17:00:10 <hberaud> #startmeeting releaseteam
17:00:11 <openstack> Meeting started Thu Mar 25 17:00:10 2021 UTC and is due to finish in 60 minutes.  The chair is hberaud. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:14 <openstack> The meeting name has been set to 'releaseteam'
17:00:15 <hberaud> #link https://etherpad.opendev.org/p/wallaby-relmgt-tracking Agenda
17:00:18 <hberaud> Ping list: ttx armstrong elod damani
17:00:22 <armstrong> o/
17:00:27 <ttx> o/
17:00:38 <hberaud> We're way down on line 507 now.
17:00:53 <hberaud> Will just wait a couple minutes for folks.
17:01:27 <ttx> time flies
17:03:01 <hberaud> #topic Review task completion
17:03:09 <hberaud> 1) Process any remaining library branching exception.
17:03:16 <hberaud> The only exception is this one => https://review.opendev.org/c/openstack/releases/+/781175
17:03:27 <hberaud> Matthew was ok
17:03:31 <ttx> just approved
17:03:36 <hberaud> awesome
17:03:46 <hberaud> 2) On the Monday, generate release requests for all deliverables that have do not have a suitable candidate yet.
17:03:54 <hberaud> https://review.opendev.org/dashboard/?title=Releases+Inbox&foreach=is%3Aopen&Wallaby=project%3Aopenstack%2Freleases+file%3A%5Edeliverables%2Fwallaby%2F.%2A+NOT+label%3AWorkflow%2D1&Victoria=project%3Aopenstack%2Freleases+file%3A%5Edeliverables%2Fvictoria%2F.%2A+NOT+label%3AWorkflow%2D1&Ussuri=project%3Aopenstack%2Freleases+file%3A%5Edeliverables%2Fussuri%2F.%2A+NOT+label%3AWorkflow%2D1&Train=projec
17:03:56 <hberaud> t%3Aopenstack%2Freleases+file%3A%5Edeliverables%2Ftrain%2F.%2A+NOT+label%3AWorkflow%2D1&Independent=project%3Aopenstack%2Freleases+file%3A%5Edeliverables%2F_independent%2F.%2A+NOT+label%3AWorkflow%2D1&Other=project%3Aopenstack%2Freleases+NOT+file%3A%5Edeliverables%2F.%2A&Jobs=project%3Aopenstack%2Fproject%2Dconfig+file%3A%5Eroles%2Fcopy%2Drelease%2Dtools%2Dscripts%2Ffiles%2Frelease%2Dtools%2F.%2A&T
17:03:58 <hberaud> ools=%28%28+project%3Aopenstack%2Freleases+file%3A%5Etools%2F.%2A+%29+OR+project%3Aopenstack%2Frelease%2Dtest+OR+%28+project%3Aopenstack%2Freleases+file%3A%5Eopenstack_releases%2F.%2A+%29+OR+project%3Aopenstack%2Freno%29&All+Releases=is%3Aopen+project%3Aopenstack%2Freleases
17:04:04 <hberaud> sorry
17:04:06 <hberaud> https://review.opendev.org/q/topic:%22wallaby-rc1-deadline%22+(status:open%20OR%20status:merged)
17:04:35 <hberaud> So tomorrow we will see more clearly
17:05:00 <hberaud> and the last one task
17:05:04 <hberaud> 3) Tracking yarn fail
17:05:07 <hberaud> Done!
17:05:15 <hberaud> The job is now fixed
17:05:38 <hberaud> and deliverables have been properly uploaded
17:05:52 <hberaud> Next topic
17:05:59 <hberaud> #topic Assign R-2 tasks
17:06:44 <elod> o/ sorry for being late :X
17:06:57 <hberaud> Some tasks are already assigned. Any volunteer for the remaining tasks?
17:07:13 <hberaud> By example to ping the QA team
17:07:46 <gmann> hberaud: ack on QA items.
17:08:18 <ttx> looking
17:08:21 <hberaud> ok thanks gmann
17:09:04 <hberaud> "Ensure that all projects that are publishing release notes have the notes link included in their deliverable file." Any volunteer?
17:09:15 <ttx> got it
17:09:20 <hberaud> thanks
17:09:38 <hberaud> so everything seems assigned
17:09:53 <rosmaita> (can someone ping me when open discussion starts?)
17:10:01 <armstrong> I am available
17:10:11 <hberaud> rosmaita: ok
17:10:25 <armstrong> ttx:
17:10:36 <armstrong> if you need any help I am available
17:10:53 <hberaud> armstrong: I put your name on the QA stuff
17:11:00 <hberaud> with me
17:11:08 <armstrong> hberaud: OK thanks
17:11:28 <hberaud> Ok thanks. Next topic
17:11:35 <hberaud> #topic Review countdown email contents
17:11:41 <hberaud> https://etherpad.opendev.org/p/relmgmt-weekly-emails
17:14:08 <ttx> looking
17:14:13 <openstackgerrit> Merged openstack/releases master: Proposing Wallaby RC1 for octavia-dashboard  https://review.opendev.org/c/openstack/releases/+/781727
17:14:35 <openstackgerrit> Merged openstack/releases master: Proposing Wallaby RC1 for designate-dashboard  https://review.opendev.org/c/openstack/releases/+/781644
17:14:38 <ttx> lgtm
17:14:55 <hberaud> thanks
17:15:28 <hberaud> Ok move on
17:15:34 <hberaud> #topic Moving ovn-octavia-provider to cycle-with-rc
17:15:40 <hberaud> #link https://review.opendev.org/c/openstack/releases/+/780673
17:16:01 <openstackgerrit> Merged openstack/releases master: New heat-translator version 2.3.0  https://review.opendev.org/c/openstack/releases/+/781175
17:16:18 <hberaud> Any opinion about ^? c.f comments
17:16:58 <hberaud> I think it's not too late to switch the model
17:17:44 <openstackgerrit> Merged openstack/releases master: Proposing Wallaby RC1 for storlets  https://review.opendev.org/c/openstack/releases/+/781759
17:18:35 <johnsom> The octavia team doesn't have an opinion. It's really up to neutron on how to manage that.
17:20:02 <hberaud> Ok thanks for feedback I'll update the patch accordingly
17:20:02 <ttx> oops sorry was disconnected
17:20:02 <ttx> ok so both could work, it;s very late to change it
17:20:02 <ttx> so I'd say it depends who wants to change it and how badly
17:20:23 <haleyb> hberaud: sorry, got bumped with ssl update
17:21:01 <haleyb> i'm probably the only one that cares about the ovn-octavia-provider
17:21:48 <hberaud> Do you want to update the patch to move to cycle-with-rc now?
17:22:12 <haleyb> at this point i'm more interested in the version numbering, but i could make it cycle-with-rc too
17:22:38 <hberaud> ttx: FYI I proposed to move the model nooow
17:22:40 <hberaud> now
17:22:42 <ttx> one option is to make Wallaby 1.0.0
17:22:44 <ttx> but keep it cycle-with-intermediary
17:22:46 <ttx> Currently it's pre-1.0 so not following semver
17:23:01 <hberaud> LGTM
17:23:10 <ttx> if you want to be able to do semver bump you need 1.0.0, you don;t need -with-rc really
17:23:32 <hberaud> haleyb: WFY?
17:23:47 <haleyb> ttx: yes, that would work too, i want to be able to backport things that might require a minor version bump, which 1.0.0 would get us
17:23:58 <ttx> basically pre-1.0 you just do bumps however you like it and semver is ignored by users
17:23:59 <hberaud> make sense
17:24:44 <ttx> haleyb: so you should also feel empowered to backport things taht require minor bump and only bump the .z
17:24:50 <haleyb> ttx: i guess i don't understand keeping it 0.x.y then?
17:25:01 <ttx> 1.0.0 is when you start signalling you'll respect semver. Before, all bets are off
17:25:15 <ttx> you can break everything when you want
17:25:27 <hberaud> (with 0.y.z)
17:25:47 <haleyb> semver being semantic versioning?
17:26:02 <hberaud> yes
17:26:07 <ttx> yes. it really depends how ready it is to follow semantic versioning
17:26:49 <ttx> 3 options:
17:27:07 <ttx> (to backport things that should warrant a minor bump)
17:27:21 <hberaud> haleyb: https://semver.org/#spec-item-4
17:27:33 <ttx> 1- release wallaby as 0.4.0nd backport things and release tehm as 0.4.1
17:27:59 <ttx> 2- release wallaby as 1.0.0 and backport things and release them (following semver) as 1.1.0
17:28:19 <ttx> 3- same as (2) but as cycle-with-rc
17:28:37 <ttx> (3) does not really improve on (2)
17:28:54 <hberaud> 3bis - each series start with a major bump
17:28:58 <ttx> The choice between (1) and (2) is about how ready you are to follow semver for this deliverable
17:29:17 <ttx> You seem to be pretty ready, so I'd say ... pick (2)
17:29:35 <ttx> no late change in release model needed
17:29:42 <hberaud> yeah
17:29:48 <hberaud> Agreed
17:30:27 <hberaud> haleyb: WFY?
17:30:31 <ttx> obviously 1.0.0 also conveys some "readiness to be used" so if it's utterly broken right now I'd recommend option (1)
17:30:46 <ttx> but I suspect it's perfectly functional
17:30:52 <haleyb> thanks for the info, i think it's fine to do 1.x.y as it is not a library so i'm not worried about breaking any API
17:31:01 <ttx> ack
17:31:03 <haleyb> sorry, my connection is wonky
17:31:28 <ttx> it's wonky for everyone, we are in the middle of a freenode maintenance window :)
17:32:06 <haleyb> ttx: it's being used downstream already, so only broken when we merge a bug :)
17:32:21 <hberaud> :)
17:33:14 <haleyb> so i'll update to 1.0.0 and leave cycle-with-interbediary, which seemed to be the consensus
17:33:16 <hberaud> haleyb: feel free to submit a new PS to propose a version 1.0.0
17:33:24 <hberaud> ack
17:33:26 <hberaud> thanks
17:33:32 <haleyb> np, thanks for the discussion
17:33:40 <hberaud> you're welcome
17:33:54 <hberaud> Ok next topic
17:34:07 <hberaud> #topic Abandoning a couple of patches
17:34:29 <hberaud> Those are trailing projects
17:34:42 <hberaud> So I think we can abandon them for now
17:34:52 <hberaud> Any opinion?
17:34:55 <hberaud> https://review.opendev.org/c/openstack/releases/+/781661
17:35:02 <hberaud> https://review.opendev.org/c/openstack/releases/+/781660
17:35:11 <hberaud> https://review.opendev.org/c/openstack/releases/+/781657
17:35:18 <hberaud> https://review.opendev.org/c/openstack/releases/+/781731
17:35:39 <elod> when do we need to release trailing projects? (I guess when every non-trailing have released)
17:36:01 <ttx> yes abandon
17:36:19 <ttx> they have until like 3 months after release
17:36:22 <hberaud> However I think we need to update our doc to ignore them next time
17:36:57 <elod> 3 months sounds quite far, so agree, let's abandon them
17:37:27 <hberaud> Done
17:37:44 <hberaud> I'll try to update the doc soon
17:37:50 <hberaud> Ok next topic
17:37:56 <hberaud> #topic Recent job failure
17:38:11 <hberaud> #link http://lists.openstack.org/pipermail/release-job-failures/2021-March/001522.html
17:39:14 <hberaud> Was related to https://opendev.org/openstack/releases/commit/391616bfee043dc3f17eb8eaa1942dedcd6319a9
17:39:32 <hberaud> pypi have been properly updated => https://pypi.org/project/kuryr-kubernetes/4.0.0.0rc1/
17:39:48 <hberaud> stable branch have been created => https://opendev.org/openstack/kuryr-kubernetes/src/branch/stable/wallaby
17:39:57 <hberaud> however the git tag is missing
17:40:11 <ttx> hmm, that's weird
17:40:14 <hberaud> I didn't what was exactly the problem in logs
17:40:20 <elod> I see both the tag and the branch @ https://opendev.org/openstack/kuryr-kubernetes#
17:40:27 <ttx> rigth ^
17:40:41 <ttx> There is no way it could upload to Pypi without the tag benig pushed
17:40:49 <hberaud> Ah right
17:40:56 <hberaud> I missed the tag previously
17:41:12 <ttx> The releases change merged, the post job picks it up and tags the repository
17:41:24 <ttx> Then a release job on that repo is publishing artifacts
17:41:31 <hberaud> Every pieces seems there
17:41:49 <ttx> if the tag is present, the failure can probably be ignored
17:41:59 <ttx> taht job only does that tagging
17:42:06 <hberaud> Probably
17:42:37 <ttx> oh, and then publish-tox-docs-releases was skipped
17:42:38 <ttx> but that's refreshed on the next run
17:42:43 <hberaud> ok
17:42:51 <hberaud> I think we can ignore it
17:42:52 <ttx> yes
17:43:09 <elod> cool
17:43:37 <elod> (I even see the tarball @ https://tarballs.opendev.org/openstack/kuryr-kubernetes/ )
17:44:06 <hberaud> anyway final RC is not far so doc things will be reexecuted soon
17:44:25 <hberaud> ok next topic
17:44:35 <hberaud> #topic Open Floor
17:44:41 <hberaud> rosmaita: ping
17:44:53 <hberaud> Anything else?
17:45:03 <rosmaita> hello
17:45:08 <hberaud> o/
17:45:17 <rosmaita> ok, we found a bug in os-brick that can lead to data loss
17:45:25 <rosmaita> of course, os-brick for wallaby has been released
17:45:37 <rosmaita> this is the bug : https://review.opendev.org/c/openstack/os-brick/+/782992
17:45:46 <rosmaita> so i was wondering what our options are
17:46:18 <hberaud> If you are able to merge this patch asap then we could consider an RFE
17:46:25 <ttx> fix bug on main branch, backport to wallaby, release bugfix
17:47:02 <ttx> that definitely qualifies as an exception if data loss is on the table
17:47:05 <rosmaita> ok, so i need to send a RFE to the ml as soon as the backport patch is up
17:47:07 <hberaud> yeah I backport is needed
17:47:24 <hberaud> s/I/a/
17:47:30 <hberaud> Yes
17:47:34 <rosmaita> ok, so it doesn't seem out of the question for wallaby
17:47:38 <hberaud> prometheanfire: FYI ^
17:48:06 <hberaud> however time flies
17:48:26 <rosmaita> yeah, so would tomorrow be the deadline?
17:48:32 <hberaud> the window is short
17:48:46 <rosmaita> yeah, i figured
17:49:06 <hberaud> hm requirements freeze is during R-2
17:49:06 <hberaud> this is our deadline
17:49:14 <rosmaita> ok, so that is next week
17:49:26 <hberaud> until requirements aren't freezed we can land that
17:49:36 <ttx> hberaud: we'll fix a data loss no matter what
17:49:43 <hberaud> *early next week
17:50:10 <hberaud> ttx: ok
17:50:14 <rosmaita> ok, thanks i will aim for monday
17:50:20 <hberaud> excellent
17:50:29 <hberaud> do not hesitate to ping us if needed
17:50:33 <ttx> IIRC the freeze is there so that we do not have to bump requirements on every affected project
17:50:33 <rosmaita> thanks, i will leave the meeting and go review code
17:50:52 <hberaud> thanks
17:50:57 <rosmaita> it would be cinder and nova, mostly nova really
17:51:07 <rosmaita> though possibly ironic
17:51:15 <rosmaita> anyway, i will do my best to have it on monday
17:51:20 <ttx> I bet prometheanfire remembers better than i do. But in the past we fixed critical regressions the day before release
17:51:36 <ttx> so rosmaita is definitely on time :)
17:51:46 <rosmaita> for a change!
17:51:55 <rosmaita> ok, thanks everyone
17:52:06 <hberaud> I comfident about this topic
17:52:25 <hberaud> rosmaita: You're welcome
17:52:31 <hberaud> Anything else?
17:54:21 <hberaud> OK, thanks everyone. Let's wrap up.
17:54:23 <hberaud> #endmeeting