15:00:21 #startmeeting releaseteam 15:00:22 Meeting started Fri Aug 25 15:00:21 2017 UTC and is due to finish in 60 minutes. The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:23 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:25 The meeting name has been set to 'releaseteam' 15:00:36 Ping list: dims, fungi, tonyb, stevemar, lbragstad 15:00:54 o/ 15:01:06 lbragstad: Hey Lance! :) 15:01:18 #link https://etherpad.openstack.org/p/pike-relmgt-tracking 15:01:38 Agenda for the meeting is the R-1 week section 15:02:08 o/ 15:02:25 dhellmann: Just noticed I missed you in the ping list. Sorry about that. 15:03:40 np 15:04:16 how're things looking with rcs? 15:04:28 I think I saw that we're expecting one more from neutron? 15:04:41 dhellmann: Sorry just looking at that. 15:04:49 Neutron is the only one I'm aware of so far. 15:05:20 we probably want to wait until after that before setting up the patch to propose the final release tags 15:05:38 just to make sure we don't make a mistake and tag the wrong release 15:05:59 Is there an example patch for that from ocata I can take a look at? 15:06:13 there's a tool to create the patch, check PROCESS.rst 15:06:16 * smcginnis needs to take a look at the docs again 15:06:29 it's something like propose-final-release I think 15:06:31 Hah, yep. Just thought of that after I asked. :) 15:06:39 I can never remember the exact names for these things :-) 15:06:51 Good thing everything is written down. ;) 15:06:52 but yeah, the tool creates the patch for you, then we just need to get the PTLs and liaisons to +1 15:07:09 then we all +2 and ttx can W+1 15:07:37 So all PTLs for all projects? 15:08:04 it's anyone who wants to sign on, but the patch will only contain changes for the milestone projects 15:08:21 and we don't *need* them to, but it's a nice gesture, since the names all go into the tag metadata 15:08:53 Very true. 15:09:22 #topic Missing intemediary releases 15:09:31 Did I see senlin go through? 15:09:53 list-deliverables shows no unreleased projects 15:10:05 there's an rc for senlin 15:10:26 OK, thought I at least saw a patch there. 15:10:31 and list-deliverables shows no projects missing rcs 15:10:52 it's not smart enough to tell us if there are any projects with really old releases, though 15:10:59 Does that also show whether a stable/pike branch was created? 15:11:09 it can, let me run that 15:11:42 http://paste.openstack.org/show/619437/ 15:11:46 5 projects 15:11:49 tempest doesn't branch 15:11:51 #topic Missing stable branch 15:11:57 I don't know about patrole 15:12:08 the others are cycle-trailing, so I think it's ok for them to not have branched yet 15:12:21 patrole is part of/related to tempest I think? 15:12:29 it's a qa thing, but I don't know what it is 15:12:43 "rbac integration tempest plugin" 15:12:46 so that may not branch either 15:12:55 Ah: Tempest plugin for testing and verifying..." 15:13:07 And the others there are cycle-trailing. 15:13:26 So we're in good shape for branches too it appears. 15:13:28 I should add a stable-branch-type for "branchless" so we can skip those 15:13:32 in the query, that is 15:13:36 ++ 15:14:00 We should have channel bots that record anything that starts with "I should..." 15:14:32 or someone else could write that patch to learn how to make those sorts of changes :-) 15:14:39 haha, I like the bot idea 15:14:56 Was that a I detected? 15:14:56 "someone should" could add to a wishlist 15:15:06 yes, that's an elbow you feel in your side 15:15:08 Ooh, I like where this is going. 15:15:22 The ShouldBot. 15:15:27 perfect name 15:15:48 I'll put it on my list to check out. 15:15:57 it could either add it to a list, or it could just reply "you're someone!" 15:16:17 #action smcginnis to look into "branchless" stable-branch-type query support 15:16:25 Hah, love it. 15:16:40 #topic Assign R-0 week tasks 15:17:49 Approve the final release patch created earlier - that patch will be created once everything is finally through, if I understand correctly. 15:18:13 yeah, we should create that early Monday 15:18:40 OK, I can take creating the patch to get the experience there. 15:19:19 dhellmann: Oh, I see that was something with your name on it for this weeks actions. 15:19:53 Did you want to create the patch? 15:20:02 oh, sure, I can do that 15:20:06 I forgot I signed up for that 15:20:23 Either that or you were signed up while you were on vacation. :) 15:20:29 that sounds likely 15:20:55 Reset gerrit ACLs - what needs to be done there? 15:21:17 I'm signed up for release testing, but I think since we've done releases pretty recently it's safe to skip that as a separate step 15:21:18 Looks like another that is scripted - yay 15:21:25 yeah, there are 2 pieces there 15:21:35 dhellmann: Yep, I agree on release testing. 15:21:36 change the group membership and change the gerrit acls files 15:22:19 Did we have something with ACLs for telemetry? 15:22:55 I'm not sure? 15:22:59 I had taken an action at some point to talk to jd about an issue there, but it's still lower on my list. 15:23:32 oh, I vaguely remember something there. 15:23:33 maybe they weren't set up the same way as the other projects, so the scripted stuff wasn't working? 15:23:56 I thought it was something like they did their own tagging. 15:24:06 I need to go back and remember too. 15:24:08 that might be it, yeah 15:24:25 So would this script fix the glitch for us? :) 15:25:22 no, I think because of the way it's set up the script skips it 15:25:29 we'll have to see, though 15:25:39 OK. I'll circle back to that later then. 15:25:57 Looks like the two doc items on the action list are no longer needed after the changes there? 15:26:32 we need to do 0.5 but not 0.6 15:27:04 we can add the doc link today, in fact 15:27:43 I can take that one unless someone else wants it. 15:28:12 we should update the process doc to remove that step for the future, because as soon as we open the series on releases.o.o the docs page should exist on docs.o.o so we can just always have the link present 15:28:15 go for it 15:28:28 I'm working with the docs team on 0.6 15:28:35 OK, great. 15:29:06 the patches for 0.7 and 0.8 can be stacked up on top of the final release patch, so I'll do that 15:29:20 Since you're so deeply involved in the docs work, do you want to update the process doc to reflect the changes? 15:29:27 yeah, I'll take that 15:29:42 Cool, thanks. 15:29:56 And ttx should probably take the final release announcement? 15:30:07 0.9 15:30:27 yeah 15:30:39 and 0.10 15:30:43 And 0.10 probably tonyb I would assume 15:30:46 Oh, OK. 15:30:58 yeah, that's the releases repo not the requirements repo 15:31:13 Oh, right. I read that too fast. 15:31:27 * smcginnis will go with that instead of admitting his eyes are getting bad 15:31:43 somewhere in here we should approve the open stable releases, and that's something for tonyb 15:32:12 I don't see it on R+1 or R+2, it should go on one of those weeks 15:32:19 don't steal the thunder of the final pike releases 15:32:27 and don't confuse people about what versions are part of what series 15:32:29 Hah, good point. :) 15:32:43 R+1 probably 15:32:54 that way the releases in R+2 are also on their own 15:33:02 Just want it noted somewhere so it's on the radar. 15:33:11 ++ 15:33:29 Great, that's it for tasks. 15:33:48 Should one of us (or me) send out a countdown email yet? 15:34:11 yeah, if we haven't done that this week it would be good to remind folks 15:34:19 let's see what we should say 15:35:02 Oh, there was one for R-1: http://lists.openstack.org/pipermail/openstack-dev/2017-August/121266.html 15:35:43 ah, ok, then I think we're set 15:35:56 those other things can be handled with a separate email early next week 15:36:06 OK, that sounds like a plan to me. 15:36:14 Anything else to cover? 15:36:19 nothing from me 15:36:32 nothing from me 15:36:50 Alright. Thanks everyone. Enjoy your weekend. 15:36:54 o/ 15:36:55 thanks, you too! 15:36:58 #endmeeting