15:00:49 <smcginnis> #startmeeting releaseteam
15:00:49 <openstack> Meeting started Fri Aug 17 15:00:49 2018 UTC and is due to finish in 60 minutes.  The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:50 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:53 <openstack> The meeting name has been set to 'releaseteam'
15:01:02 <smcginnis> Hi dhellmann and annabelleB
15:01:17 <dhellmann> I know ttx is around, too
15:01:18 <smcginnis> ttx: I can't ping, but I can do individual ones I guess.
15:01:33 <smcginnis> Maybe fungi and armstrong too.
15:01:39 <armstrong> hello
15:02:14 <smcginnis> #link https://etherpad.openstack.org/p/rocky-relmgt-tracking Tracking etherpad / Agenda
15:02:29 <smcginnis> Currently at R-2, line 633
15:02:46 <ttx> o/
15:02:47 <smcginnis> ttx did a check of the deliverables this morning.
15:02:54 <smcginnis> #topic Release Status
15:03:09 <ttx> checking if anything changed since this morning
15:03:09 <smcginnis> Good to see cycle-with-milestones is good now.
15:03:16 <smcginnis> ttx: I don't believe so.
15:03:39 <smcginnis> cycle-with-intermediary still has more than I'd like to see at this point, but they have until Thursday.
15:03:57 <smcginnis> One of note is python-tricircleclient.
15:04:10 <smcginnis> We talked about that one last week and I was going to force a release.
15:04:19 <smcginnis> They had issues in their README though, so that got delayed.
15:04:22 <ttx> the other one I'd like not to fall into a black hole is magnum-ui
15:04:41 <smcginnis> ttx: What is the issue with that one?
15:05:07 <ttx> magnum released and not magnum-ui, so some might have gone PTO thinking their work was done
15:05:26 <smcginnis> Ah. Yeah, I could see that happening.
15:05:27 <ttx> usually those are released in tight succession
15:05:51 <smcginnis> Anyone want to take an action to ping that team?
15:06:02 <ttx> My last one of concern would be heat-agents, which hasn't been released since April
15:06:05 <dhellmann> most of the changes I see for magnum-ui are translations or ci/testing changes
15:06:13 <ttx> I bet dhellmann has more from the list-changes output
15:06:32 <dhellmann> http://paste.openstack.org/show/728318/
15:07:06 <dhellmann> given those testing changes we do want to release and branch
15:07:17 <dhellmann> and given the lack of anything else significant it's probably safe to do it
15:07:31 <dhellmann> 18f4af1 2018-05-07 09:28:19 +0000 Fix unicode type judgment appears to be a bug fix
15:07:41 * fungi truly hopes the magnum team hasn't forgotten they have that repo
15:07:43 <smcginnis> Bump min version of magnumclient should get out there.
15:07:45 <dhellmann> oh, and there are some new features around april 3
15:07:56 <smcginnis> Add * action for quotas
15:08:27 <ttx> we can ping strigazi here to make sure he will request a release for it soon
15:08:44 <dhellmann> ++
15:08:48 <fungi> looks like smcginnis had a health check update for the team ~2 months ago https://wiki.openstack.org/wiki/OpenStack_health_tracker#Magnum
15:08:57 <ttx> done :) Will repeat on #openstack-rekease
15:09:03 <ttx> +l-k
15:09:13 <smcginnis> fungi: No, that was the next one after for Manila
15:09:24 <fungi> d'oh, yep
15:09:28 <fungi> i completely misread
15:09:33 <smcginnis> TC contacts are mnaser and dims
15:09:36 <dhellmann> mnaser and dims haven't recorded anything about magnum
15:09:50 <smcginnis> fungi: I did at first too and was confused why I didn't remember doing that. ;)
15:09:51 <mnaser> sorry. i have not updated anything, but we have been in contact and i've personally spoken with the PTL
15:10:10 <fungi> i suppose asking them about the state of releases for magnum-ui would be a good thing to add
15:10:24 <mnaser> though, as an actual consumer of magnum-ui.. it is relatively stable
15:10:41 <mnaser> it doesn't really need much changing.  most magnum changes involve changing things inside the way it does things in the backend and talking to the COE's
15:10:48 <dhellmann> that's not too surprising, but there are some features and bug fixes in master that haven't been released
15:11:41 <ttx> ok, let's give strigazi and chance to pick it up, otherwise we'll see
15:11:57 <ttx> My last concern would be around heat-agents
15:12:14 <ttx> where we have a release, but it's pretty old
15:12:31 <ttx> analysis says "1 bugfix, 1 feature"
15:12:56 <smcginnis> I don't see ricolin in here or -release
15:13:08 <smcginnis> Nor -dev
15:13:27 <ttx> a part of me would like to release the last-released version to make an example
15:13:34 <dhellmann> branch?
15:13:52 <ttx> well, we'd branch from that (rocky) release
15:13:58 <dhellmann> yeah, I think branching from their previous release makes sense
15:14:09 <ttx> I mean yes, "release" as in "openstack-release" so branch
15:14:12 <dhellmann> it's only 4 patches if they do need to backport them
15:14:31 <dhellmann> ah, "release" not "tag"
15:14:34 <smcginnis> If they don't do another. They have until next week, but if we don't see anything, I think that's an issue for them to sort out.
15:14:58 <smcginnis> I will send a reminder to openstack-dev about the cycle-with-intermediary releases too.
15:15:19 <dhellmann> that's a long list of tempest plugins
15:15:37 <dhellmann> it seems like those should have been tagged for anything that has a branch at this point
15:15:45 <smcginnis> I'm not too concerned about those if they are late, but I think we do need to make sure they are done.
15:15:54 <ttx> ackyeahm the call triggered a few responses, but not so many
15:16:03 <smcginnis> I wonder if we should just go through and get the pypi package registered and publish-to-pypi jobs set up.
15:16:08 <dhellmann> I guess the tempest plugins can't be too incompatible with the branch if they're late
15:16:30 <ttx> Those don't need a branch, right
15:16:35 <smcginnis> Right
15:16:36 <dhellmann> they should not have branches
15:16:43 <ttx> I added them to the output because th script reports them :)
15:16:57 <dhellmann> but keystone-tempest-plugin might become incompatible with keystone master, for example
15:17:13 <dhellmann> we want tags, but not branches
15:17:15 <smcginnis> Looks like we should have that script filter out tempest-plugin types.
15:17:23 <ttx> so heat-dashboard would be the only thing that's recently-released but not branched
15:17:39 <ttx> the others being deployment stuff
15:17:46 <smcginnis> lbragstad is out, maybe cmurphy knows (just using keystone as an example since it was brought up)
15:18:35 <smcginnis> (we have a magnum-ui release now)
15:18:59 <fungi> the tags for tempest plugins are more about record-keeping than anything
15:19:10 <evrardjp> o/
15:19:17 <fungi> if we forget to tag them we won't necessarily notice because there's no real impact to upstream ci
15:19:19 <smcginnis> Yeah. Could cause issues as things diverge, but less time critical I think.
15:19:41 <smcginnis> Missing branches we can do after the RC final deadline.
15:19:52 <fungi> well, mainly that we do a disservice to downstream consumers of those tests if we fail to tag them in an appropriate timeframe
15:19:52 <smcginnis> If they don't get covered by then.
15:20:19 <dhellmann> right, tagging those plugins was a request from (or at least on behalf of) downstream consumers
15:21:05 <dhellmann> I wonder if we need to build in some sort of checkpoint process. Like the first time a team requests any tag, we talk to the PTL/liaison about the plans for *all* of their deliverables for the cycle.
15:21:07 <smcginnis> Do we want to force those if not done by RC-final? Or final-final?
15:21:27 <ttx> RC-final
15:21:38 <dhellmann> yeah, I think that's right, rc-final
15:22:01 <ttx> OK, so the list appears long but it looks like all the problem cases are covered
15:22:07 <smcginnis> OK, I will mention that in my reminder post about c-w-i deliverables.
15:22:25 <ttx> Considering how many deliverables we have, it's not that bad
15:23:12 <smcginnis> OK, any other concerns about Rocky before we move on?
15:23:37 <smcginnis> #topic Stein PTG
15:23:43 <smcginnis> I started the tracking etherpad
15:23:44 <smcginnis> #link https://etherpad.openstack.org/p/stein-relmgt-tracking Stein etherpad started
15:23:45 <ttx> (788 deliverables)
15:23:55 <smcginnis> ttx: Wow, not bad then.
15:24:20 <ttx> not all cycle-tied, but still
15:24:21 <smcginnis> I also created a PTG etherpad even though I think we will just end up grabbing a quiet corner again.
15:24:24 <smcginnis> #link https://etherpad.openstack.org/p/relmgt-stein-ptg PTG Topic etherpad
15:25:10 <smcginnis> If you know of any other travel or holiday plans already, please add them to the Stein tracking etherpad.
15:26:00 <ttx> Stein tracking etherpad... fancy
15:26:05 <smcginnis> It's looking like Denver will be the start of a three week travel sprint for me now, but timing is probably OK on that.
15:26:26 <smcginnis> I'm being "proactive" ;)
15:26:27 <dhellmann> oof
15:27:42 <smcginnis> Anything else we should discuss ahead of Denver?
15:27:51 <ttx> wow, looks like I'll be around for FF week this time
15:27:58 <smcginnis> \o/
15:28:01 <ttx> \o/
15:28:21 <ttx> Snowboarding the week before, so unlikely to do anythign that week
15:28:33 <smcginnis> Just be safe out there then.
15:29:08 <smcginnis> #topic Open Discussion
15:29:21 <smcginnis> I didn't have anything else for today. Anyone have anything else?
15:29:51 <fungi> i have no travel plans aside from ptg, so happy to be infra on point for release problems
15:29:56 <ttx> smcginnis: btw I'll likely have to skip OSS EU in Edimburg. Plans have me traveling for the two weeks prior to that, and I hesitated to add a third week
15:30:09 <ttx> then them not selecting my talk pushed me over the decision line
15:30:25 <dhellmann> I'm going to be out next friday
15:30:29 <smcginnis> ttx: Probably a good call.
15:30:41 <smcginnis> dhellmann: Ack, thanks.
15:30:50 <smcginnis> Hopefully all is quiet by then.
15:31:09 <smcginnis> I'll wrap up then if there aren't any more topics.
15:31:12 <smcginnis> Thanks everyone.
15:31:13 <ttx> nothing else to raise
15:31:15 <annabelleB> thanks!!
15:31:19 <ttx> Thanks smcginnis !
15:31:21 <armstrong> thx
15:31:24 <smcginnis> #endmeeting