15:00:06 <smcginnis> #startmeeting releaseteam
15:00:06 <openstack> Meeting started Fri Jan 19 15:00:06 2018 UTC and is due to finish in 60 minutes.  The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:07 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:11 <openstack> The meeting name has been set to 'releaseteam'
15:00:16 <smcginnis> ping dhellmann, dims, fungi, tonyb, lbragstad, ttx
15:00:44 <smcginnis> armstrong: Want to add your nick to the ping list here? https://etherpad.openstack.org/p/queens-relmgt-tracking
15:00:47 <lbragstad> o/
15:01:08 <smcginnis> Morning lbragstad. Warming up there yet?
15:01:19 <lbragstad> not too bad
15:01:31 <lbragstad> nice enough to get out of the house and go do stuff :)
15:01:34 <armstrong> @smcginnis: Yes I want to
15:01:51 <smcginnis> lbragstad: Same here. It's no longer painful to step out of the house. :)
15:01:59 <fungi> i can finally see some of my driveway once again
15:02:18 <smcginnis> fungi: Oh right, you guys got hit, didn't you?
15:02:32 <fungi> second time in as many weeks
15:03:20 <smcginnis> Just going to wait one more minute to see if ttx is around.
15:04:34 <smcginnis> OK, we are currentl at R-6 (line 224) in the tracking etherpad: https://etherpad.openstack.org/p/queens-relmgt-tracking
15:04:40 <smcginnis> #topic Task report
15:04:55 <smcginnis> I think this was added by Thierry.
15:05:16 <smcginnis> I will follow up and see what's left to do with the ceilometer-powervm note.
15:05:58 <smcginnis> I am not aware of any other immediately pressing items, other than normal activities at this point in the cycle.
15:06:10 <smcginnis> Anyone have any tasks in progress that we should all be aware of?
15:06:33 <ttx> oops
15:06:39 <ttx> sorry for lateness
15:06:43 <ttx> managery-stuff
15:06:44 <smcginnis> Hey ttx.
15:06:56 <ttx> Yes, so task report
15:06:59 <smcginnis> "managery-stuff" == cake coma wearing off? :)
15:07:10 <ttx> Just wanted to flag an oddness
15:07:43 <ttx> The script that updates -release-branch group membership expects to be named $projectteamname-release-branch
15:07:55 <ttx> in the case of ceilometer-powervm that is Telemetry
15:08:14 <ttx> so it tries to update Telemetry-release-branch group
15:08:21 <ttx> Err
15:08:42 <ttx> I mean it tires to add Telemetry-release to Telemetry-release-branch
15:09:00 <ttx> But Telemetry-release doesn't exist. Only ceilometer-powervm-release
15:09:07 <fungi> can we just fix their acl and rename the group?
15:09:16 <smcginnis> So worth adding special handling to the script, or just expect the failure and manually follow up?
15:09:28 <ttx> That are teh 3 options yes
15:09:39 <ttx> 1 Fix ACLs/group names
15:09:51 <ttx> 2 Adding special handling in script
15:10:01 <ttx> 3 Expect failure and manually followup
15:10:03 <smcginnis> For consistency, I like 1.
15:10:05 <ttx> I did (30 this time around
15:10:11 <ttx> (3)
15:10:14 <fungi> note that a group rename takes a little care, but can be done without gerrit downtime (though it does need a gerrit admin to assist)
15:10:31 <smcginnis> Any other gotchas to be aware of with renaming?
15:10:50 <ttx> I mean, it's not urgent, just wanted to flag it before I forget about it
15:11:00 <ttx> so that we can add it to some future plan
15:11:25 <ttx> I think one issue here is that it's really a ceilometer-powervm specific group
15:11:27 <smcginnis> ttx: Want to add a note to https://etherpad.openstack.org/p/rocky-relmgt-plan while it's fresh in your mind?
15:11:43 <ttx> smcginnis: sounds good
15:11:51 <fungi> no real gotchas, process is pretty simple but have to take precautions to avoid racing our manage-projects script since it will create any groups referenced in an acl if they don't exist yet
15:12:40 <fungi> so basically stage the acl patch, pause configuration management, approve the patch and wait for it to merge, run a gerrit api command to rename the group itself, then unpause configuration management
15:13:29 <smcginnis> And the telemetry team would just need to be aware of the new group name for any future modifications, right?
15:13:55 <fungi> yeah, though honestly they likely won't notice since the group id number and stable url for it remain unchanged
15:14:15 <smcginnis> Doesn't sound too painful.
15:15:11 <ttx> also they likely don't care/ powervm is the only thing using that group (only thing using milestone-driven in Telemetry)
15:15:26 <fungi> however we've also got a project rename we need to schedule soon, so maybe it makes more sense to just roll the group rename into that
15:15:27 <smcginnis> OK, added those notes to the etherpad so we don't need to try to remember all that.
15:15:55 <ttx> hmm /me looks up ceilometer
15:15:56 <smcginnis> fungi: Oh good, that would be a safe opportunity to change it.
15:16:30 <ttx> yeah, switched to intermediary
15:16:36 <fungi> it's just eth one group which needs this? ceilometer-powervm-release -> telemetry-release?
15:16:46 <fungi> er, the one group
15:16:54 <ttx> Telemetry-release yes
15:17:27 <ttx> with a capital T
15:17:43 <fungi> oh, weird. why is that one capitalized?
15:17:59 <smcginnis> It's a special snowflake. :)
15:18:01 <ttx> The script uses the project team name as in projects.yaml
15:18:20 <ttx> and yes it's the only capitalized team name using milestone-driven deliverables
15:18:37 <fungi> but, like, nova-release isn't capitalized even though projects.yaml says their team name is Nova
15:18:46 <ttx> hmmm
15:18:49 <fungi> or is that not the case?
15:18:58 <ttx> MAYBE it's case-insensitive search
15:19:25 <smcginnis> Ah, and nova already exists, so it doesn't make a difference?
15:19:30 <ttx> Like when you search for Nova-release it returns existing nova-release
15:19:47 <ttx> wouldn't surprise me at all.
15:20:11 <smcginnis> ttx: Just for reference, which is the script in question?
15:20:19 <fungi> oh, i think i see what you're saying. the top-level keys in reference/projects.yaml being capitalized or not
15:20:21 <ttx> aclmanager
15:20:30 <ttx> fungi: yes
15:20:31 <smcginnis> ok
15:20:38 <fungi> and yeah, "nova" is the key for the nova team's entry in projects.yaml
15:20:41 <fungi> lower-case
15:20:44 <fungi> okay
15:21:05 <ttx> it's basically the only team that is not named after its only deliverable
15:21:19 <ttx> so it obviously triggers edge cases
15:21:54 <fungi> i find the case mixing in that file strange, but i'll refrain from letting it overwhelm my ocd tendencies
15:22:00 <smcginnis> :)
15:22:25 <smcginnis> fungi: Do you know when that rename is scheduled?
15:22:55 <fungi> smcginnis: we haven't set the schedule yet. still trying to work out the new project rename process with zuulv3
15:23:24 <fungi> distributed job configuration means potentially more places to update repository name references
15:23:32 <smcginnis> fungi: OK. We'll have to keep this in mind when that comes up then.
15:23:39 <smcginnis> Any more on this topic?
15:24:01 <ttx> nope
15:24:13 <smcginnis> #action team to try to coordinate with project name restarts to fix acl and group name for ceilometer-powervm
15:24:16 <smcginnis> #topic Pike-3/FF tunnel coming up
15:24:32 <smcginnis> #link https://etherpad.openstack.org/p/queens-relmgt-plan Planned tasks
15:25:00 <ttx> I added that one too -- I think it's a good time to check how much we'll defer to R
15:25:01 <smcginnis> So yeah, I thought I had taken care of most of mine, but when I started updating for rocky I realized there were several things in there I overlooked.
15:25:39 <smcginnis> I think there may be a few things that we've just taken care of without noting there, but I noticed a few things that definitely still require some action.
15:27:02 * ttx updates a couple things
15:27:32 <ttx> So the only thing I missed was "TODO Support storyboard in openstack/releases (ttx)"
15:27:51 <ttx> I don't expect to close it in the coming weeks
15:27:59 <smcginnis> So I guess action for all of us - please look through the list again and see what can be done with anything with your name on it.
15:28:01 <ttx> I have a bit of backlog on my coding time
15:28:05 <fungi> i saw some patches go by related to that, i thought
15:28:32 <fungi> though i didn't recognize the submitter's name
15:28:32 <smcginnis> There's a patch out there now, but it has a -1.
15:28:37 <fungi> ahh
15:29:33 <ttx> so yeah, check what you can still complete before we go too deep in release mode
15:29:49 <ttx> and mark the others as DEFER
15:30:00 <smcginnis> #action team to review tasks in planning etherpad
15:30:07 <fungi> planning to do the rocky cycle release signing key while at the ptg
15:30:28 <fungi> since it's a good task to involve other infra people on
15:30:41 <smcginnis> I've copied over everything to the rocky planning etherpad, but will re-sync that with wherever we end up by the end of the cycle.
15:30:47 <smcginnis> fungi: ++
15:31:03 <smcginnis> #topic Review FF week actions in PROCESS
15:31:33 <smcginnis> #link http://git.openstack.org/cgit/openstack/releases/tree/PROCESS.rst
15:31:57 <ttx> So I won't have a lot of time mnext week, just wanted to check that you guys have the process udner control for next week
15:32:26 <ttx> did we finish all the pre-M3 tasks ?
15:33:07 <ttx> this week was the lib freeze I see
15:33:43 <ttx> was the step on line 150 done ?
15:33:49 <smcginnis> Just looking at that. No.
15:33:53 <smcginnis> I will do that after the meeting.
15:34:08 <smcginnis> #action smcginnis to propose feature list update to devstack-gate
15:34:15 <ttx> did we have any "final release
15:34:18 <ttx> "
15:34:36 <ttx> since we are supposed to ensure they come with a stable branch request
15:35:06 <ttx> I suspect we did not, since the new branch name would have likely failed tests :)
15:35:35 <smcginnis> I was looking for that, but will be honest I may have forgotten on a couple.
15:36:04 <smcginnis> Hmm, no. Guess I'll need to revisit that.
15:36:24 <ttx> It looks like there is no step to create stable/queens for everyone
15:36:45 <smcginnis> IIRC, don't we have a script that will do that too?
15:37:09 <ttx> yes I think we do, it just doesn't appear here
15:37:23 <ttx> probably something to catch up with dhellmann on Monday
15:37:31 <smcginnis> propopse_library_branches.py in openstack_releases/cmds it appears.
15:37:36 <smcginnis> ack
15:37:49 <ttx> as we should probably run that after the patches from line 150 are done
15:38:12 <smcginnis> OK, I'll get 150 taken care of today, then Monday will follow up to make sure the branching is done.
15:38:22 <ttx> Line 165 will be fun
15:38:39 * smcginnis is afraid to look
15:39:20 <ttx> otherwise FF week doesn't have that many more tasks than FF-1 week
15:40:04 <smcginnis> We should be in good shape.
15:40:09 <ttx> smcginnis: ok
15:40:23 <smcginnis> #topic Rocky PTL
15:40:38 <ttx> smcginnis: you're running again right
15:40:49 <smcginnis> I feel like I'm just starting to get into things, so I'm fine going another cycle unless someone wants to boot me.
15:40:55 <ttx> We do one-year PTLships around here
15:40:59 <smcginnis> :)
15:41:02 <smcginnis> Works for me.
15:41:21 <ttx> ok, next topic :)
15:41:28 <fungi> in preparation for the year-long release cycle? ;)
15:41:36 <smcginnis> Darn, I suppose I still need to think of something to write up.
15:41:45 <ttx> been toying with the idea of 9-month cycles lately
15:41:53 <ttx> :)
15:42:18 <smcginnis> There is something nice about cycles fitting within year blocks, but I don't want to go down that path right now. ;)
15:42:25 <smcginnis> #topic PTG track assignment
15:42:36 <ttx> So I posted the proposed layout
15:42:43 <ttx> RelMgt is one of the "orange" teams
15:42:57 <fungi> seems dangerous in dublin
15:42:58 <ttx> meaning we are encouraged to schedule ourselves when we happen to be available and free
15:43:08 <ttx> fungi: nice
15:43:13 <smcginnis> #link http://lists.openstack.org/pipermail/openstack-dev/2018-January/126335.html
15:43:24 <ttx> I think that's a great thing for us, matches how we did it in Denver
15:43:33 <ttx> and avoids uselessly booking space
15:43:39 <smcginnis> Yeah, that seemed to work well.
15:43:51 <smcginnis> And we all have so many other things going on, it was nice having the flexibility.
15:43:55 <ttx> Note that we can also schedule MORE than 0.5 days if we need extra time
15:44:03 <smcginnis> Still way too many times I needed to be in more than one place.
15:44:05 <ttx> or LESS
15:44:08 <fungi> sure, just grab me if/when you need me though i expect to be split at least 4 ways most of the week
15:44:32 <ttx> that is all
15:44:46 <smcginnis> Good plan. Looking forward to seeing everyone in Dublin again.
15:44:49 <ttx> smcginnis: shall we post a PTG prep  etherpad
15:45:06 <smcginnis> ttx: Yeah, that wouldn't hurt. I'll take another action.
15:45:25 <smcginnis> #action smcginnis to add relmgt PTG etherpad
15:45:32 <smcginnis> OK, anything else?
15:45:57 <smcginnis> #topic Open discussion
15:46:11 <smcginnis> I started next weeks email draft, but I still need to work on it.
15:46:11 <ttx> nope
15:46:21 <smcginnis> If there's anything else anyone wants to add there, feel free: https://etherpad.openstack.org/p/queens-email-r4
15:46:42 <smcginnis> OK, otherwise, I guess we can wrap for today.
15:46:53 <smcginnis> Thanks everyone.
15:46:58 <ttx> smcginnis: thanks !
15:46:59 <lbragstad> o/
15:47:05 <smcginnis> #endmeeting