14:00:06 #startmeeting tripleo 14:00:07 Meeting started Tue Jun 8 14:00:06 2021 UTC and is due to finish in 60 minutes. The chair is marios. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:10 The meeting name has been set to 'tripleo' 14:00:20 #topic agenda 14:00:20 * Review last minutes & action items 14:00:20 * One off agenda items 14:00:20 * Bugs & Blueprints 14:00:20 * Projects releases or stable backports 14:00:22 * Specs 14:00:25 * open discussion 14:00:27 Anyone can use the #link, #action and #info commands, not just the moderatorǃ 14:00:29 marios: wow, that's a on-time start. Even a Swiss couldn't do better :D 14:00:30 hello OFTC #tripleo o/ 14:00:40 who is around today ? o/ 14:00:40 hello 14:00:48 hi 14:00:54 o/ 14:01:00 o/ 14:01:42 great thanks all for joining ... lets get started and folks will catch up as usual ;) 14:01:55 #topic review last meeting logs & action items 14:02:07 #link http://eavesdrop.openstack.org/meetings/tripleo/2021/tripleo.2021-05-25-14.00.html 14:02:19 anyone have something to bring up from last meeting? now is the time 14:02:22 * marios checks logs 14:02:35 hola 14:03:00 0/ 14:03:05 o/ 14:03:11 i think the biggest item of discussion last time was the proposal for moving to independent and and stein/queens etc but we have an item for that today as well 14:03:20 lets move on t 14:03:21 o/ 14:03:30 #topic one off agenda items 14:03:35 #link https://etherpad.openstack.org/p/tripleo-meeting-items 14:03:42 o/ 14:03:44 taking them in order as usual 14:03:51 Tengu: please go ahead when you're ready 14:03:52 oh, meaning that's for me 14:04:12 So, I have 2 open specs for review, and would love to get some more feedback, or even merge them 14:04:18 #link https://review.opendev.org/c/openstack/tripleo-specs/+/733652 14:04:30 that one is about improving logging withint tripleo deployment (and, well, overall CLI) 14:04:41 I've addressed latest comment by cloudnull - soooo a new review would be nice 14:04:45 did we discuss the ansible logging one at PTG /me doesn't recall 14:04:53 weeee did, yes 14:04:53 (not that it matters just wondering) 14:05:00 Tengu: k thanks i didn't review that one yet 14:05:01 Simon Dodsley proposed openstack/tripleo-heat-templates master: Add Pure Storage FlashBlade Manila driver https://review.opendev.org/c/openstack/tripleo-heat-templates/+/792850 14:05:04 there was an on-going PoC 14:05:18 0/ 14:05:21 marios: it's an old spec actually, but since it didn't merge in the previous release... 14:05:22 Tengu: ah it was posted in 2020 14:05:23 o/ 14:05:26 :) 14:05:31 exactly. So a revival. 14:05:33 or reboot. 14:05:35 or whatever. 14:05:41 Tengu: so fwiw i think we can/should merge the healthcheck one 14:05:43 we've also made changes since it was originally proposed 14:05:45 #link https://review.opendev.org/c/openstack/tripleo-specs/+/787535 14:05:51 that's the second spec about healthchecks 14:05:57 Tengu: it has enough votes, it got some feedback and there is no -1 14:06:03 mwhahaha: yeah, I've adapted that one accordingly 14:06:14 I'm not certain the logging one should be done as we don't know if we'll continue to leverage ansible int eh same way 14:06:19 i think that one needs more discussion 14:06:27 marios: yep, guess we can merge the healthcheck one and drop some of the things. 14:06:27 anyone object to me merging https://review.opendev.org/c/openstack/tripleo-specs/+/787535 (healthcheck) ? 14:06:43 * marios counts to 10 14:06:51 mwhahaha: yep, that's also why I've pushed that topic to the meeting - get some attention, and open discussion. The logging is... well, it's complicated. 14:07:03 * cloudnull no objections 14:07:24 Tengu: k merged +/787535 14:07:34 well, worfklowed anyway 14:07:36 I'll prepare a patch disabling the healthchecks by default, that's a first step. 14:07:39 Merged openstack/tripleo-heat-templates master: Add OS::TripleO::UndercloudUpgradeEphemeralHeat https://review.opendev.org/c/openstack/tripleo-heat-templates/+/785297 14:07:40 marios: thanks :) 14:07:57 Tengu: do we need to discuss the logging one or some aspect of it here? 14:08:04 and, yeah, the one about logging would love some more feedback here: https://review.opendev.org/c/openstack/tripleo-specs/+/733652 14:08:27 marios: I didn't prepare things for a live discussion, but if there are concerns right now, we can take a moment? depends on the other topics. 14:08:37 or we can go with the comments, maybe better 14:08:55 Tengu: should we leave it to the open discussion, depending on time if there isn't something very specific to mention at this point? 14:09:08 marios: good idea. 14:09:31 OK thanks Tengu does anyone have any more questions comments concerns about the items raised by tengu? 14:09:42 Tengu, I just left one comment for healthcheck one but it is not a blocker 14:09:56 I'm ok to merge that now as it is 14:09:59 tkajinam: should i remove workflow? 14:10:01 * marios checks 14:10:03 sorry to be late 14:10:04 Simon Dodsley proposed openstack/puppet-tripleo master: Add Pure Storage FlashBlade backend https://review.opendev.org/c/openstack/puppet-tripleo/+/792849 14:10:05 marios, no 14:10:20 even if we'd change the plan then we can update the spec at that time 14:10:40 tkajinam: ok 14:11:13 so moving on if there is nothing else... holding for a few more seconds 14:11:23 tkajinam: guess your comment would be a "step 2" - thanks for raising this one. 14:12:07 We have quite a few community items: 14:12:14 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-June/022824.html * [TripleO] Proposing ysandeep for tripleo-ci core 14:12:14 Leif Madsen proposed openstack/tripleo-heat-templates master: Fix typo in ceilometer-write-qdr environment file https://review.opendev.org/c/openstack/tripleo-heat-templates/+/795345 14:12:21 congrats and thanks to ysandeep|ruck++ 14:12:43 i haven't seen any objections so we can add him to the core group now 14:12:43 thanks all ++ 14:12:57 Merged openstack/openstack-tempest-skiplist master: Add rhos-17 test skip where wallaby test is included https://review.opendev.org/c/openstack/openstack-tempest-skiplist/+/795327 14:13:00 #info http://lists.openstack.org/pipermail/openstack-discuss/2021-June/022859.html * [TripleO] tripleo repos going Extended Maintenance stable/train OK? (not yet IMO) 14:13:15 #info train is now in extended maintenance https://review.opendev.org/c/openstack/releases/+/790778 14:13:25 any comments on that ^^^ ? 14:13:34 (we didn't really have much choice ;) ) 14:13:46 and we can still merge things to it, it just means no more tagged release for train 14:14:06 #info proposal for making a list of folks in each of the tripleo squads so new contributors (or even old contributors ;)) can know who to add to reviews 14:14:09 tkajinam: ^^^ 14:14:38 this came from an email thread and the idea is to have some list of folks per squad so you know who to add to reviews 14:14:42 thanks to tkajinam for raising it 14:14:53 for example there https://docs.openstack.org/tripleo-docs/latest/contributor/index.html 14:14:53 that would be very useful for newcomers to the community ;) 14:14:57 ;-) 14:15:19 are there are comments about that? what do folks think? Will you agree to be listed there? 14:15:25 marios: iirc one can check project owner list, and then ping the relevant folks... That's available on gerrit already. 14:15:27 we can only make it work if folks agree to be included 14:16:19 Tengu: does it really work though? I am guessing one or two people are listed as owners of our repos (and not say the members of the current squad as it changes? ) 14:16:50 hmm, I'm not sure how to access an example with the new gerrit, but I think it's listing the groups of actual cores 14:16:54 Tengu, to ask very generic question I think that list works. but what I expected initially is something explaining a bit more about coverage of each person 14:16:55 which is already a nice thing 14:17:39 Tengo: yes, the issue is that we dont really know which cores know more about a specific area of tripleo 14:17:43 for example to find somebody who can answer questions specific to nova 14:18:08 Merged openstack/tripleo-specs master: Healthcheck cleaning and consolidation https://review.opendev.org/c/openstack/tripleo-specs/+/787535 14:18:31 tkajinam: jlarriba: so i think we can proceed by posting a review to add this into docs, and seeing what the reception about that is 14:18:33 Tengu, sorry i mispelled your name ^^ 14:18:51 tkajinam: jlarriba: for example, we can add everyone to the review and if anyone objects we remove their name 14:18:54 thoughts? 14:18:58 jlarriba: ^^ 14:19:04 marios: i think it is a good idea 14:19:13 well, it doesn't even have to be in tripleo-docs that is just my suggestion maybe there are others? 14:19:16 marios: that would make too much spam to the non-involved ppl... 14:19:41 Tengu: ;) well, we could just add them and not ask so they don't have any spam? ;) 14:19:54 Tengu: i am open to suggestions about how to proceed 14:20:03 gerrit notifications are... annoying at some point 14:20:34 but maybe it's really a matter of getting the list up-to-date in the project itself. though, apparently, the new gerrit masks this feature 14:20:47 still didn't find that link back. 14:21:04 * marios needs to wrap this up in a sec as we have a lot of items we can carry on in open discussion time allowing 14:21:07 Isn't there a way reviewer get added to reviewer list on a particular patch based on repo/files? 14:21:13 Tengu: I asked people for that and they werent able to find that list in gerrit either 14:21:34 ysandeep|ruck: not that i know of but that sounds useful ;) 14:21:44 https://review.opendev.org/admin/repos/openstack/nova,access 14:21:57 Browse -> Repositories 14:21:58 https://review.opendev.org/admin/groups/0319cee8020840a3016f46359b076fa6b6ea831a,members 14:22:09 and, yeah, there's Browse -> members 14:22:16 Browse -> groups sorry 14:22:32 In rdo gerrit.. i noticed some people get added automatically on some repos.. need to check how it works. 14:22:41 Tengu: when i tried tripleo-heat-templates i get Cannot parse URL as a Gitiles URL 14:22:47 gor instance, nova_core: https://review.opendev.org/admin/groups/54f6a1ec13b7453596635e8708f1b60bfd281ebd,members 14:22:47 Tengu: https://review.opendev.org/plugins/gitiles/openstack/tripleo-heat-templates/+log/7c03dc03868fe3adb9a11fa6f4921379c2a41857/project.config 14:23:02 Tengu: but the request isn't just about 'who works on tripleo-heat-templates' 14:23:09 Tengu: more like who is in each squad 14:23:16 Tengu: deployment, upgrades, storage, CI etc 14:23:37 yeah.. I'm just trying to find out if that info is already available, but apparently it's a nope. 14:23:39 Tengu: because tripleo-heat-templates is something we all work on for eaxmple 14:23:40 it is kind of the roster for tripleo-upstream, I guess 14:23:58 jbadiapa: exactly. 14:24:05 k so i haven' heard a counter proposal to 17:18 < marios> tkajinam: jlarriba: so i think we can proceed by posting a review to add this into docs, and seeing what the reception about that is 14:24:29 do we have a volunteer to lead this effort and post the review? jlarriba tkajinam ;) ^^ ? 14:25:20 marios: yes, no problem I can post the review (after some out-of-the-meeting) conversation ;) 14:25:27 thank you jlarriba 14:25:44 marios, jlarriba I can help as well.. but thinking of something more flexible like etherpad 14:25:47 rather than docs 14:25:56 we can talk about that later 14:25:57 #info jlarriba to post review to tripleo-docs so we can start the discussion about that and see who wants to be included 14:26:03 thank you jlarriba tkajinam 14:26:15 moving on now in a second ... holding for any final comments 14:26:16 tkanijam lets discuss later 14:26:50 #info tripleo repos to move to independent (disucssion part #2 - discussed in last meeting) 14:27:14 tkajinam ^^ 14:27:20 so we talked about this in the last meeting and hopefully folks have had some time to digest 14:27:33 i said we would discuss it at least once more before proceeding so... 14:27:41 comments or concerns about this? 14:27:52 #link https://releases.openstack.org/reference/release_models.html#independent 14:27:57 weshay|ruck: ^^ 14:28:00 aye 14:28:03 thanks for raising 14:28:14 there will be an email thread soon as well for comment 14:28:17 if there are no concerns then we are planning on sending something out to the list this week 14:28:24 ++ 14:29:10 ok... moving on in a few seconds if there is nothing raised about that here (you can comment on the list too once we post it) 14:29:11 do we plan to make the change for xena? 14:29:25 slagle: well, don't know if we have enough time for that. maybe not 14:29:31 Cedric Jeanneret proposed openstack/tripleo-specs master: Remove now useless placeholder https://review.opendev.org/c/openstack/tripleo-specs/+/795351 14:29:46 slagle: we may cut the Xena branch and then it will take effect afterwards, but it depends on how long the review process takes etc 14:30:00 slagle: if it is really quick and there are no objections, then yes in time for xena 14:30:01 ok, thanks 14:30:56 ok, moving to the next item then 14:31:00 #info retiring stein and queens (tag as EOL upstream - disucssion part #2 - discussed in last meeting) 14:31:20 this is the second part of the proposal raised in the last meeting, less controversial though... ^^^ 14:31:37 some of the discussion last time was about whether we want to keep Q for a bit, since it gives us feedback upstream 14:31:50 but i don't think it is worth the effort, because the traffic is very low 14:31:57 queens - for tripleo-heat-templates & tripleo-common & python-tripleoclient 23 commits since Nov 21 2020 https://gist.github.com/marios/b3155fe3b1318cc26bfa4bc15c764a26#gistcomment-3752102 14:32:15 stein - 9 commits https://gist.github.com/marios/b3155fe3b1318cc26bfa4bc15c764a26#gistcomment-3755127 14:32:38 stein was broken for a long time 14:32:45 so i think it just got bypased 14:32:49 i'd eol that one at least 14:32:55 'worth the effort' because we have to fix some of those queens jobs first, and then it would just continue to take resources away from newer branches and as mentioned we have a very low commit count there so... 14:33:10 mwhahaha: yes i think stein is without question 14:33:19 mwhahaha: the discussion really here is if we want to keep queens for a bit longer 14:33:31 mwhahaha: we did rocky last cycle so stein/queens are next 14:33:56 ok lets make it very specific 14:34:00 imho we probably should just eol queens since the ci is broke too. and we dropped most good coverage 14:34:06 does anyone object to us EOL queens? 14:34:25 (we will post something to the list for this as well so you can comment there too, just trying to get any intital discussion here) 14:34:56 ramishra: ^^^ ? 14:35:51 k then we can proceed i think weshay|ruck ? at least lets post it to openstack-discuss and see if there is further discussion there 14:36:00 moving on in a sec if there is nothing further here... 14:36:43 marios: I've been asking for it since long, so I won't object:) 14:37:14 ramishra: yeah i was more referring to keeping queens i though it was your suggestion that we keep it for the check/gate coverage it gives us 14:37:39 ramishra: which i understand in theory but looking at the commits i don't know if we can justify it in practice 14:38:05 I've changed side since long :D , I don't propose queens patches upstream anymore 14:38:24 ramishra: ack thanks ok then 14:38:41 marios, yup.. we'll get the email out 14:38:44 moving on 14:38:52 (jlarriba) Question: during PTG, there were some arguments in favor of removing ansible from tripleo because of distribution changes. Has any decision been taken? 14:38:56 question from jlarriba 14:39:09 this was exploratory work from mwhahaha and cloudnull ... can you answer this ? ^ 14:39:14 i don't know what the status of that work is 14:39:28 still investigating, no time line 14:39:36 yes, we are working heavily in B&R with tripleo-ansible, and we dont know if we should start planning to switch away from ansible or continue to invest on it 14:39:38 but we don't want to increase reliance on ansible specific stuff 14:40:07 we'll likely still support ansible playbook execution 14:40:11 but it may change how that occurs 14:40:17 so careful 14:40:17 i guess we can say not in X timeframe mwhahaha ? 14:40:24 Takashi Kajinami proposed openstack/puppet-tripleo master: Allow tuning timeouts for rabbitmq pacemaker resource https://review.opendev.org/c/openstack/puppet-tripleo/+/795302 14:40:29 TBD 14:40:34 :) 14:40:46 mwhahaha, is the directord thing ? or is there something else? 14:40:51 TBD 14:40:58 ok 14:41:15 we can neither confirm nor deny the existence of an non ansible deployment tool 14:41:16 :D 14:41:27 XDDDD 14:41:29 sorry couldn't resist 14:41:34 like i said, we'll likely still support the execution of playbooks 14:41:38 but that may change how it happens 14:42:14 mwhahaha, ack 14:42:25 ok moving on i think ? 14:42:30 is the a plan or idea for puppet as well? 14:42:54 we don't have a plan for puppet at the moment 14:43:02 thx 14:43:03 all things may be on the table 14:43:07 we'll see 14:43:44 man we are really eating time today 14:43:48 moving on folks sorry 14:44:03 i am going to try and get through the rest of the community items quickly to allow storage to raise their items 14:44:09 please interrupt me if you want to raise something 14:44:15 releases 14:44:21 #info https://review.opendev.org/c/openstack/releases/+/794251 * Make a release for tripleo victoria repos 14:44:22 storage only needs 30 seconds 14:44:30 #info https://review.opendev.org/c/openstack/releases/+/794583 * Make final train tags tripleo repos before extended maintenance 14:44:41 i will post ussuri soon too 14:45:08 note from tkajinam FYI: We are creating the final stable/train release now for puppet moduels so once it is cut off then we expect consistent release between puppet and TripleO 14:45:12 https://review.opendev.org/c/openstack/releases/+/792820 14:45:19 #info tripleo and oooq moved to OFTC 14:45:30 obviously everyone here knows this already ;) ^^ 14:45:38 #info doc update at https://review.opendev.org/c/openstack/tripleo-docs/+/793824 14:45:47 #info posted chan ops/admin update https://review.opendev.org/c/openstack/project-config/+/793827 14:45:54 I just added some folks so we can have representation across timezones if th is is needed (e.g. to deal with trolls?). If any other tripleo-core would like to have op access please post a review similar to project-config/+/793827 and add at least one other tripleo-ci core as reviewer? 14:46:19 (tkajinam) Q. Do we still use tripleo-puppet-elements ? https://opendev.org/openstack/tripleo-puppet-elements 14:46:29 tkajinam: i think we do, at least the image_yaml refers to some of those elements 14:46:40 if that is still required then I'll fix it 14:46:41 tkajinam: ^^ please go ahead while i go fetch an example 14:46:50 if that is no longer used then I'll drop it 14:46:59 it's heavily outdated 14:47:03 yes we still use parts of it 14:47:17 tkajinam: e.g. there https://opendev.org/openstack/tripleo-common/src/branch/master/image-yaml/overcloud-images-python3.yaml 14:47:28 tkajinam: so that one refers to overcloud-base for example which is from tpe 14:47:29 mwhahaha, marios ok 14:47:58 k thanks tkajinam mwhahaha 14:48:08 any comments or concerns about any of those community items please? 14:48:12 the main proboel is puppet-modules which is included in the list 14:48:25 so I'll continue fixes. thx 14:48:56 thanks tkajinam. k lets move on then 14:48:59 storage :D 14:49:02 fultonj: please go ahead 14:49:09 #info cephadm integration updates 1: hci working + documented, progress on ganesha + monitoring stack, standalone-scenario004 will be updated soon to use cephadm 14:49:14 #link https://review.opendev.org/q/topic:%22cephadm_ganesha%22+(status:open%20OR%20status:merged) 14:49:17 #link https://review.opendev.org/q/topic:%22cephadm_monitoring%22+(status:open%20OR%20status:merged) 14:49:27 #info cephadm integration updates 2: cephadm/ceph-ansible packages in -stream respos used by quickstart, working on RBD before overcloud, planning to remove {deployment,environments}/ceph-ansible/* by end of X 14:49:34 #link https://cbs.centos.org/koji/packageinfo?packageID=8439 14:49:35 #link https://cbs.centos.org/koji/packageinfo?packageID=3726 14:50:22 #info tl;dr progress on ceph-ansible removal, we expect to remove it before X ends 14:50:27 (that's all) 14:50:40 thank you fultonj 14:50:41 thanks fultonj 14:50:52 fultonj: are the HCI stuff yours? 14:50:56 sorry fmount i meant ^ 14:51:02 yeah 14:51:17 #link https://docs.openstack.org/project-deploy-guide/tripleo-docs/latest/features/cephadm.html#scenario-deploy-hyperconverged-ceph 14:51:36 requires pacific-dev container ATM 14:52:33 are there are comments concerns questions for the storage team please ? 14:53:17 moving on... 14:53:24 thanks all for the items raised 14:53:35 #topic Bugs and blueprints 14:53:35 #link https://bugs.launchpad.net/tripleo/ 14:53:35 #link https://storyboard.openstack.org/#!/project/openstack/tripleo-ansible 14:53:35 #link https://launchpad.net/tripleo/+milestone/xena-1 14:53:37 #link https://launchpad.net/tripleo/xena 14:54:03 usual reminder please keep your bug status up to date, especially check https://launchpad.net/tripleo/+milestone/xena-1 any bugs you may have been involved with 14:54:14 anything bug related folks want to raise here? 14:54:29 (me moving quickly unless folks speack up cos time) 14:54:47 #topic Project releases or stable backports 14:54:59 we already covered the releases, unless there is any specific request or comment here? 14:55:21 https://review.opendev.org/q/I86c6d1d060885d0a99b0e0c72cade97ee7833313 14:55:27 #topic specs 14:55:29 #info https://review.opendev.org/q/project:openstack/tripleo-specs 14:55:48 ci team would like to merge https://review.opendev.org/c/openstack/tripleo-specs/+/772442 soon 14:55:56 does anyone object to us doing so? ^^^ 14:56:00 rlandy: ^^ 14:56:21 marios: thanks for raising 14:56:34 * marios cant see any other spec in https://review.opendev.org/q/project:openstack/tripleo-specs apart from the ones Tengu already mentinoed 14:56:42 looks like we merged the stuff we discussed at ptg, or almost merged 14:56:45 marios: https://review.opendev.org/c/openstack/tripleo-specs/+/795351 some cleanup 14:56:54 not a spec, just.... cleaning :) 14:56:57 Tengu: thanks 14:57:03 (sorry, I'm on another mtg) 14:57:07 Tengu: np :D 14:57:22 rlandy: i didn't hear any objections ;) so lets merge 14:57:30 yay - thanks 14:57:37 done# 14:57:47 and just in time 14:57:52 #topic open discussion 14:57:53 Anything else that folks want to bring up to the meeting? 14:58:25 jfyi we haven't been testing some modules in tripleo-ansible 14:58:33 https://review.opendev.org/q/I5f15e26332d2b8b1e53539d58fe095997f3151c9 14:58:37 trying to fix but it's complicated 14:58:50 master has been fixed, stable branches need some work 14:59:01 thanks mwhahaha i had a look at the wallaby one this morning but looks like deps conflicts 14:59:15 yea we'll likely need to change how we do doc requirements 14:59:24 was going to poke at it today 14:59:40 right now we symlink doc-requirements to molecule-requirements 14:59:51 which causes an issue with upper-constraints 14:59:57 fun with source 15:00:04 OK we are at time now 15:00:25 so thanks all for the discussion today, next meeting Tue 22 June 15:00:40 #endmeeting tripleo