16:00:03 <smcginnis> #startmeeting releaseteam
16:00:04 <openstack> Meeting started Thu Sep 12 16:00:03 2019 UTC and is due to finish in 60 minutes.  The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:07 <openstack> The meeting name has been set to 'releaseteam'
16:00:15 <smcginnis> Ping ttx dhellmann diablo_rojo hberaud evrardjp armstrong tonyb
16:00:16 <hberaud> o/
16:00:26 <armstrong> o/
16:00:27 <evrardjp> o/
16:00:34 <smcginnis> #link https://etherpad.openstack.org/p/train-relmgt-tracking Agenda tracking
16:00:38 <fungi> a deh ya!
16:00:46 <fungi> you're right on time
16:01:05 <smcginnis> Gotta be punctual. ;)
16:01:56 <openstackgerrit> Merged openstack/releases master: Release tosca-parser 1.6.0  https://review.opendev.org/681240
16:02:04 <smcginnis> I just saw ttx around, so just waiting a little more before starting.
16:02:13 <ttx> o/
16:02:19 <smcginnis> #topic Review week tasks completion
16:02:29 <smcginnis> Looks like most are crossed off as done.
16:02:50 <smcginnis> evrardjp: I saw you ping diablo_rojo_phon earlier for that client lib task.
16:03:05 <evrardjp> yeah
16:03:06 <evrardjp> so
16:03:15 <evrardjp> I haven't proposed stuff
16:03:17 <evrardjp> https://etherpad.openstack.org/p/train-m3-unreleased-for-autoreleases
16:03:44 <evrardjp> I have all those patches in my env ready, but I haven't digged yet into what must be proposed or not
16:03:55 <smcginnis> Since some teams are still trying to get last minute changes in before their final client release, we could probably wait until later today or early tomorrow to see what hasn't been covered by the teams by then.
16:03:58 <evrardjp> Either I can just propose everything and cancel what's not necessary
16:04:08 <evrardjp> yeah that's what I thought too
16:04:30 <evrardjp> but when I see that list of names ... I doubt things will change :)
16:04:42 <evrardjp> let's see if I was wrong :D
16:04:50 <smcginnis> Maybe one or two of them, but yeah, I see some usual suspects there.
16:05:01 <evrardjp> yeah exactly :)
16:05:37 <evrardjp> Would you mind if I propose them all, and cancel the releases that don't make sense? I know it's not the usual way, but our tox output is so nice.
16:05:46 <evrardjp> and I love zuul
16:05:49 <ttx> ++
16:05:59 <ttx> Better to put those out there now
16:06:04 <smcginnis> At least we have pretty low overhead for our jobs.
16:06:13 <evrardjp> ok
16:06:17 <smcginnis> So a few extra patches won't slow everyone down or anything.
16:06:17 <evrardjp> let me push those now then
16:06:34 <smcginnis> It might make it easier for some teams too since they can just do a quick update of the hash if needed.
16:06:47 <smcginnis> Thanks for doing that evrardjp
16:07:04 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-barbicanclient  https://review.opendev.org/681793
16:07:10 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-congressclient  https://review.opendev.org/681794
16:07:13 <evrardjp> that took me like 3 minutes, I can do that :)
16:07:15 <smcginnis> #topic Release requests that may be affected by freezes
16:07:17 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-cyborgclient  https://review.opendev.org/681795
16:07:23 <evrardjp> sorry for the spam
16:07:24 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-glanceclient  https://review.opendev.org/681796
16:07:32 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-mistralclient  https://review.opendev.org/681797
16:07:36 <smcginnis> Right in the middle of the meeting logs! :D
16:07:37 <ttx> yeah, I was about to approve that one this morning and was like... wait a minute
16:07:38 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-monascaclient  https://review.opendev.org/681798
16:07:46 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-octaviaclient  https://review.opendev.org/681799
16:07:53 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-saharaclient  https://review.opendev.org/681800
16:07:59 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-searchlightclient  https://review.opendev.org/681801
16:08:01 <evrardjp> it's almost over...
16:08:06 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-swiftclient  https://review.opendev.org/681802
16:08:10 <ttx> In theory we are at library freeze
16:08:14 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-tackerclient  https://review.opendev.org/681803
16:08:17 <smcginnis> I thought oslo was already in freeze since they try to do it a week before the other freezes.
16:08:21 <openstackgerrit> Jean-Philippe Evrard proposed openstack/releases master: Auto releasing python-tricircleclient  https://review.opendev.org/681804
16:08:38 <ttx> so should we really be approving those ? I guess it's ~ok
16:08:45 <ttx> since requirements is not frozen yet
16:09:11 <smcginnis> I thought we would still look for a requirements FFE for anything past the lib freeze.
16:09:12 <evrardjp> should we implement some check to make it clear?
16:09:19 <smcginnis> Even if requirements in total are not frozen.
16:09:20 <bnemec> We're in feature freeze.
16:09:23 * diablo_rojo appears late
16:09:29 <smcginnis> bnemec: Oh good, was just going to ping.
16:09:36 <smcginnis> bnemec: Are the updates in https://review.opendev.org/#/c/681333/1 important for train?
16:09:39 <bnemec> The releases I proposed were bug fixes.
16:10:10 <ttx> yeah it's a bit in grey area
16:10:21 <bnemec> I guess they're not critical. All of the changes are backportable if we have to.
16:10:36 <bnemec> Two are just release notes for things that landed recently.
16:10:39 <ttx> libraries are frozen a week before so that people consume the latest features
16:10:43 <smcginnis> prometheanfire: Happen to be around? Any preference on getting FFE approval for lib releases at this point?
16:10:55 <ttx> and then they are frozen at RequirementsFreeze
16:11:00 <smcginnis> Bugfixes would be good to get out there.
16:11:11 <ttx> so in theory bugfix releases are still OK this week
16:11:28 <smcginnis> And if there's no feature changes, then should be low risk for consumers of those libs.
16:11:33 <smcginnis> I'd say let's get them out then.
16:11:34 <ttx> I would approve it now and freeze requirements after
16:11:44 <ttx> just wanted us to discuss it first
16:12:11 <smcginnis> So looks like one bug fix and two release notes. Ship it.
16:12:39 <ttx> shipped
16:12:57 <smcginnis> That was the only lib one out there now, right?
16:13:03 <ttx> yes
16:13:05 <smcginnis> #topic Discuss no-change libraries
16:13:15 <ttx> So I found only two cases of no-changes-at-all
16:13:26 <ttx> cinderlib: has changes, just no releases (is cycle-trailing, so not really a library that services depend on. Should probably be moved to type:'other')
16:13:42 <smcginnis> That is planned for release after cinder.
16:13:52 <ttx> cinderlib does not behave like a library-you-depend-on
16:14:02 <ttx> it's a bit of a one-off
16:14:10 <ttx> so it shows on reports
16:14:25 <ttx> I'd advise making it a "other" thing if it is to stay a cycle-tariling thing
16:14:27 <smcginnis> It's technically a library, for other things outside of openstack, but probably does make more sense for our purposes to call it "oterh"
16:14:40 <ttx> We don;t do cycle-trailing libs
16:14:44 <smcginnis> Or "other" even.
16:14:52 <ttx> OK, I'll propose that
16:14:56 <smcginnis> ++
16:15:08 <ttx> the other is os-acc and that is being abandoned
16:15:18 <ttx> I'll propose a removal of the deliverable file
16:15:27 <ttx> since that is what throws me off
16:15:36 <smcginnis> ++ againo
16:15:40 <smcginnis> *again
16:15:43 <smcginnis> Sheesh, typing.
16:16:28 <smcginnis> Only two is actually pretty good. And they weren't bad ones either.
16:17:18 <smcginnis> #topic Remind the requirements team to freeze changes to openstack/requirements by applying -2 to all open patches.
16:17:19 <prometheanfire> smcginnis: ya, I'd say we are in freeze now, so FFEs for libs would be good
16:17:45 <openstackgerrit> Thierry Carrez proposed openstack/releases master: Remove os-acc deliverable  https://review.opendev.org/681808
16:17:47 <smcginnis> prometheanfire: Would you like bnemec to do it for oslo? There's one minor bugfix and two release note changes, so they are pretty minor.
16:17:52 <smcginnis> Going through gate right now.
16:18:06 <prometheanfire> I'd approve it, just needs the request
16:18:57 <bnemec> Okay, I'll send a request to the list.
16:19:07 <smcginnis> Thanks!
16:19:21 <prometheanfire> there are a bunch of things still getting released to master that'll need it I think, unless the cutoff is 'now' or any review after x
16:19:34 <evrardjp> I missed that part for FFE when reviewing, sorry. I thought it was not yet active.
16:19:55 <prometheanfire> many things in the same timeframe
16:20:25 <prometheanfire> the date for freeze could go up til tomorrow though (the 13th)
16:20:34 <prometheanfire> looking at the schedule
16:20:41 <openstackgerrit> Thierry Carrez proposed openstack/releases master: Make cinderlib an "other" deliverable  https://review.opendev.org/681809
16:21:44 <smcginnis> OK, we'll watch for anything else coming through that could have ripple effects on other deliverables.
16:23:11 <prometheanfire> cool
16:23:43 <smcginnis> #topic R-4 email content final review
16:23:51 <smcginnis> #link https://etherpad.openstack.org/p/relmgmt-weekly-emails
16:23:57 <evrardjp> L424?
16:24:03 <ttx> I did a draft, but there was two sections I was not sure of
16:24:13 <ttx> evrardjp: yes
16:24:22 <ttx> constraints update patches should not be approved until a stable/train branch has been created for openstack/requirements ??
16:24:31 <smcginnis> I don't believe that is still the case with constraints.
16:24:44 <smcginnis> Now that tonyb got the static URLs, those can be approved at any point.
16:24:53 <ttx> The patch itself says something like "will be tested with master until stable is created"
16:25:15 <smcginnis> Very small window of risk that some new library will be updated in master before the requirements stable/train branch is created.
16:25:23 <ttx> OK removed the *
16:25:27 <smcginnis> But not very likely and seamless to the teams.
16:25:32 <ttx> fill train goal info ??
16:25:41 <smcginnis> I don't think we're doing that now either.
16:25:48 <ttx> I was wondering if teams still filled those
16:25:53 <ttx> I'm pretty sure they don't
16:25:54 <openstackgerrit> Merged openstack/releases master: Oslo releases for 2019-9-10  https://review.opendev.org/681333
16:26:07 <smcginnis> There are the storyboard stories, but not like before where each team had to submit a patch to update the goal.
16:26:10 <smcginnis> So I think that's out too.
16:26:24 <ttx> done
16:26:43 <ttx> I simplified StringFreeze to only mention dashboards since that is the only thing translated
16:27:12 <smcginnis> Looks good to me.
16:27:22 <smcginnis> Anyone else have any comments on the email content?
16:27:40 <smcginnis> Otherwise I will get that sent out after the meeting so I don't forget about it.
16:27:47 <ttx> Also I did send a separate email to cycle-with-intermediary no-release-yet peeps
16:27:59 <smcginnis> Nice, thanks
16:28:03 <ttx> because that would be noise for 80% of the teams
16:28:28 <ttx> next week we have a remidner for cycle-with-intermediary no-release-in-last-2-months peeps
16:28:53 <smcginnis> Man, another cycle flying by. Only feels like I've been PTL for a few weeks. :D
16:28:53 <ttx> ok, good to go
16:29:05 <ttx> maybe wait until tomorrow to send it since it says "we are in FF!"
16:29:17 <ttx> or late tonight
16:29:23 <evrardjp> nothing to say on the email, but would be happy to discuss after the meeting about the constraints url innerworkings. I suppose it's with the nice redirections, but I don't know how it internally works
16:29:33 <smcginnis> Oh, true. I did try to soften the note under General Information, but there's more in there.
16:29:36 <ttx> evrardjp: magic
16:29:44 <smcginnis> I will make myself a note and wait until tomorrow.
16:29:51 <evrardjp> ttx: there is no such thing Michael
16:30:20 <ttx> well, a kind of magic
16:30:28 <ttx> One dream, one soul, one prize, one goal...
16:30:38 <evrardjp> :)
16:30:40 <smcginnis> Hah
16:31:00 <smcginnis> OK, if anyone has any updates for the email draft, feel free to propose changes in the etherpad. I will get that sent out tomorrow.
16:31:07 <smcginnis> #topic Review next weeks tasks
16:31:14 <ttx> I grabbed a couple
16:31:42 <ttx> can also do the List cycle-with-intermediary ... thing
16:31:48 <ttx> unless someone wants it
16:32:14 <smcginnis> Might as well since you've done all the others. :P
16:32:20 <smcginnis> I'll do the stable branch creation.
16:32:21 <evrardjp> haha
16:32:26 <evrardjp> one man army ttx
16:32:26 <ttx> yeah, that is what I was thinking
16:32:32 <smcginnis> Unless someone else really wants to do that one.
16:32:45 <smcginnis> It's reeeallly difficult.
16:32:53 <ttx> This rage that lasts a thousand years
16:33:01 <ttx> Will soon be gone
16:33:12 <smcginnis> :)
16:33:23 <evrardjp> Is your second name Freddie?
16:33:45 <ttx> You should see me doing the Show must go on. My best karaoke thing
16:34:14 <fungi> we ought to karaoke the entire flash gordon soundtrack
16:34:15 <smcginnis> I don't think I've seen a good OpenStack karaoke party since Sydney.
16:34:15 <evrardjp> Expectations are high now, and I am glad to join you on the karaoke (just for watching that)
16:34:29 <smcginnis> fungi: ++
16:34:51 <smcginnis> OK, tasks are covered then.
16:34:56 <smcginnis> #topic Open discussion
16:35:04 <smcginnis> Anything else we need to cover?
16:35:10 <fungi> (my wife has just gotten me to agree to adopting another kitten, and this one is named "freddy mercury" apparently)
16:35:22 <diablo_rojo> Seems like noone has done cycle highlights
16:35:31 <smcginnis> Not enough cats in the car on the way to the mountains?
16:35:31 <fungi> er, freddie
16:35:47 <fungi> smcginnis: yeah, don't get me started
16:35:53 <smcginnis> :)
16:35:56 <ttx> diablo_rojo: not true, lots of them on the changes list
16:35:57 <fungi> in real news, gerrit project rename maintenance commences 1400z monday
16:36:03 <diablo_rojo> Oh
16:36:08 <ttx> diablo_rojo: but yeah, far from everyone
16:36:08 <diablo_rojo> I am behind this morning
16:36:10 <diablo_rojo> my bad
16:36:13 <smcginnis> Yeah, lot's I've been holding off for your approval diablo_rojo.
16:36:21 <diablo_rojo> I sent another email reminder yesterday
16:36:23 <ttx> same here, holding off
16:36:32 <diablo_rojo> I'll check them out shortly.
16:36:47 <fungi> so there will be some gerrit/zuul/gitea disruption monday, but also some release requests were waiting for their respective projects to be renamed i think
16:37:00 <smcginnis> We had the one.
16:37:11 <smcginnis> https://review.opendev.org/#/c/673234/
16:37:14 <clarkb> ya kayobe
16:37:49 <ttx> ++
16:37:51 <clarkb> I wanted to point out that the gate has been super flaky the last week or so
16:38:06 <smcginnis> It knows the release calendar.
16:38:06 <clarkb> seems like overnight we've managed to get a number of fixes in and it seems fairly stable today
16:38:07 <fungi> not to be confused with super freaky
16:38:17 <smcginnis> Totally different.
16:38:35 <clarkb> we can enqueue bug fixes to the gate and promote them above other changes
16:38:55 <clarkb> but also release team might need to be ready for projects not getting code in and making hard decisiosn around whether or not thigns get RC'd
16:39:00 <fungi> (i wonder if next christine will want to adopt a dog named "rick james")
16:39:29 <smcginnis> We're usually pretty flexible about spilling over into the next week for teams that have been impacted by stuff like that.
16:39:31 <smcginnis> Should be OK.
16:39:55 <smcginnis> Alrighty, anything else?
16:40:26 <diablo_rojo> None frm me :)
16:40:35 <smcginnis> OK, thanks everyone for helping with release management.
16:40:44 <smcginnis> #endmeeting