15:00:42 <hberaud> #startmeeting oslo
15:00:43 <openstack> Meeting started Mon Oct 26 15:00:42 2020 UTC and is due to finish in 60 minutes.  The chair is hberaud. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:44 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:46 <openstack> The meeting name has been set to 'oslo'
15:01:04 <hberaud> Courtesy ping for bnemec, smcginnis, moguimar, johnsom, stephenfin, bcafarel, kgiusti, jungleboyj, sboyron, damani
15:01:14 <moguimar> o/
15:01:38 <bcafarel> o/ (sharing attention with neutron ptg session)
15:01:56 <hberaud> #link https://wiki.openstack.org/wiki/Meetings/Oslo#Agenda_for_Next_Meeting
15:02:38 <johnsom> o/
15:03:02 <kgiusti> o/
15:04:02 <hberaud> #topic Red flags for/from liaisons
15:04:24 <hberaud> nothing from our side
15:04:25 <johnsom> Nothing from Octavia
15:04:43 <sboyron> o/
15:05:15 <moguimar> nothing from Barbican
15:06:03 <hberaud> #topic Releases liaison
15:06:04 <bnemec> o/
15:06:45 <hberaud> I seen few releases proposed by damani
15:06:54 <hberaud> nothing more here
15:07:03 <hberaud> #topic Security liaison
15:07:41 <bnemec> Just the gerrit breach audit, which was completed without incident.
15:07:49 <hberaud> right
15:07:52 <sboyron> yep
15:08:00 <bnemec> Thanks to everyone who helped out with that!
15:08:05 <hberaud> +1
15:08:20 <hberaud> #topic TaCT SIG liaison
15:08:43 <sboyron> Nothing is on fire
15:08:54 <hberaud> awesome
15:09:04 <sboyron> or the fire for gerrit breach masked everything ^^
15:09:21 <sboyron> did not saw any mails
15:09:25 <hberaud> :)
15:09:38 <hberaud> #topic Action items from last meeting
15:10:29 <hberaud> just 1 thing here
15:10:31 <hberaud> hberaud to review https://review.opendev.org/#/c/742160
15:10:34 <hberaud> done
15:10:49 <hberaud> #topic Weekly Wayward Wallaby Review
15:11:40 <hberaud> https://review.opendev.org/#/c/698814/
15:12:15 <hberaud> this one is WIP since 2019
15:12:35 <hberaud> I wonder if we should abandon these changes
15:12:46 <hberaud> smcginnis: any opinion? ^^^
15:13:41 <smcginnis> Would be great to see that work done if anyone can take it on. It would be nice to have.
15:13:52 <hberaud> ack
15:13:56 <smcginnis> But if no one else can tackle it, we should probably just abandon it.
15:14:17 <hberaud> I'll try to take a look more deeply to this one
15:14:25 <hberaud> #topic Open discussion
15:14:49 <hberaud> anything else?
15:15:13 <hberaud> moguimar: you proposed to run our retrospective, maybe it's the right time
15:15:23 <hberaud> moguimar: as you want
15:15:26 <moguimar> sure
15:15:35 <moguimar> do you have the link for the ptg?
15:15:46 <moguimar> maybe we have some topics there already?
15:16:12 <hberaud> yep sec
15:16:26 <hberaud> #link https://etherpad.opendev.org/p/oslo-wallaby-topics
15:17:29 <hberaud> so few items are in the "what went well" part
15:17:59 <moguimar> ok
15:18:04 <moguimar> lets start with What went well
15:18:18 <moguimar> #topic Retrospective - What went well
15:18:24 <moguimar> no powers =P
15:18:36 <moguimar> zuulv3 migration
15:18:54 <moguimar> comments?
15:19:07 <bnemec> I included this because it got done on time and a number of different people contributed to it.
15:19:07 <hberaud> AFAIK this topic is well done
15:19:10 <moguimar> tosky, around?
15:19:23 <bnemec> Pretty much exactly how we want community goals to work.
15:20:02 <moguimar> cool
15:20:18 <bnemec> And yeah, big thanks to tosky for driving that!
15:20:23 <moguimar> what do you mean in time bnemec ?
15:20:40 <moguimar> like early in the cycle?
15:20:49 <hberaud> I suppose during the cycle
15:21:04 <bnemec> Yeah. We weren't desperately merging patches late in the cycle, or even after it was over.
15:21:07 <hberaud> this was a community goal
15:21:17 <bnemec> Which has happened on other community goals.
15:21:18 <moguimar> ok
15:21:23 <moguimar> moving on
15:21:25 <moguimar> hberaud took over most of the release duties
15:21:32 <moguimar> hberaud++
15:21:50 <bnemec> I guess this was kind of our first step toward the DPL model.
15:22:08 <moguimar> speaking of that, did we get a new oslo.utils release?
15:22:56 <hberaud> bnemec: yep I agreed
15:23:12 <hberaud> damani: ^^^ (I'm not sure)
15:23:35 <moguimar> yeah, I think damani started working on that one
15:23:55 <hberaud> I think too, but I didn't seen related release for now
15:24:30 <moguimar> any other things that went well?
15:24:50 <moguimar> I have one that is mixed feelings
15:25:02 <moguimar> the pre-commit patches
15:25:11 <hberaud> :)
15:25:18 <moguimar> It covered 35 repos I guess
15:25:22 <hberaud> yeah
15:25:36 <moguimar> we were able to kinda automate a good part of it
15:25:52 <moguimar> it was very good to first get good in a single repo and then replicating
15:26:05 <hberaud> yep
15:26:23 <moguimar> the part I'm not sure it was that good was how long it took to get it reviewed
15:26:32 <moguimar> not sure about it*
15:27:13 <moguimar> which is a sign that we are going short on cores
15:27:46 <hberaud> Agreed
15:28:29 <hberaud> fortunately new few people seems interested by oslo
15:28:44 <moguimar> I think we planted a good seed
15:28:51 <hberaud> +1
15:28:59 <moguimar> lets see how pre-commit will affect our work in the next cycles
15:29:19 <bnemec> I know last cycle I had toyed with the idea of changing our approval process to only require one +2 on a change submitted by a core.
15:29:43 <hberaud> it could be dangerous
15:29:51 <bnemec> I kind of tabled it because it seemed like we were keeping up on reviews okay.
15:30:25 <moguimar> In that case me and hberaud could have taken care of pre-commit alone
15:30:56 <moguimar> what do you think could be dangerous about it hberaud ?
15:30:56 <hberaud> yep
15:31:08 <hberaud> I'm not fan
15:31:30 <hberaud> pre-commit isn't a good example
15:31:32 <bnemec> FWIW, you wouldn't _have_ to single approve if the change was complex or anything like that.
15:31:47 <moguimar> I see
15:31:59 <bnemec> It would just leave it up to the core team's discretion whether to use that.
15:32:49 <bnemec> Our disagreement rate in Oslo is pretty low, so in practice I don't think there would be a huge change in quality as a result.
15:33:02 <bnemec> It's not all that common that a core +2's something and another comes along and -1's.
15:33:23 <hberaud> but by example on oslo.messaging I really prefer to see double check from an expert (kgiusti) rather merge patch too quickly to compensate for a lack of person
15:33:31 <bnemec> Anyway, it's something to think about. We don't necessarily have to make a decision immediately.
15:33:45 <hberaud> sure
15:33:49 <moguimar> ok
15:33:52 <moguimar> moving on
15:33:59 <moguimar> What didn't go well
15:34:04 <moguimar> focal ci migration
15:34:14 <bnemec> This was on me.
15:34:29 <moguimar> I think this is a good example of goal happening late during the cycle
15:34:31 <bnemec> I dropped the ball on this because I wasn't paying enough attention to the goal update emails.
15:34:39 <bnemec> Yeah.
15:34:49 <hberaud> :)
15:34:52 <bnemec> We shouldn't have been making major ci changes as late as we did.
15:35:18 <hberaud> agreed
15:35:38 <bnemec> I guess the silver lining is that post-ff there wasn't a lot that should have been merging anyway so it didn't matter if ci got broken for a while. ;-)
15:36:45 <bnemec> And I guess the takeaway from this is to make sure we have someone who owns community goals so someone is making sure they get done.
15:37:34 <hberaud> any idea about how to manage this with the DPL?
15:37:46 <bnemec> We saw the effect of that this cycle: the zuul migration got done because tosky took ownership of it, focal didn't because nobody had a specific interest in it.
15:38:08 <bnemec> I think we just need to make sure as a team that when community goals are assigned, someone takes responsibility for driving it.
15:38:19 <hberaud> ack
15:38:45 <bnemec> Which I'm pretty sure is the same thing I said last cycle when this happened, and then didn't follow through on it. :-/
15:39:17 <hberaud> I see
15:40:37 <bnemec> I will also note that since the privsep conversion goal was selected for Wallaby, we're likely to need a point of contact for other teams.
15:41:04 <hberaud> good point
15:43:15 <hberaud> Personally I'll try to be a little more attentive to community goal and the privsep stuff for this cycle
15:43:22 <bnemec> I guess I don't have anything else. That was all the topics I had for the retro. Any thoughts from the rest of the team on things that didn't go well?
15:43:44 <bnemec> hberaud: Cool, thanks! I'll try to help out with the privsep stuff as much as I can, but I can't make any promises.
15:44:22 <hberaud> no problem
15:44:23 <bnemec> Hopefully I've done enough work in privsep to answer some of the basic questions, at least.
15:44:25 <hberaud> thanks
15:44:50 <moguimar> That's all on my end too
15:45:11 <hberaud> anybody else is interested to become the bnemec's backup on this topic?
15:46:58 <hberaud> the last topic on our etherpad lack of present author
15:47:19 <hberaud> so I think we are near from the end
15:47:26 <hberaud> anything else?
15:47:27 <bnemec> That sounds like a spec to me.
15:47:41 <hberaud> yep
15:47:47 <bnemec> And we'll want kgiusti to look at it.
15:48:46 <hberaud> I left a comment
15:49:11 <hberaud> I hope he will seen it
15:49:25 <hberaud> so I think we are done
15:49:50 <hberaud> Thanks for joining!
15:49:56 <hberaud> #endmeeting