15:02:26 #startmeeting releaseteam 15:02:27 Meeting started Fri Jan 20 15:02:26 2017 UTC and is due to finish in 60 minutes. The chair is dhellmann. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:02:29 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:02:32 The meeting name has been set to 'releaseteam' 15:02:45 this is week R-5 15:03:32 our agenda is in the etherpad, as ususal 15:03:34 #link https://etherpad.openstack.org/p/ocata-relmgt-tracking 15:03:48 #topic review late releasing libraries 15:04:03 looking in the inbox, I only see a client lib so that can wait 15:04:16 and the oslo branches can be processed when harlowja +1s that patch 15:04:27 any idea if we have cycle-with-intermediary libraries that weren't refreshed within ocata ? 15:04:28 o/ 15:04:32 and do we care ? 15:04:48 ttx: so far only client libs, I sent a list to the ML today 15:05:01 http://lists.openstack.org/pipermail/openstack-dev/2017-January/110582.html 15:05:23 oops, forgot to courtesy ping ttx, dims, sigmavirus, fungi, stevemar 15:05:31 everything else was refreshed at least once ? 15:05:35 o/ 15:05:42 ttx: yes, according to list-deliverables 15:05:52 let's trust list-deliverables! 15:06:12 all of the things that existed in newton had stub data files copied to the ocata dir, and list-deliverables produced the list in the email linked above 15:06:28 not bad, given ocata is short and a bit low 15:06:33 yeah 15:06:46 I'd like to figure out a way to notice that clients aren't releasing earlier in the cycle, though 15:06:57 maybe add a step to the process at milestones or something 15:07:06 we coudl introduce a notive by milestone-2 15:07:11 notice 15:07:15 good idea 15:07:20 * ttx writes it down 15:07:39 ideally we wouldn't have patches to libs go unreleased for more than a week 15:08:11 because if a release breaks something, and the change is 2 months old, it's not going to be fresh in anyone's mind for debugging 15:08:25 dhellmann: ++ 15:08:38 maybe a periodic job could send a list out 15:08:47 maybe a week is a bit short, but yes 15:08:48 we have the tools to produce the list 15:09:16 anyway, ideas for next cycle 15:09:18 moving on 15:09:20 #topic team availability next week for b3 tagging 15:09:41 it's not a very big concern, but I wanted to make sure I knew who would be around next week 15:09:50 i'll be around dhellmann 15:09:53 I expect to be, though it's possible I'll be working remotely 15:09:58 I'll be around 15:10:15 fungi : does the infra team have someone lined up to fight fires for the feature freeze deadline? 15:10:16 travel madness starts Saturday morning next week 15:10:18 i may be off for some time on 26th 15:10:37 ttx : where are you off to? 15:10:40 dims : ack, noted 15:10:48 so many places I can't list them all 15:10:53 oof 15:10:56 safe travels :) 15:11:16 because everyone knows January/February are the best months to travel 15:11:28 i'll be sequestered january 29 through february 3, but am otherwise available 15:11:38 at least up until ptg time 15:11:46 "sequestered" sounds ominous 15:11:52 but accurate 15:12:25 ok then 15:12:32 but yeah, next week you can count on me to focus on any release-blocking fires 15:12:40 cool, thanks 15:12:40 thanks fungi 15:12:53 ttx, I think you're up next here 15:12:55 #topic Review remaining tasks in Ocata plan 15:12:58 yep 15:13:08 I listed the things still in Ocata plan 15:13:20 want to discuss if they should stay there and be completed in the coming month 15:13:22 or moved to pike 15:13:24 or abandoned 15:13:51 #link https://etherpad.openstack.org/p/ocata-relmgt-plan 15:14:03 First one is Automation 7 "we need to add to the checklist for marking a release EOL to update the info on releases.o.o" 15:14:22 ok, let's comment on the etherpad directly 15:18:05 OK, so a couple TODOs and otherwise add to backlog for final call in ATL 15:18:12 sounds good 15:18:14 Nothing needing to be completed by EOC 15:18:24 except the one HIGH TDO 15:18:33 the pike release schedule? 15:18:35 #action ttx to move last TODOs to Pike plan 15:18:38 yes 15:18:40 next topic 15:18:48 yes, I'm getting questions about that more often 15:20:00 is there anything else on this topic? 15:20:03 no 15:20:06 ok 15:20:09 #topic open discussion 15:20:15 that's all we had on the formal agenda for today 15:20:21 does anyone have anything else to bring up? 15:20:22 err no there was one more 15:20:28 ? 15:20:33 we use "security-supported" on https://releases.openstack.org/, do we define it anywhere (exactly?) 15:20:41 HIGH TODO status review - Define and publish Pike release schedule 15:20:50 oops 15:20:51 #undo 15:20:52 Removing item from minutes: #topic open discussion 15:20:58 #topic pike schedule 15:21:02 So we are now 60% sure that we can have PTG2 during Sept11 week (rather than Sept18) so I think we can start working on a strawman schedule 15:21:07 I misread the indentation level there 15:21:20 in the spare time between release work next week 15:21:38 try to see where we can make milestones fall wrt. events / holidays and all 15:21:54 Release August 31, Ff July 27, RC1 Aug 10 15:22:00 oh, good 15:22:10 It's not sure at all though 15:22:19 It's just that it's the most likely option at this stage 15:22:25 used to be 40/40/20 15:22:31 Now it's 60/40/0 15:22:48 but I think strong enough that we can prepare something 15:23:03 should we prepare copies of both schedules, or focus on 11 sept? 15:23:29 I guess we could prepare two versions 15:23:49 it's only a week, they won't be *that* different 15:23:57 won't work on it today anyway, and might have more news early next week 15:23:58 i have a feeling it wouldn't look too different as they're only a week offset 15:24:01 ok 15:24:17 let's just plan to spend a few minutes into strawmen next week 15:24:21 declare an extra rc week or extra dead week or something 15:24:22 let's focus on the slightly shorter schedule. no one is going to complain if we end up giving them an extra week. 15:24:46 that is all I had 15:24:47 ttx: do you want to take that on, or should I start it? 15:25:08 I can work on it you can break it down 15:25:12 ok 15:25:16 takes two to play strawman 15:25:21 ++ 15:25:29 #topic open discussion 15:25:35 dims: support phases are defined (at least) in the stable branches section of the project team guide 15:25:36 dims, you had a question about security-supported 15:25:38 #link http://docs.openstack.org/project-team-guide/stable-branches.html#support-phases 15:25:44 ah, yes, thanks, fungi 15:25:58 we should probably link to that 15:26:05 sgtm 15:26:16 I'll throw together a patch 15:26:36 fungi : dhellmann : does "security-supported" mean phase II or III? 15:26:58 "Current stable release, security-supported" == Phase II? 15:27:02 phase 3 is when we only support in a security capacity 15:27:11 Security-supported == Phase III? 15:27:25 so depending on how you interpret the term, security support would be from phase 1 through 3 15:27:51 fungi : dhellmann : so we should publish phase number in the https://releases.openstack.org/ ? 15:27:53 critical fix support is provided only in phases 1 and 2 15:27:54 it would be good of the 2 sites used the same terminology 15:28:06 right 15:28:12 yeah, i agree we should settle on one term and not use conflicting models for the same concepts 15:28:22 ++ 15:28:26 maybe we can update the PTG to use the more descriptive terms? 15:28:33 or combine them somehow? 15:28:42 or stop using confusing acronyms :) 15:29:10 we should combine the project team guide with the project teams gathering 15:29:31 http://www.writethedocs.org 15:29:35 LOL 15:29:49 fungi: I thought they were the same thing 15:30:09 alright, I'll take a stab at clarifying/unifying the guide and the releases site early next week 15:30:16 thanks dhellmann 15:30:27 * ttx jumps to next meeting 15:30:37 I think that's it, then 15:30:42 thanks, everyone! 15:31:07 #endmeeting