Thursday, 2020-02-27

*** tosky has quit IRC00:21
*** jamesmcarthur has joined #openstack-tc00:35
*** jamesmcarthur has quit IRC00:51
*** jamesmcarthur has joined #openstack-tc00:52
*** jamesmcarthur has quit IRC00:57
zanebspotz: my bad if that was misinterpreted. the point was that in order to ensure we have volunteers to fill the slots the bylaws oblige us to have, we will have to accept that the people being voluntold are not going to actually have time to do any work01:16
zanebI think that was clearer from my first message, but by the time of the second the context had been diluted :/01:17
*** gagehugo has joined #openstack-tc01:19
*** jamesmcarthur has joined #openstack-tc01:22
*** jamesmcarthur has quit IRC01:23
*** jamesmcarthur has joined #openstack-tc01:23
*** jamesmcarthur has quit IRC01:55
*** jamesmcarthur has joined #openstack-tc02:00
openstackgerritZane Bitter proposed openstack/ideas master: Add Project Teapot idea  https://review.opendev.org/71017302:24
*** jamesmcarthur has quit IRC02:31
*** jamesmcarthur has joined #openstack-tc02:43
*** jamesmcarthur has quit IRC02:49
gmanntc-members: we need to close the 2nd Ussuri goal (contribution guide) implementation ASAP we are already late on this. This is one proposal for new updates -https://review.opendev.org/#/c/709617/02:50
gmannthis is ML discussion- http://lists.openstack.org/pipermail/openstack-discuss/2020-February/012784.html02:51
gmannproactive voting will be appreciated either - or +.02:51
*** jamesmcarthur has joined #openstack-tc02:53
*** diablo_rojo has quit IRC02:57
*** jamesmcarthur has quit IRC03:07
*** jamesmcarthur has joined #openstack-tc03:31
*** jamesmcarthur has quit IRC03:58
*** gagehugo has quit IRC05:03
*** gagehugo has joined #openstack-tc05:03
*** evrardjp has quit IRC05:34
*** evrardjp has joined #openstack-tc05:35
*** e0ne has joined #openstack-tc07:16
*** e0ne has quit IRC07:23
*** e0ne has joined #openstack-tc07:24
*** e0ne has quit IRC07:29
*** e0ne has joined #openstack-tc07:48
*** e0ne has quit IRC07:52
*** slaweq has joined #openstack-tc07:53
*** e0ne has joined #openstack-tc08:06
*** e0ne has quit IRC08:08
*** tosky has joined #openstack-tc08:17
*** witek has joined #openstack-tc08:20
*** rpittau|afk is now known as rpittau09:09
*** dmellado has quit IRC09:59
ricolinjungleboyj, can you update https://review.opendev.org/#/c/698582 when you got time for it? :)10:07
*** rpittau is now known as rpittau|bbl11:12
*** iurygregory has joined #openstack-tc11:59
*** gagehugo has quit IRC12:19
*** gagehugo has joined #openstack-tc12:20
TheJuliaI have not had a chance to chime in, but it seems inappropriate to be changing goal completion criteria and work scope this late into the cycle.12:26
*** iurygregory has quit IRC12:32
*** jamesmcarthur has joined #openstack-tc12:36
*** iurygregory has joined #openstack-tc12:36
smcginnisI agree with TheJulia on that. I think the contributor docs goal should just move to victoria. It came in *way* too late in the cycle in the first place, and it's still not clear exactly what the implementation is.12:37
smcginnisSure, it may be a noop for a lot of teams (maybe), but it's too late.12:38
TheJulia++12:40
TheJuliaWe have to keep in mind team are resource constrained as well. It is going to be easy to miss as well12:40
TheJuliaTeams are,yay for autocorrect12:40
*** jamesmcarthur has quit IRC13:00
*** jamesmcarthur has joined #openstack-tc13:00
*** dmellado has joined #openstack-tc13:05
*** jamesmcarthur has quit IRC13:06
*** jamesmcarthur has joined #openstack-tc13:10
*** rpittau|bbl is now known as rpittau13:24
*** jamesmcarthur has quit IRC13:32
*** jamesmcarthur has joined #openstack-tc13:32
*** e0ne has joined #openstack-tc13:35
*** lpetrut has joined #openstack-tc13:36
*** ijolliffe has joined #openstack-tc13:43
*** iurygregory has quit IRC13:46
*** e0ne has quit IRC13:46
*** jamesmcarthur has quit IRC13:47
*** jamesmcarthur_ has joined #openstack-tc13:47
*** e0ne has joined #openstack-tc13:57
gmannI agree with the concern of time and resource constraint for completing goal. we will be more streamlined from V or surly from W cycle to have goal ready at that starting of cycle.13:59
gmanncontributor guide goal should be easy to do but i can agree about time left issue.14:01
* fungi just got an image of a surly wallaby in his head14:03
gmannoh s/surly/surely :)14:05
spotzzaneb: It almost seems like we should makee the change to 3 members for the UC as we had discussed and then talk merger or at least have that effort with thee other one14:07
*** e0ne_ has joined #openstack-tc14:08
*** e0ne has quit IRC14:08
zanebspotz: that's a bylaws change though :(14:08
zanebbecause the number is hardcoded14:08
spotzzaneb: yeah no matteere what iit seems likee by-laws would need changing14:09
gmannsmcginnis: TheJulia i will bring this in today office hour if we can get consensus on moving the goal or doing it in Ussuri only. because still there are still lack of votes on goal change doc proposal so i cannot say when it will be ready.14:11
spotzAnd I have told melsakhawy and jayahn about the discussions here. Jaesuk wasn't feeling well so they're not having their meeting this morning14:11
*** melsakhawy has joined #openstack-tc14:15
fungii don't think it's a foregone conclusion that a bylaws change is required, at least not for the option #1 ttx proposed on the ml14:15
fungior the similar option i proposed in here last week14:15
fungiofficially renaming committees or changing the number of seats declared for the uc would almost certainly require a bylaws change, but that will take years to complete, so it might be nice to have other options which can be employed in the meantime14:16
zanebI think a bylaws change could be done in 1 year, but yeah14:17
gmannbut option#1 from ttx is not just for Bylaw change time ? it is a try for permanent or long term solution if work well14:18
gmannif it is till we change Bylaw then looks good option to go.14:19
melsakhawywe aren't aiming for the merger before August anyway14:20
melsakhawywe'r still discussing within the UC, we should have a meeting including TC and UC sometime in March to discuss plans to proceed14:21
*** melsakhawy has quit IRC14:25
*** melsakhawy has joined #openstack-tc14:25
fungizaneb: the earliest we could reasonably have a vote of the membership and be able to expect enough turn-out for it to be a binding result is coincident with the board elections next january, but the back-and-forth with legal counsel for all the companies represented by the board can easily take longer than that and would result in pushing any vote out to 202214:27
fungiso yes it *could* take just under a year to complete the bylaws change, but i wouldn't place any bets on that14:28
fungigmann: ttx's option #1 could easily fill in the gap until any bylaws change is enacted to remove the uc from the bylaws (if that's the consensus direction the community and board wind choosing)14:31
njohnstonI am concerned that this is going to suck a lot of attention over a long period of time and distract from more productive focuses.14:34
*** jamesmcarthur_ has quit IRC14:35
njohnstonI think that if a bylaws change is needed, if ttx's option #1 is not a valid option, we should really assess what the downsides are of doing nothing at all.14:36
njohnstonI'm not sure they are that huge.  Even if the UC ends up with zero members - a scenario the bylaws does not seem to address - and the TC informally starts handling it's responsibilities, that could work.14:38
njohnstonI just see a danger in committing to a process that will be a significant distraction while not really delivering the kinds of changes our constituent communities are most concerned about.14:40
*** jamesmcarthur has joined #openstack-tc14:40
ttxnjohnston: if option 1 is not satisfactory, I think we should still pursue encouraging operators / users to join the TC14:44
ttx(while keeping the UC around)14:45
ttxI personally still think it's a level of complexity and load balancing that is no longer necessary, but yes we are constrained by the bylaws as to what is easily doable and what's not14:46
*** jamesmcarthur has quit IRC14:48
ttxRenaming things is costly from a bylaws perspective, but also from a documentation update perspective, governance URLs that include '/tc' to redirect etc. It's just too much work for limited gain.14:49
fungiyeah, we have countless groups in our community who do amazing work without their existence being demanded in the bylaws. if this were the "users and operators sig" (or something like that) without pressure from the bylaws for it to justify its continued existence, it could dissolve and re-form any time it needed14:50
fungiand could scale up or down, organize however was convenient for its members, et cetera14:51
fungiand renaming the one governing body besides the board which does have responsibilities delegated to it would help of course... but the more of the bylaws we propose changing the greater the chance that it will take 2 (maybe more) years to get done14:54
ttxoffice-hour!15:00
ricolino/15:01
gmanno/15:01
ttxWould be good to get to the bottom of what to do with that contributor guide goal... so that diablo_rojo knows where to apply champion energy15:01
gmanntc-members one topic i would like to bring here about ussuir 2nd goal of contribution guide.15:01
gmannttx: yeah15:01
jungleboyjo/15:02
njohnstono/15:02
jungleboyjHappy Thursday.15:02
ttxIt's in the grey area with most people not caring immensiely one way or another. and a few caring deeply15:02
ttxwhich makes it hard to motivate people to choose between them15:02
TheJuliajungleboyj: It is thursday?!? :)15:03
ttxTheJulia: don't get me started15:03
jungleboyjTheJulia: It most definitely is.15:03
ttxTheJulia: hope you enjoyed Geneva15:03
ttxI'll be heading in mountain direction starting tomorrow, for one week15:03
TheJuliattx: very much so, heading back to Paris now :)15:03
ttxI intend to stay off IRC15:03
TheJuliathat sounds excellent15:04
gmannone more issue and concern is about time left for ussuri release and projects are ok to accept the goal with whatever direction we decide.15:04
*** e0ne_ has quit IRC15:04
fungii think the compromise with critical contributor information in the CONTRIBUTING.rst and then less critical information in doc/source/contributors/contributing.rst or whatever is a reasonable middle-ground. it would be nice if a sphinx include could avoid the need to specify the critical things in more than one place in the repo so they don't get out of sync, but that's not a show-stopper15:05
gmannbut before doing time check, we as TC need to get consensus on implementation way15:05
ttxI think everyone wants the text in the file to be short *but* informative. The current one is definitely too light for my taste.15:05
jungleboyjfungi:  That sounds reasonable.15:05
TheJuliagmann: my $0.02 is that the additional goal activity scoped should be scoped for the next cycle, since initial expectations were already communicated to projects. A change after the initial communication is just not great. :(15:05
ttxAlso fine with pushing it back to Victoria, which is a short cycle anyway15:06
ttxso having a bit of a head start on it can't hurt15:06
gmannTheJulia: yeah, agree on that.15:06
TheJuliattx: ++15:06
fungii feel like we run into this every cycle though. we try to specify detail for goals, then we get to implementing them and discover something needs changing. i don't really know how to fix the process. the things we've tried so far haven't changed that dynamic that i can see15:07
gmannif we push it on Victoria then we end up having single goal for Ussuri. I hope everyone is fine with that. It should ok i think15:07
gmannfungi: with new goal schedule it should be avoided at some extend. when we will process the goal discussion one cycle ahead15:07
fungior we could consider the process not broken, and just accept that all cycle goals are going to get pushed back on and redefined and completed the following cycle15:08
fungigmann: well, it's still usually at implementation time that we discover implementation details which need changing15:08
gmannand I will say we can have a deadline for not-accepting-any-change in goal after this time period.15:08
gmanns/this time period/ defined time period15:09
fungibut yeah, maybe if implementations start at (or before?) the beginning of the target cycle then the problems will get worked through in time15:09
gmannyeah.15:09
ttxAlso in doubt, I'm encouraging everyone to lean on the goal champion to decide, rather than defend their pet variant15:09
jungleboyjfungi: ++15:09
fungihowever most teams would rather wait until late in the cycle to consider goals work15:09
ttxbecause discouraging champions is not a great tactic15:09
fungior that's how it's seemed so far15:09
TheJuliaor perhaps that cycle goals should be planned/communicated further out? We're not a young community anymore. We need to be mindful of contributor bandwidth and ability to even track a moving target on a short time schedule. The longer/further out the easier for people to plan/allocate time15:10
*** melsakhawy has quit IRC15:10
gmanntrue, i am also waiting for what diablo_rojo_phon  think on that and how feasible is that to complete in Ussuri15:10
ricolingmann, I do okay with either allow it to complete within two cycles or if we enable pre-select process. Either way, we are more prepared for contribution guide to start before we ask each team to change15:10
fungiso discovering implementation problems in goals happens too late in the cycle (because that's when teams started to try to work on getting them done), and the result is "oh the goal is wrong, now we can't do it this cycle"15:11
TheJuliattx: Perhaps an important question, does the theme and or the explicit population of specific content matter?15:11
TheJulias/and//15:11
fungithough if we officially say a goal spans two cycles and there are no milestones which have to be completed in the first cycle, human nature says everyone will procrastinate and do it in the second cycle15:11
gmanni can see lack of review on goal implementation is one of the issue for that. having more review from more contributors than just TC can shape us in better way for future15:12
TheJuliafungi: I am not sure the suggestion is to say it spans two cycles, but more so plan a cycle ahead15:12
ricolinand that remind us got to start pre-select process correctly this time15:13
gmanntwo cycle would not help much i think. but keep goal ready before start of cycle and no change accepted except worst situation after start of cycle.15:13
TheJuliafungi: so there is more an ordered list execution, and if the goal is cemented earlier then maybe someone is able to get to it early even, there is of course risk with that where older goals ended up with teams jumping ahead and the plans changed on them15:13
gmannricolin: +1, we can do from W as planned.15:13
ricolingmann, ++15:14
TheJuliaI guess my issue is the change of plan/expectation, and if the TC feels that it is okay... then *shrug*15:14
fungiTheJulia: yeah, i believe we've recently enacted changes to the schedule to start planning goals a cycle ahead, though i'm still not sure that solves the fact that problems tend to go undiscovered until the goal is being implemented rather than during planning15:15
gmannfor V cycle we have one goal already ready and we can say if we do not get second goal before V start then live with single goal. that is what we have to do at some cycle to start the pre-select process15:15
fungi(i mean, problems are discovered during planning too, just not all of them from what we've seen in the past)15:16
TheJuliafungi: there can always be a test project or test execution to lead the way15:16
TheJuliathe delta just needs to be reconciled though. We've had a bad history of not reconciling that delta due to resource constraints15:16
gmannfungi: for that we can make the step of 'implemented in one of the project' as mandatory instead of preferred.15:16
fungiyeah, i think some of our better goals have started out that way... seed projects implement poc, goal is proposed, solution is copied form seed projects to the rest of them15:17
fungiand those typically don't start out as goals15:17
fungiproject x, y and z are doing this thing, it's working well for them, let's make it a goal to implement something like that for everyone15:18
gmann+115:18
njohnston+115:18
gmanni will note down those improvement ideas and propose for review.15:19
ricolinFor the selected goals for V, maybe we can use the time we have before V to start review the plan. I don't see we need to do much for zuul v3, but it will be a good practice for TCs.15:19
gmannyeah.15:19
fungii have a feeling there is a lot of subsurface legacy zuul v2 compat shim stuff hidden in the jobs many projects copied into their repositories, fwiw15:20
gmannzuulv3 has enough poc and with single implementation way15:20
zanebthe grenade thing worries me with that one15:21
gmannback to contributor guide goal. let's wait to hear from diablo_rojo_phon what she think on postponing or doing in ussuri.  or if she would like to check on ML15:21
TheJuliafungi: I also suspect there is reasoning behind some things that were done previously that may have been ripped out, but still apply elsewhere. It will always be a case by case thing for every project15:21
ricolingmann, +115:21
gmannzaneb: do not worry :) it is almost done15:21
fungii regularly find in-repo jobs which are still parented to "legacy" jobs from the openstack-zuul-jobs repo or invoking zuul-cloner or calling devstack-gate to set something up15:21
zanebgmann: excellent :)15:21
gmannzaneb: i am hoping to get it merge by next week. it is in good shape - https://review.opendev.org/#/c/548936/10715:21
gmannfungi: yeah, there are lot.15:22
zanebre the contrib guide goal: I am committed to helping as much as possible to make it happen in U if that's the consensus. but...15:23
ricolingmann, For Zuul v3. I'm not worry about the implementation, but the docs itself. `How team can use it`, `What you might encounter`, `Where's the reference` etc. IMO It will be nice if we can review on those before V. :)15:24
zanebit seems to me that beneath the bikeshedding about which file it should go in (which is at least 50% my fault), the root cause is that we don't actually have a clear idea who the audience is and what we're trying to achieve15:24
*** e0ne has joined #openstack-tc15:24
zanebwhich, if true, is a more serious problem15:25
gmannricolin: sure, we have really good doc on zuulv3 migration and how one can define job. but yes we can double check for those thing in advance15:25
TheJuliaI suspect it is more there is an idea of audience and what is trying to be achieved, but the means to reach that are often clouded because context always differs15:26
gmannricolin: and you are one of best candidate to re-check those as you did migrated heat jobs into zuulv3 recently and very fast.15:26
ricolingmann, willing to15:27
fungito rewind the contributor docs goal, i believe the impetus was that every project eventually be represented in the openstack contributor guide, and that if they have any workflows or policies which are "special" to their team and not reflected in the general guide, that they explain them in their contributor documentation. there's a secondary need to not remove the CONTRIBUTING.rst entry point which15:29
fungimany folks who get a copy of the source repo in some way will rely on to figure out basic things like how to reach the developers, where to report defects, and how to propose patches15:29
fungithe project-specific contributor guide sections, in turn, are somewhat in service of getting a bunch of team documentation relocated out of the wiki and into project docs15:30
fungito further reduce our reliance on wiki.openstack.org15:30
* gmann going for doctor appointment. will check logs later.15:30
TheJuliaperhaps that should be an explicit goal... I mean... I've talked about doing it for ironic numerous times... but I almost never get a chance to cycle to cleaning up wiki.openstack.org15:32
zanebthat actually sounds great, but the conversations we're actually having have barely touched on any of that15:33
fungiyep, i think we usually get to focused on the implementation and forget to talk about the reasons15:34
fungier, too focused15:34
jungleboyjTheJulia:  I have been meaning to clean up the Cinder Wiki for several cycles.  :-)15:35
ricolinfungi, usually the reasons of goal show up once in ML/Forum and never hear it again:)15:36
*** jamesmcarthur has joined #openstack-tc15:38
TheJuliaThe abstraction away to implementation makes it easy to be lost... tbh.15:43
fungistuff like moving off wiki.o.o has been an implied eventual goal for so long that we forget to mention it when we're working toward it, i think15:43
jungleboyjSorry, was multi-tasking, pulled into another meeting.15:44
jungleboyjfungi:  I think your interpretations is good.15:44
*** diablo_rojo has joined #openstack-tc16:00
*** e0ne has quit IRC16:12
* mordred agrees with fungi16:16
*** lpetrut has quit IRC16:40
ricolinevrardjp, FYI, I just added two topic suggestion for next meeting if that's not too late16:40
*** jamesmcarthur has quit IRC16:45
*** jamesmcarthur has joined #openstack-tc16:46
*** e0ne has joined #openstack-tc16:53
*** njohnston is now known as neutron-fwaas17:21
*** neutron-fwaas is now known as njohnston17:22
*** rpittau is now known as rpittau|afk17:23
*** evrardjp has quit IRC17:35
*** evrardjp has joined #openstack-tc17:35
*** witek has quit IRC17:49
*** e0ne has quit IRC18:21
*** ccamel has joined #openstack-tc18:27
diablo_rojogmann, finally caught up with goal stuff. I think if we get the actual format settled in the next weekish(ideally ASAP). We still have time to finish the goal before the end of the release.18:29
diablo_rojoI will reply on the thread today.18:29
*** camelCaser has quit IRC18:30
*** jamesmcarthur has quit IRC18:33
*** jamesmcarthur has joined #openstack-tc18:33
diablo_rojozaneb, left a comment on your patch https://review.opendev.org/#/c/709617/2 if I am wrong I will +1.18:33
*** witek has joined #openstack-tc18:40
*** e0ne has joined #openstack-tc18:57
*** e0ne has quit IRC18:59
zanebdiablo_rojo: you are correct19:01
zanebI'll submit a typo fix on top19:02
diablo_rojozaneb, glad I'm not 100% insane :)19:02
zanebdiablo_rojo: I should have checked more carefully after I also screwed it up on the first attempt https://review.opendev.org/#/c/709617/1..2/goals/selected/ussuri/project-ptl-and-contrib-docs.rst19:04
diablo_rojoLol. Links are hard and not made any easier by all the files being named versions of the word 'contributing'19:05
openstackgerritZane Bitter proposed openstack/governance master: Fix transposed links  https://review.opendev.org/71033219:05
zanebthird time lucky ^19:06
zanebmy eyes glaze over every time I hit %7b%7bcookiecutter.repo_name%7d%7d19:07
diablo_rojoIts quite the mind bending repo thats for sure. You've seen my patchset history touching that thing.19:07
zaneblol19:08
*** e0ne has joined #openstack-tc19:10
*** jamesmcarthur has quit IRC19:12
-openstackstatus- NOTICE: Memory pressure on zuul.opendev.org is causing connection timeouts resulting in POST_FAILURE and RETRY_LIMIT results for some jobs since around 06:00 UTC today; we will be restarting the scheduler shortly to relieve the problem, and will follow up with another notice once running changes are reenqueued.19:12
*** ijolliffe has quit IRC19:31
*** ijolliffe has joined #openstack-tc19:32
*** jamesmcarthur has joined #openstack-tc19:40
-openstackstatus- NOTICE: The scheduler for zuul.opendev.org has been restarted; any changes which were in queues at the time of the restart have been reenqueued automatically, but any changes whose jobs failed with a RETRY_LIMIT, POST_FAILURE or NODE_FAILURE build result in the past 14 hours should be manually rechecked for fresh results19:46
*** ijolliffe has quit IRC19:53
*** ijolliffe has joined #openstack-tc19:57
*** e0ne has quit IRC20:07
*** jamesmcarthur has quit IRC20:54
*** witek has quit IRC20:58
*** slaweq has quit IRC21:15
gmanndiablo_rojo: thanks. let's see if that is ok for projects. IMO too, it should be doable in Ussuri.21:47
diablo_rojogmann, I updated the ML with progress on the most recent update thread21:47
gmannyeah, concern this morning was about timelines as we already passed the m-2 and goal is changing now so moving to V or can be done in U.21:50
diablo_rojoYeah I get all that.21:50
diablo_rojoI think if it was a code change I would vote to push it21:51
diablo_rojoSince its docs that are relatively stand alone I dont think its that hard to accomplish.21:51
gmannI agree.21:52
diablo_rojoWe still have over a month till m3 right? I think thats plenty of time to get things done.21:52
gmannthis is not tied to project release as not code change so doing till final release also fine.21:53
diablo_rojoExactly.21:53
diablo_rojoSo I think it should be fine.21:53
*** jamesmcarthur has joined #openstack-tc21:54
gmann+1.22:02
*** dhellmann_ has joined #openstack-tc22:05
*** dhellmann has quit IRC22:05
*** dhellmann_ is now known as dhellmann22:05
*** jamesmcarthur has quit IRC22:08
*** ijolliffe has quit IRC22:09
*** slaweq has joined #openstack-tc22:10
*** jamesmcarthur has joined #openstack-tc22:11
*** slaweq has quit IRC22:15
fungii think the only thing it risks breaking is docs jobs anyway22:17
*** slaweq has joined #openstack-tc22:17
*** slaweq has quit IRC22:22
*** jamesmcarthur has quit IRC22:49
*** slaweq has joined #openstack-tc23:10
*** slaweq has quit IRC23:16
*** jamesmcarthur has joined #openstack-tc23:25

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!