Monday, 2015-11-23

*** pixelbeat has joined #openstack-stable00:48
*** ihrachys has joined #openstack-stable00:55
*** ihrachys has quit IRC01:00
*** ihrachys has joined #openstack-stable02:27
*** ihrachys has quit IRC02:31
*** pixelbeat has quit IRC02:33
*** ihrachys has joined #openstack-stable02:43
*** ihrachys has quit IRC02:47
*** ihrachys has joined #openstack-stable03:23
*** ihrachys has quit IRC03:27
*** agireud has quit IRC04:18
*** agireud has joined #openstack-stable04:20
*** agireud has quit IRC04:25
*** agireud has joined #openstack-stable04:46
*** rcernin has joined #openstack-stable05:36
*** rcernin has quit IRC05:42
*** rcernin has joined #openstack-stable07:07
*** openstack has joined #openstack-stable07:18
ttxStable team meeting today at 15:00 UTC on #openstack-meeting-4 to discuss proposing a separate team. Feel free to join09:05
*** ihrachys has joined #openstack-stable09:15
*** e0ne has joined #openstack-stable09:28
*** derekh has joined #openstack-stable09:40
*** flaper87 has quit IRC10:46
*** flaper87 has joined #openstack-stable10:46
*** pixelbeat has joined #openstack-stable10:53
*** armax has joined #openstack-stable13:06
*** armax has quit IRC13:10
*** doug-fish has joined #openstack-stable13:25
*** eharney has joined #openstack-stable13:53
*** eharney has quit IRC14:26
*** mriedem has joined #openstack-stable14:38
*** eharney has joined #openstack-stable14:39
*** mriedem has quit IRC14:52
*** mriedem has joined #openstack-stable14:57
mriedemhuzzah15:03
flaper87o/15:03
*** mestery has joined #openstack-stable15:03
Jokke_o/15:03
ttxI'm pretty sure they didn't book that slot. I'll flame them later15:03
dhellmanno/15:03
ttxSo15:04
dhellmannttx: they're 12 hours off, according to my calendar15:04
ttxteam scope15:04
ttx1. Policy enforcement and definition15:04
ttxThat includes defining and applying tags, since they are the weapon we can wield15:05
ttxThe idea being to try to make sure teams that pretend to follow policy are actually following policy15:05
mriedemi agree with that one15:05
Jokke_++15:05
ttxcould be some periodc review, can't be all the time15:05
dhellmannapuimedo: ++15:05
mriedemi usually gauge by ML shame15:06
dhellmannoops, just ++15:06
ttx2. Help projects15:06
*** jordanP has joined #openstack-stable15:06
ttxThat's answering questions, encouraging releases, doing doc...)15:06
ttx3. Keep CI working on stable15:06
ttxThat's pretty essential, I think the team can't work if they don't take that one15:07
flaper87number #3 takes quite some time and it's super important!15:07
ttxyeah, number doesn't mean priority15:07
mriedem3 also ties into 115:07
mriedemb/c if 1 sucks, 3 suffers15:07
ttx4. Mentoring/growing the team (shamelessly stolen from mriedem's priorities)15:08
ttx5. Improve tooling/automation15:08
mriedemfin?15:09
ttxThat's all I had. In particular, deciding on stable support timeframes is *not* in team scope. That needs alignment between Infra, QA and Stable, and would be arbitrated at TC level if there is disagreement15:09
flaper87One of the things I'd like to see coming out of this new team is #515:09
Jokke_ttx: +++15:09
ttxSo am I missing something ?15:09
flaper87It'd be super cool for the PTL of this team to lead those efforts15:09
mriedemi'm happy to hear about timeframes not being in scope15:09
ttxmriedem: I think it's good to explicitly spell it out, otherwise the team will be under a lot of pressure15:10
dhellmanndeciding time frames isn't, but enforcing might be (coordinating with the infra team)15:10
mriedemyes, i'm already getting direct emails on keeping juno around15:10
ttxmriedem: I think I have a good idea of where those come from15:10
ttxanyway, if we agree on team scope I think we can move on15:11
mriedemi think we have 2 main gaps for #5 so agree with flaper87 there15:11
dhellmannthat's a good list of initial priorities15:11
mriedemyeah i think those are good points15:11
Jokke_I'd like to slam 2 and 4 together ... I really think inclusion and through that mentoring and growing the stable team would be the key to also grow the part of that team that does the gate fixing15:11
ttxJokke_: yes, the pool of $project-stable-maint people is the natural recuitment ground15:11
ttxSo next topic is membership / voters15:12
ttxWhat would constitute "contribution" to the team and grant voters one vote for PTL elections ?15:12
ttxI'm fine including all $project-stable-maint + stable-maint-core15:12
ttxalthough that feels a bit self-selected15:13
Jokke_ttx: I'd like to see that happening at the beginning15:13
mriedemdoes stable-maint-core stay a separate thing? i assumed it would.15:13
dhellmannttx: landing a patch on a stable branch of any project managed by the team?15:13
bknudsonyou could also include anyone that's proposed a patch to stable15:13
dhellmannbknudson : jinkx15:13
flaper87mmh, what about we start small and open for volunteers ?15:14
mriedemyeah i'm not sure if we're talking about stable-maint-core right now or voting rights?15:14
mriedemsince i think those are different thigns15:14
Jokke_bknudson: I was thinking of that as well, but that just could send wrong signal15:14
mriedem*things15:14
ttxmriedem: stable-maint-core... that's more of a failsafe to be able to quickly land gate-unblocking patches and -2 obviously-wrong things when people escalate questions to the team15:14
Jokke_mriedem: I think we're talking about stuff like who can vote for the Stable PTL15:15
ttxmriedem: should definitely be refactored15:15
ttxJokke_: +115:15
mriedemttx: ok, i wasn't sure if you were suggesting we make all $project-stable-maint part of stable-maint-core now15:15
ttxmriedem: oh no, membership to the team as in "who votes"15:15
mriedemok, then yeah, +1 on that15:15
ttxnot membership to stable-maint-core15:16
Jokke_++15:16
ttx(stable-maint-core destiny shall be decided later, it's a tool)15:16
ttxSo... authors of stable patches on managed teams ?15:16
ttxOne issue with that is...15:17
ttxIt means we have "managed teams"15:17
ttxI thought we would only have "compliant teams"15:17
mriedemright....nova-docker has a stable/icehouse still...15:17
mriedembut would not work15:17
ttxcould be "any stable patch on any compliant team"15:18
mriedemand compliant teams are tagged in the governance repo?15:18
dhellmannttx: sure, teams with the tag managed by the stable team15:18
Jokke_so which way we want to look this? Do we want to include people who got their patch backported by someone to the stable team, or people who actively backports suitable patches15:18
ttxor $project-stable-maint + at least 2 patches for a "compliant team"15:18
dhellmannJokke_ : the person doing the backport15:18
flaper87latest sounds better15:18
Jokke_so committer, not author15:19
flaper87patch owner15:19
ttxyeah, we always look at owner15:19
* dhellmann isn't sure if owner==committer or owner==author15:19
ttxit's a separate concept15:19
dhellmannk15:19
mriedemcommitter i think15:19
Jokke_ok ... that works, just wanted to confirm as author stays original with backports15:19
ttxgerrit owner vs. git author/committer15:20
Jokke_dhellmann: you cannot make either assumption15:20
ttxwould one backport be enough to vote ? Slightly concerned about crazy numbers of voters here15:20
mriedemthat's how ATCs are handled, so..15:20
ttxyeah, fair15:21
mriedemseems if you're going to open it up to people that backport and land a change,15:21
mriedemit's the same15:21
Jokke_I'd like to see bit more activity, so we don't get just those drive by backporters hoping for voting rights15:21
bknudsonI don't think it's going to be a large number based on keystone15:21
bknudsonIt might just be OpenStack Proposal Bot that gets to vote.15:21
mriedemyeah, honestly, most drive by backport people are not even knowing this is a thing we're discussing15:21
stevemari'd agree with that statement mriedem15:22
ttxso... until we get the tags straight, I propose we use release:managed as a substitute for stable:compliant15:22
ttxif we need to run an election now for example15:22
Jokke_mriedem: I'm more worried company XYZ pushing every team member to do a backport so they get maximum amount of voting power15:23
mriedemhow far back does the backport thing go? active stable branches? or also EOL?15:23
ttxJokke_: we'd likely be able to spot that15:23
bknudsonthat would be awesome if a company got interested in backporting15:23
stevemarbknudson: hehe15:23
ttxmriedem: active stable branches15:23
mriedemok15:23
ttxmriedem: so at the end of liberty elections, we'd do stable/liberty stable/kilo15:24
ttxerr15:24
dhellmannJokke_ : they have to *land* the backports15:24
ttxat the N elections at the end of the mitaka cycle we'd do stable/liberty and stable/kilo15:25
Jokke_so do we limit the membership to 1 cycle and all supported stable branches in it?15:25
ttxelections are at the end of a cycle, so current stable branches then15:25
bknudsonI prefer all active. There might just not be a lot of fixes required for 1 cycle back.15:25
ttxall active15:25
mriedem+115:26
dhellmannyeah, it would be simpler to say all active branches15:26
Jokke_ttx: so for N stable/[liberty,kilo,juno] within mitaka cycle15:26
ttxOK, agreed: membership is recent owners of backports on any stable branch active by election time15:26
ttxJokke_: I wouldn't count Juno, but I think it's a detail15:27
ttxI guess we could say "active during the cycle"15:27
mriedemheh, i guess it depends on when you hold the election15:27
Jokke_I'd like that as there is overlap between start of cycle and EOL15:28
ttxmembership is last 6 months worth of backport owners on active stable branches15:28
Jokke_that++15:28
ttxby definition if you did the backport then those were active branches15:28
dhellmannhow far back do we go for other teams?15:28
ttxdhellmann: 6 months15:28
dhellmannk15:28
flaper87++15:28
ttxerr, checking15:29
* flaper87 has to drop now!15:29
dhellmannare we at risk of having a stable branch removed before the electorate is calculated?15:29
flaper87super happy to see this team kicking off!15:29
* Daviey_ is here now... Reading scrollback15:29
ttxdhellmann: actually it's "last 2 cycles"15:29
mriedemDaviey_: are you connected via OnStar?15:29
dhellmannttx: that's what I thought I remembered15:29
ttx"Members who committed a change to a repository of a project over the last two15:29
ttx6-month release cycles are considered APC for that project team."15:29
ttxoh, well, make that a year then15:29
dhellmannso even after we remove the stable/juno branch at the end of november we'll need a way to determine who contributed patches to it so we can produce the electorate for the N cycle15:30
ttxdhellmann: the branch may be removed, the data is still there thanks to the tag ?15:30
dhellmannI think that's right, yes15:30
ttxanyway, technical detail15:30
bknudsonmaster..juno-eol15:31
dhellmannI'm not sure how fungi's scripts do that now. It may not be a big deal.15:31
ttxAlright so... membership is last year worth of backport owners on active stable branches15:31
ttxNext topic is PTL candidates. Two people stepped up for leading the team, Jokke_ and mriedem15:32
* dhellmann is impressed with the amount of interest15:33
ttxI think both would do great, although I think mriedem experience with fixing stable gate might be critically useful as we set up the team, to set the tone there15:33
Daviey_Two great candidates!  but i agree that mriedem's visibility and willingness to own technical issues is +1.15:33
ttxWe can hold an election if both want to run15:34
ttxI'd say Jokke_ has more experience with stable policy, but that's more of a traditional skill of the team15:35
ttxso I'm slightly less worried on that side15:36
Jokke_I'm still up for it and gotta say that which ever way it turn at least I will be happy member of the new team :)15:36
Daviey_Seems reasonable to hold an election.15:36
ttxtotally15:36
Jokke_I'm really happy that I wasn't the only one15:36
ttxOK, I can call for other candidates and try to set up something15:37
ttxor call for candidates and have you post, for reference15:38
dhellmannttx: ++ to having each candidate post for reference15:38
ttxWondering about the order now. Should we get the team blessed first ? Or have a PTL ?15:39
ttxprobably have a ptl15:39
dhellmannttx: one of the requirements for an official team is having a leader15:39
dhellmannit's not clear how much stock we put in the election itself15:40
ttxyeah15:40
ttxalright, so actions... I'll call for candidates (and explain the voting membership) on a thread soon15:40
ttxI'll document team scope on the wiki somewhere for reference15:41
Jokke_ttx: sounds good15:41
ttxand we'll wait for the PTL to be selected to formally propose the team for inclusion as an official project15:41
ttx(no hurry on that)15:41
dhellmannttx: should we put the team scope in the project team guide instead of wiki?15:42
ttxuntil then, the Release team is fine hosting the project if anything is needed15:42
ttxdhellmann: maybe yes15:42
ttxIf everybody is fine with it15:42
Daviey_ttx: How about you be interim leader until the election?15:42
*** Daviey_ is now known as Daviey15:43
ttxDaviey_: so that we speed up team adoption ? I don't think it needs to be sped up that much15:43
mriedemso there can be a lightning rod for keeping juno around for another year15:44
mriedem:)15:44
ttxand I think the new leader will signal that the team changed (rather than keep the same lead from last cycle and call it separate)15:44
Jokke_dhellmann: are you ok hosting stable under the wings of release until the paperwork is done? :P15:44
dhellmannJokke_ : sure15:44
Davieyttx: fair enough15:44
ttxDaviey: in practice, dhellmann is the lead until we spin out15:44
ttxAlright, I have all the points I wanted to cover15:45
ttxQuestions ? Thoughts, Comments ?15:45
Jokke_ttx: thanks for taking the lead on this15:46
dhellmannthanks again to mriedem and Jokke_ for stepping up as leaders, and to everyone else who is interested in participating in the new stable team15:46
Davieyttx: I think we could do with summarizing the outcome of this meeting.15:46
ttxHave a few things I think we'll need to work on soon -- like rewrapping stable-maint-core, evolve how to manage additions to $project-stable-maint, and replacing the horrible has-stable-maint tag by something about policy compliance15:46
dhellmannDaviey : ++15:47
ttxDaviey: I'll summarize it as part if the PTL nomination email15:47
ttxa.k.a. "here are the rules of the game, wanna play ?"15:47
Jokke_ttx: ehat do you mena by rewrapping stable-maint-core?15:47
Jokke_what even15:47
ttxJokke_: cleaning it up15:48
Jokke_ttx: oh, that's stale as well? :(15:48
ttxLike I said earlier, it's a failsafe group, doesn't need to keep inactive members around15:48
Jokke_ttx: indeed15:48
ttxOK, looks like we are done. Thanks everyone for gathering on such short notice15:49
Jokke_thanks15:49
bknudsonand on thanksgiving week no less.15:49
ttxI'll get that email done before tomorrow morning15:50
ttxbknudson: thanksgiwhat?15:50
ttxAlright, consider the meeting over, even if it was not started15:50
ttxAnd now I'll troll the Kuryr meeting15:50
jordanP:)15:50
Davieyttx: thanks!15:52
*** eharney has quit IRC15:53
*** jordanP has left #openstack-stable15:54
*** eharney has joined #openstack-stable16:08
*** rcernin has quit IRC16:08
*** armax has joined #openstack-stable16:09
*** Pablo|off| has joined #openstack-stable16:14
*** Pablo|off| is now known as pcaruana16:15
ttxJokke_, mriedem: sent the PTL self-nomination opening email at http://lists.openstack.org/pipermail/openstack-dev/2015-November/080249.html17:12
mriedemok, thanks17:12
*** rcernin has joined #openstack-stable17:21
Jokke_ttx: thnx, got it17:21
*** ihrachys has quit IRC17:24
*** derekh has quit IRC17:31
*** e0ne has quit IRC17:39
*** zul has quit IRC17:53
*** zul has joined #openstack-stable17:58
*** sghanekar__ has joined #openstack-stable18:05
*** armax has quit IRC18:10
*** pixelbeat has quit IRC18:17
*** pixelbeat has joined #openstack-stable18:17
*** e0ne has joined #openstack-stable18:22
*** e0ne has quit IRC18:29
*** e0ne has joined #openstack-stable18:29
*** mriedem has left #openstack-stable18:42
*** mriedem has joined #openstack-stable18:43
*** ihrachys has joined #openstack-stable18:45
mriedemanyone for a +W so i can continue backporting this? https://review.openstack.org/#/c/247825/18:47
*** pixelbeat has quit IRC18:48
*** ihrachys has quit IRC18:59
*** ihrachys has joined #openstack-stable18:59
*** ihrachys has quit IRC19:23
*** ihrachys has joined #openstack-stable19:29
*** ihrachys_ has joined #openstack-stable19:35
*** ihrachys has quit IRC19:35
*** pixelbeat has joined #openstack-stable19:38
*** ihrachys_ has quit IRC19:49
*** ihrachys has joined #openstack-stable19:49
*** e0ne has quit IRC19:55
*** dims has joined #openstack-stable19:56
*** ihrachys has quit IRC19:58
*** ihrachys has joined #openstack-stable20:01
*** ihrachys has quit IRC20:16
*** ihrachys has joined #openstack-stable20:21
*** armax has joined #openstack-stable20:22
*** pixelbeat has quit IRC20:28
*** e0ne has joined #openstack-stable20:42
*** pixelbeat has joined #openstack-stable20:46
*** pixelbeat has quit IRC20:56
*** e0ne has quit IRC21:08
*** e0ne has joined #openstack-stable21:17
*** david-lyle has quit IRC21:31
*** david-lyle has joined #openstack-stable21:34
*** dims_ has joined #openstack-stable21:43
*** e0ne has quit IRC21:43
*** dims has quit IRC21:44
*** david-lyle has quit IRC21:55
*** david-lyle has joined #openstack-stable22:03
*** david-lyle has quit IRC22:04
*** rcernin has quit IRC22:20
*** david-lyle has joined #openstack-stable22:21
*** david-lyle has quit IRC22:25
*** armax has quit IRC22:30
*** ihrachys has quit IRC22:30
*** ihrachys has joined #openstack-stable22:31
*** mriedem is now known as mriedem_away22:46
*** pixelbeat has joined #openstack-stable22:48
*** dims_ has quit IRC22:55
*** eharney has quit IRC23:02
*** david-lyle has joined #openstack-stable23:03
*** david-lyle has quit IRC23:05
*** doug-fish has quit IRC23:17
*** david-lyle has joined #openstack-stable23:23
*** dims_ has joined #openstack-stable23:25
*** david-lyle has quit IRC23:25
*** dims_ has quit IRC23:26
*** ihrachys has quit IRC23:29
*** david-lyle has joined #openstack-stable23:35
*** david-lyle has quit IRC23:41
*** david-ly_ has joined #openstack-stable23:41
*** armax has joined #openstack-stable23:47

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!