone-to-test-the-validation-framework-itself 14:09:56 adding a scenario and specific job just for validations 14:10:04 a standalone scenario and job.. 14:10:09 yeah, right - this is basically asking for a dedicated job, with dedicated|custom scenario and so on for the validations. 14:10:15 this would be the only job upstream where validations are enabled 14:10:28 outside of the promotion pipeline, yes. 14:10:33 looking for objections, comments on that thought before proceeding 14:10:59 promotions and component are technically 3rd party 14:11:04 yup 14:11:04 but use the upstream bits 14:11:25 the goal is to still catch issues, but not block gate :) 14:11:32 exactly 14:12:06 the dedicated job itself will also allow to test the framework on its own. 14:12:19 if/when they are reliable why not block the date? 14:12:21 gate 14:13:30 owalsh: well, it was stable and reliable, but a tiny change in the healthchecks made it mad. 14:13:51 owalsh, when the validations or spread throught multiple jobs.. it's difficult to manage and validations can inconsistenctly pass/fail based on non-tripleo code 14:13:54 and since it was voting, in addition to another issue, it blocked gate for what, 2, 3 days? not really good. 14:14:35 tripleo is responsible for the tripleo deployment bits.. not ensuring neutron services validate correctly 14:14:38 fact is, validation is a sensitive topic, and can indeed raise issues due to some 3rd party, like port change, dead service due to whatever reason and so on 14:14:53 but if it's not blocking something is anybody going to look at it? 14:14:55 more in depth validation and testing is now possible in the compoment pipeline 14:15:11 owalsh: if it blocks the promotion pipeline, it will be worked out. 14:15:20 owalsh, each project will be informed when their validations fail 14:15:26 ok 14:15:31 and via the new component thingy, like weshay|ruck said, it would be visible as well 14:15:55 additionally in this case.. validations were turned on for upstream check / gate W/O being enabled in the promotion jobs 14:16:02 that is a really bad thing 14:16:29 the latest non-tripleo bits were not tested w/ validations prior to hitting our gates 14:17:01 so.. if you have more questions or thoughts.. let's add them to the etherpad https://hackmd.io/3mNnVaVHS-SE793vRtmiAA 14:17:10 you can comment on the side of each section 14:17:16 Tengu, thanks for your help on this 14:17:29 weshay|ruck: thank you for your patience and explanations ;). 14:17:31 ah.. one more thing.. forgot to add to the agenda 14:17:43 Proposal to restructure the tripleo squads 14:18:01 #link https://etherpad.openstack.org/p/tripleo-ci-squad-meeting 14:18:01 upgrade 14:18:01 #link https://etherpad.openstack.org/p/tripleo-upgrade-squad-status 14:18:01 edge 14:18:01 #link https://etherpad.openstack.org/p/tripleo-edge-squad-status 14:18:01 integration 14:18:03 #link https://etherpad.openstack.org/p/tripleo-integration-squad-status 14:18:05 validations 14:18:07 #link https://etherpad.openstack.org/p/tripleo-validations-squad-status 14:18:09 networking 14:18:11 #link https://etherpad.openstack.org/p/tripleo-networking-squad-status 14:18:13 transformation 14:18:15 #link https://etherpad.openstack.org/p/tripleo-ansible-agenda 14:18:17 these are OLD 14:18:26 I think we need upgrades, mistral and tripleo-ansible 14:18:29 ci maybe 14:19:16 so validations is still on their own - and since we get our weekly right after this #tripleo mtg, it's hard to push a status update in there :( 14:19:24 not sure what to do with that one, to be fair... 14:19:24 if we narrow the squads down to the work that is being done.. and expect status from each.. 14:19:58 My expectation of squads is at the very least.. they are coming to this meeting w/ a list of reviews that need attention 14:20:20 meh.. I get that list something like 15 minutes after this mtg XD 14:20:41 k.. so I think I'll email openstack-discuss[tripleo] w/ this 14:20:44 will check if I can move it to some earlier time, should be possible. 14:20:55 :) 14:21:36 or maybe I can just link the pad used for the notes. Will add a point for today's call :] 14:21:38 so I think I'll call out folks like Tengu, d0ugal, cloudnull, for a list of reviews and a little summary 14:21:56 aye.. we'll figure it out.. just wanted to plant the seed atm 14:21:58 Reviews: https://review.opendev.org/#/q/topic:mistral-removal+OR+topic:mistral_to_ansible 14:22:13 we need more people reviewing code.. this meeting can be more focused on that 14:22:27 it is hard to give a little summary, because there is so much going on :) but basically there is quite a bit of progress going on across the board 14:22:44 aye.. 14:22:47 Emilien Macchi proposed openstack/tripleo-heat-templates master: Add support to run Container Puppet tasks without Paunch https://review.opendev.org/702323 14:23:08 it's worth the experiment.. and adjusting in time 14:23:20 d0ugal: :] same for the validations. I'll check what I can do, point added to my mtg. 14:23:28 weshay|ruck: -^ 14:23:40 weshay|ruck sorry in dualing meetings. 14:23:41 thanks 14:23:41 weshay|ruck: the squad status is good place for review requests but generally speaking it is better/more effective if folks speak up during this meeting to highlight a series that needs reviews 14:23:59 agree 14:24:21 weshay|ruck, the conversion for node import to ansible seems to be making introspection timeout. I'm unsure of why and investigating. 14:24:34 ok.. so expect a follow up via the email list 14:24:44 introspection.. always a hoot 14:24:51 but we'd love review on the following https://review.opendev.org/#/q/status:open+label:verified%253D%252B1%252Cuser%253Dzuul+topic:mistral_to_ansible 14:25:09 all of those are mistral_to_ansible patches that are passing zuul checks and ready for review 14:25:51 thanks all :) Please help review!!! 14:25:54 cloudnull: is there some story behind that series its a lot of patches are they related somehow? 14:25:56 20min a day at least :) 14:25:59 moving on 14:26:03 https://releases.openstack.org/ussuri/schedule.html 14:26:13 Ussuri is ml2 next week 14:26:15 they're all part of our tripleoclient conversion epic 14:26:21 https://storyboard.openstack.org/#!/story/2007212 14:26:24 cloudnull: thanks 14:26:27 one more milestone to get your features in 14:26:31 we've been tracking them all in storyboard 14:26:41 and we still don't have CentOS 8 :( 14:26:44 * weshay|ruck cries 14:27:06 #link https://launchpad.net/tripleo/+milestone/ussuri-2 14:27:18 Blueprints: 14:27:18 1 Unknown, 1 Started, 1 Good progress 14:27:18 Bugs: 14:27:18 5 New, 19 Incomplete, 4 Invalid, 3 Won't Fix, 5 Confirmed, 361 Triaged, 117 In Progress, 76 Fix Released 14:29:06 latest new specs 14:29:06 cloudnull: hey i can help here https://storyboard.openstack.org/#!/story/2007212 I have wrote some of them (for update in particular) 14:29:06 https://review.opendev.org/#/c/698828/ 14:29:25 ekultails, has a new one.. please look ( reminder again ) 14:29:37 #topic open discussion 14:29:37 Anything else that folks want to bring up to the meeting? 14:29:48 matbu absolutely ! 14:30:08 please feel free to take from the board 14:30:47 * weshay|ruck notes.. gate is red.. https://bugs.launchpad.net/tripleo/+bug/1861803 14:30:47 Launchpad bug 1861803 in tripleo "ubuntu-bionic | ERROR: Could not find a version that satisfies the requirement oslo.concurrency===4.0.0 (from -c /home/zuul/src/opendev.org/openstack/requirements/upper-constraints.txt " [Critical,In progress] - Assigned to wes hayutin (weshayutin) 14:30:58 upper constraints bit us this morning.. patches in progress 14:31:06 anything else folks? 14:31:09 cloudnull: ack i will do 14:31:22 if you have questions please ping us 14:31:32 happy to help in any way we can 14:31:35 going once 14:31:40 going twice 14:31:44 #endmeeting tripleo