16:01:43 <smcginnis> #startmeeting releaseteam
16:01:44 <openstack> Meeting started Thu Jan 31 16:01:43 2019 UTC and is due to finish in 60 minutes.  The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:45 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:47 <openstack> The meeting name has been set to 'releaseteam'
16:02:01 <smcginnis> People need to stop pinging me right before meeting start times. :)
16:02:26 <smcginnis> ttx: Not referring to you - others that are making me late.
16:02:30 <smcginnis> https://etherpad.openstack.org/p/stein-relmgt-tracking
16:02:37 <smcginnis> Ping list: smcginnis ttx dhellmann diablo_rojo hberaud evrardjp
16:04:36 <evrardjp> o/
16:04:45 <smcginnis> OK, good...
16:04:51 <smcginnis> #topic Review days
16:05:15 <smcginnis> Now that we have a few more folks, it would be good to redo what our review days are.
16:05:21 <smcginnis> And actually get it written down.
16:05:31 <smcginnis> I've added a section to the top of the tracking etherpad.
16:05:35 <evrardjp> sounds good
16:05:39 <smcginnis> diablo_rojo_phon: You around?
16:05:42 <hberaud> o/
16:05:50 <smcginnis> Welcome!
16:06:27 <diablo_rojo_phon> Kinda..
16:06:48 <diablo_rojo_phon> In Brussels about to figure out dinner, but can follow along here :)
16:07:07 <smcginnis> diablo_rojo_phon: OK, just wanted to make sure you were at least aware of the review days.
16:07:27 <smcginnis> diablo_rojo_phon: When you have time, please add your name to one of the days on the top of the tracking etherpad.
16:07:32 <diablo_rojo_phon> Much appreciated smcginnis :)
16:07:35 <diablo_rojo_phon> Will do!
16:08:06 <smcginnis> Any questions, concerns, comments on review days?
16:08:13 <smcginnis> ttx, dhellmann: or additional commentary?
16:08:46 <dhellmann> I've signed up for friday, which is usually a light day, so I can float on other days for a while, too
16:09:06 <smcginnis> ++
16:09:09 <diablo_rojo_phon> Nothing from me. I picked Tuesday but if we want me to shift I can do that too.
16:09:19 <ttx> LGTM
16:09:33 <smcginnis> dhellmann: I will do the same. Will be helpful while evrardjp and diablo_rojo_phon get up to speed.
16:09:35 <evrardjp> I picked the last day
16:10:04 <smcginnis> If any of these end up to be particularly troublesome days for anyone, just let us know and we can adjust.
16:10:21 <dhellmann> ++
16:10:28 <smcginnis> #topic Clarify Stein membership status for tripleo-common-tempest-plugin
16:10:41 <smcginnis> ttx: You have the background on this?
16:11:03 <ttx> sure
16:11:17 <ttx> It's the last leftover on the MembershipFreeze analysis
16:11:38 <ttx> (a repo that is in governance but not defined in release management yet)
16:11:53 <smcginnis> Do they not expect to be ready for any releases in stein?
16:11:57 <ttx> When I last asked EmilienM he said he was not sure
16:12:02 <ttx> as the created got reassigned
16:12:26 <ttx> I think we could just consider it Train material if it ever gets picked up
16:12:32 <EmilienM> hello
16:12:41 <smcginnis> Hey EmilienM
16:12:51 <ttx> just wanted to double-check with y'all if that was ok
16:13:00 <ttx> (including EmilienM )
16:13:02 <dhellmann> it's also a tempest plugin, so it could be independent if needed for now
16:13:08 <EmilienM> jaosorior: do you know the status on tripleo-common-tempest-plugin ?
16:13:28 <EmilienM> tbh I don't think this repo is going anywhere, it has a super low activity
16:13:41 <ttx> ok, so safe to skip
16:13:43 <smcginnis> Are there any tests defined?
16:14:02 <jaosorior> EmilienM: let me figure that out
16:14:21 <EmilienM> it is safe to skip yes
16:14:26 <jaosorior> EmilienM: chandankumar might know. He's not online though. so I'll ask him tomorrow
16:14:26 <ttx> let's say that we'll ignore it unless you tell us otherwise
16:14:29 <jaosorior> EmilienM: but yeah, lets skip it
16:14:31 <smcginnis> Looks like just the framework with no tests.
16:14:44 <smcginnis> I'm fine skipping it for now and reevaluating during train.
16:15:18 <smcginnis> Any objections to that plan or other thoughts?
16:15:42 <EmilienM> +2
16:16:12 <jaosorior> +1
16:16:23 <smcginnis> OK, thanks everyone.
16:16:28 <jaosorior> thank you!
16:16:33 <smcginnis> #topic Organize pre s-final-lib tasks
16:16:53 <smcginnis> #link https://github.com/openstack/releases/blob/master/doc/source/reference/process.rst#between-milestone-2-and-milestone-3
16:17:14 * EmilienM disappears in shadow
16:17:15 <ttx> So I copied a bunch to the Stein-3 - 2 week
16:17:31 <ttx> since they were marked "two weeks before"
16:17:53 <smcginnis> That looks good.
16:18:12 <smcginnis> The artifact signing key is an important one.
16:18:21 <ttx> We did That left two, one that you covered in a weekly email
16:18:33 <ttx> the other I left to discuss
16:18:42 <ttx> "Follow up with PTLs and liaisons for projects that missed the second milestone, or still haven’t done their library releases yet ?"
16:19:19 <ttx> Also for R-7 I won't be much around so it might be a good idea to pre-assign those tasks
16:19:32 <ttx> just in case people have questions on how to do them
16:19:39 <smcginnis> So we should probably change that wording in the process doc since the cycle-with-rc change makes "missing the second milestone" the norm.
16:20:01 <dhellmann> good point
16:20:05 <smcginnis> We have the automatic releases for libraries, but from the paste of a run I did this morning, looks like there maybe should have been a couple that were somehow missed.
16:20:22 <ttx> did we follow up with those intermediary who did not release before milestone-2 though
16:20:56 <smcginnis> I thought so? I should look again.
16:21:09 * ttx tries to remember if those should be switched to cycle-with-rc immediately, or after stein-3
16:21:42 <smcginnis> Probably after stein-3?
16:21:57 <ttx> hmm
16:22:05 <ttx> "Two weeks before milestone 3, warn cycle-with-intermediary projects that had changes over the cycle but no release yet that the release team will tag HEAD of master for their project if they have not prepared a release by the following week so that there is a fallback release to use for the cycle and as a place to create their stable branch."
16:22:25 <smcginnis> Well, it's possible there's a script issue, but "list-deliverables --unreleased --model cycle-with-intermediary" doesn't look like any missed unless I'm misinterpreting the output.
16:22:32 <dhellmann> step 5 under https://releases.openstack.org/reference/process.html#between-milestone-2-and-milestone-3
16:22:38 <ttx> not sure we documented that "switch to rc " policy
16:22:50 <smcginnis> I don't think so.
16:24:05 <ttx> smcginnis: I haven't tracked closely which emails of the new policy ended up being sent
16:24:12 <smcginnis> I've added a note for the R-7 countdown email to make sure to include that message.
16:24:14 <ttx> https://etherpad.openstack.org/p/stein-relmgt-auto-release-change has in "email 4":
16:24:30 <ttx> "To encourage (force?) this, we would require at least two releases of a non-library cycle-with-intermediary project during the cycle. If a release is not done by milestone 2, these projects would be switched to the new cycle-with-rc. "
16:25:35 <smcginnis> I thought we (I) had sent all of those, but now I'm wondering about that last one.
16:25:49 <dhellmann> wouldn't hurt to send a reminder, even if we did send that
16:26:03 <ttx> It's a bit late to send it if we haven't though
16:26:17 <smcginnis> Yeah
16:27:47 <dhellmann> I get a rather long list of things that haven't released
16:28:06 <ttx> not surprising
16:28:09 <dhellmann> although a bunch of those are tempest plugins so I should filter it more
16:28:14 <smcginnis> I guess so: http://paste.openstack.org/show/744320/
16:28:17 <dhellmann> 91 without filtering
16:28:38 <ttx> I feel like we missed our window
16:29:06 <dhellmann> we could change the deadline
16:29:16 <smcginnis> I agree. We can still propose to change these to cycle-with-rc and have a conversation with the teams over the patch for it.
16:29:17 <ttx> stein-3?
16:29:28 <ttx> So/// mandate at least one, and done before stein-3?
16:29:28 <dhellmann> list-deliverables --unreleased --model cycle-with-intermediary --type library --type client-library --type service  --type horizon-plugin -> 45
16:29:33 <dhellmann> yeah
16:29:40 <ttx> rather than mandate at least 2 with one done before stein-2
16:29:52 <ttx> and keep that "evolution" for Train
16:29:55 <smcginnis> I can include that in today's countdown email.
16:29:56 <dhellmann> well, at least 2 with 1 before s-3
16:30:27 <ttx> dhellmann: I think it's a bit late to announce that if we have not mentioned it before
16:30:40 <dhellmann> we should make sure we haven't actually mentioned it, I guess
16:30:47 * ttx is searching email archives
16:30:56 <dhellmann> there are only 10 unreleased services
16:31:10 <dhellmann> aodh, ceilometer, cloudkitty, magnum, panko, solum, tacker, tricircle, watcher, zun
16:32:26 <smcginnis> I can include that list in the countdown and refer to our stated changes that these should consider switching to cycle-with-rc.
16:32:41 <ttx> yeah can't find anything beyond email 1
16:33:03 <smcginnis> I know I sent at least up to 3.
16:33:36 <ttx> oh
16:33:43 <ttx> Proposed changes for cycle-with-intermediary services releases, December 3
16:33:57 <ttx> "If a release is not done by milestone 2, these projects would be switched to
16:33:59 <ttx> the new cycle-with-rc.
16:34:06 <ttx> so it was announced!
16:34:09 <ttx> and ignored
16:34:32 <dhellmann> so a reminder with a link to the patch making the change seems like a good idea
16:34:35 <evrardjp> I had the feeling it was, but no proof :)
16:34:52 <evrardjp> thanks for the grep skills ttx :p
16:34:59 <smcginnis> OK, so I think I can propose patches changing the deliverables in that list to -rc and add the PTLs to it. Then we can discuss on there.
16:35:11 <smcginnis> ttx: Do you have a date on that email?
16:35:20 <evrardjp> He said december3?
16:35:29 <smcginnis> Oh, sorry. :)
16:35:30 <ttx> Yep. I guess if someone replies with an immediate release request that's fine too
16:35:45 <dhellmann> sure, that works for me
16:35:53 <ttx> It's one of the first to be sent to -discuss
16:36:03 <ttx> maybe it was missed
16:36:11 <evrardjp> oh that's why I didn't find it in dev: )
16:36:13 <smcginnis> High probability of that.
16:36:24 <ttx> So some reminder email would not hurt
16:36:30 <smcginnis> I'll mention in the countdown, then get some patches up to change those.
16:36:48 <ttx> and offer the option to immediately do an intermediary release if they really don't want to be cycle-with-rc
16:36:58 <smcginnis> ++
16:37:15 <smcginnis> We said milestone 2, but I think we can be flexible.
16:37:45 <ttx> yeah, we did not really handle comms on that perfectly around milestone-2 :)
16:37:59 <dhellmann> yeah :-/
16:38:38 <ttx> that also points to missing doc in PROCESS
16:38:50 <ttx> we trusted PROCESS but it was not updated
16:39:00 <dhellmann> good point
16:39:00 <ttx> which is why we missed
16:39:12 <ttx> at least why I missed it
16:39:17 <smcginnis> Yeah, I think we need to go through that whole doc and make sure it's still accurate too.
16:39:45 <evrardjp> that sounds a good idea. Is it too late to do it at the next summit together?
16:39:45 <ttx> It's not, since it does not mention anythign about switching to cycle-with-rc after milestone-2
16:39:47 <dhellmann> yeah, we really need to review that every cycle
16:40:23 <ttx> that December 3 email is not reflected in there afaict
16:40:51 <ttx> ok, let's add that as tasks for R-9
16:41:20 <smcginnis> OK, thanks for adding that.
16:41:27 <smcginnis> Anything else we should cover on this topic?
16:42:45 <smcginnis> #topic Open discussion
16:42:51 <ttx> smcginnis: do the tasks I added for next week correctly cover the work that needs to be done?
16:43:09 <smcginnis> I think so. That covers the notes I had written down.
16:43:12 <dhellmann> I think that covers it
16:43:13 <dhellmann> ttx: we should make the change patch first so we can link to it in the email
16:43:32 <smcginnis> ++
16:43:40 <ttx> switched
16:43:58 <smcginnis> OK, anything else?
16:44:28 <dhellmann> nothing from me
16:44:29 <ttx> nope
16:44:35 <smcginnis> Have fund at FOSDEM.
16:44:38 <smcginnis> *fun
16:44:42 <smcginnis> And have funds...
16:44:56 <evrardjp> one is required for the other
16:45:08 <smcginnis> Alright, thanks everyone.
16:45:19 <ttx> Thanks smcginnis !
16:45:23 <smcginnis> #endmeeting