18:00:05 #startmeeting keystone 18:00:05 Meeting started Tue Jan 23 18:00:05 2018 UTC and is due to finish in 60 minutes. The chair is lbragstad. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:07 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:08 ping ayoung, breton, cmurphy, dstanek, edmondsw, gagehugo, henrynash, hrybacki, knikolla, lamt, lbragstad, lwanderley, kmalloc, rderose, rodrigods, samueldmq, spilla, aselius, dpar, jdennis, ruan_he 18:00:09 o/ 18:00:10 The meeting name has been set to 'keystone' 18:00:11 o/ 18:00:12 #link https://etherpad.openstack.org/p/keystone-weekly-meeting 18:00:17 agenda ^ 18:00:17 o/ 18:00:19 let's do this! 18:00:34 good energy :) 18:00:49 shakin' things up 18:01:07 o/ 18:01:34 o/ 18:02:05 we'll give folks another minute to join 18:02:31 o/ 18:02:35 o/ (food took too long to arrive, will be monitoring the meeting from phone) 18:03:05 #topic announcements 18:03:17 let's go through the status of things 18:03:28 #topic announcement: release status 18:03:36 we're getting down to the wirte 18:03:41 wire* 18:03:50 feature freeze and client freeze are both friday 18:04:13 with that, we did make a bunch of progress on reviews last week 18:04:18 i think client freeze is thursday 18:04:20 as noted in cmurphy's weekly recap 18:04:31 cmurphy: oh - good call 18:04:52 on that note - we should also be planning to have things *gating* on thursday if we want to land them this releas 18:04:54 release* 18:05:01 Do we need to request any FFEs? 18:05:54 given the RC period is only two weeks, i'm not sure we'll have much time 18:06:01 * hrybacki nods 18:06:05 requirements freeze is also friday http://lists.openstack.org/pipermail/openstack-dev/2018-January/126475.html 18:06:12 but i expect we can assess on wednesday evening 18:06:29 o/ i am unlurking now 18:06:54 we have three major efforts left 18:07:08 cmurphy: has been whipping up the application credential stuff 18:07:32 which looks like it's waiting on a bug gix 18:07:33 fix* 18:07:49 cmurphy: anything you want to share there? 18:08:22 the ansible folks brought up https://bugs.launchpad.net/keystone/+bug/1744948 with me so i'd like to get that fixed before merging anything else 18:08:23 Launchpad bug 1744948 in OpenStack Identity (keystone) "allow_application_credential_creation contraint issue with suse + mariadb 10.2" [High,In progress] - Assigned to Colleen Murphy (krinkle) 18:08:42 ack - i'll be reviewing that right after this meeting 18:08:46 waiting for zuul results on that patch and they are testing it for me 18:09:09 but everything else is looking good 18:09:15 awesome 18:09:22 nice 18:10:11 the unified limits implementation is looking good, but we're working on pulling out a FK constraint that merged with the migration 18:10:39 so that's gating on https://review.openstack.org/#/c/536644/ 18:10:54 dstanek is also looking into https://review.openstack.org/#/c/126030/25 18:11:09 to make sure we captured all context of the problem before moving foward 18:11:12 forward* 18:11:34 the system scope stuff is just pending reviews 18:11:49 there is a weird issue with tempest i started digging into with andreaf last night 18:12:09 and that is happening when we add a system admin role to the admin user in bootstrap... 18:12:24 not sure what the story is there yet, but that's one of the last patches in the series 18:13:12 we're going to have a fast transition to rc period 18:13:35 i still need to parse all keystone bugs for any show stoppers so that we can focus on those during that two week period 18:14:15 if you interested in helping build out that list, let me know.. i want to try and have that done by monday at the latest so we have a list of things to shoot for for RC 18:14:30 lbragstad: I can probably help you there 18:14:35 * cmurphy too 18:14:39 bug stomp? 18:14:49 hrybacki: cmurphy that'd be a massive help, i appreciate it 18:14:50 Whooza 18:15:35 feel free to tag things as queens-rc if it is something we need to fix before queens goes out the door 18:15:54 #topic announcements: priority reviews 18:16:21 for your reviewing convenience - we now have a review board dedicated to queens feature work we need to land this week 18:16:27 #link https://goo.gl/NWdAH7 18:16:50 if there is a review missing from there, let me know and i'll get it added (or removed) 18:17:12 any questions on release related activity? 18:18:33 Wow that list looks long for me 18:18:49 yeah - 46 keystone patches and 4 client patches 18:18:58 Priorities? 18:19:13 d.) everything 18:19:14 :) 18:19:27 gates are gonna be in pain this week 18:19:28 the good thing is that most of them are in good shape 18:19:30 Cool, simple 18:19:36 yes - that too 18:19:52 I will try to do a few reviews this week 18:20:02 i've been trying to baby sit them in off hours so we can maximize our reviews, but nothing beats being as early as possible 18:20:04 As much as I can on that list 18:20:23 hrybacki: that wont help :( 18:20:40 samueldmq: I know -- but it happens every cycle :| 18:20:53 good strategy lbragstad 18:21:18 lbragstad: agreed, lets try to kill a few of then today and Tomorrow still 18:21:23 office hours == review session? 18:21:34 i'm good with that 18:21:59 if people want to break up and review things in batches we can do that too 18:22:25 lbragstad: thats a good strategy 18:22:34 2 - 3 people per feature? 18:22:40 +1 18:22:44 we can figure out that in office hours, too 18:22:49 Make sure to distribute the cores wsll in the groups 18:22:54 +1 18:23:23 alright - i'm going to hand it off to hrybacki 18:23:26 #topic RH1 (CI) is coming down 18:23:29 ack 18:23:43 ++ 18:23:46 general awareness, the RH1 cloud (running TripleO CI OVB jobs) is coming down 18:23:51 ahead of schedule actually 18:23:59 https://bugs.launchpad.net/tripleo/+bug/1744763 18:24:00 Launchpad bug 1744763 in tripleo "Ci: moving OVB jobs to RDO third party CI" [High,In progress] - Assigned to Sagi (Sergey) Shnaidman (sshnaidm) 18:24:08 :( 18:24:19 just general awareness -- the TripleO CI team is hammering away on this 18:24:51 How representative is that in our current resource pool? 18:25:22 samueldmq: that's hard to say. I don't think it will affect us too much. I don't think we have a billion gate jobs running OVB rn 18:25:33 So that we could have an Idea of How much impact in current Ci capacity 18:25:57 I'm honestly not sure -- been out of the upstream CI game for a cycle and things have changed a bit 18:26:07 what is an ovb job? 18:26:19 OpenStack Virtual Baremetal 18:26:38 basically OpenStack on TripleO 18:26:43 or quintupleo 18:27:09 RH is pushing resources into a new cloud, RDO Cloud, and TripleO CI is expanding CI there 18:27:24 is this something that affects keystone or is this related to infra resources? 18:27:54 more related to infra resources -- I don't think it will hurt this team but I'm not 100% on that either 18:28:49 * hrybacki tries to find list of repos affected 18:30:09 you can see the list in the 'same topic' section of this review: https://review.openstack.org/#/c/536507/2 18:31:04 so mostly TripleO and THT related (didn't mean to eat up this much time in the mtg) 18:31:45 that's all I have on that lbragstad 18:31:53 no worries - thanks for the heads up 18:32:04 #topic core reviewers 18:32:23 for those who don't know yet, gagehugo became core last week :) 18:32:32 congrats gagehugo :) 18:32:38 :) 18:32:43 i'm not sure who put this topic on the agenda thoguh 18:33:06 wasn't me 18:33:17 me -- just curious if we need to plan a way to gain more cores 18:33:17 hehe 18:33:29 Congrats gagehugo 18:33:30 I know we talked about it at the last PTG -- and we've lost some since 18:33:36 Well deserved 18:33:57 samueldmq :) thanks 18:34:36 i think a plan would be good - but i'm not sure how to approach building one 18:34:39 i don't think we're ever going to have enough cores 18:34:47 since it usually depends on the reviewer 18:34:54 cmurphy: ++ 18:35:04 make your first patch, get a free beer in dublin 18:35:19 flight included? XD 18:35:43 I guess the question is -- are nearing the point of /needing/ more core reviewers? 18:36:06 i think review pressure has gone up - but i'm curious to hear what others think 18:36:09 we'll have a better answer after the progress today during office hours i think 18:36:23 fair point knikolla 18:37:13 okay, well let's put a pin in this for now 18:37:27 ok 18:37:44 #topic open discussion 18:38:13 I feel like the pressure isn't just on keystone, a lot of projects seem to be hurting for reviewers 18:38:24 yeah it's openstack-wide 18:38:31 agreed 18:38:53 :( 18:39:12 the volume of work or need for it hasn't changed, though 18:39:31 at least in my opinion 18:39:39 or based on the people i talk to 18:40:09 i think it all comes down to how easy it is to have part time contributors 18:40:38 it's harder today to get paid to work a large chunk of your time upstream. 18:40:53 ^^ +2 18:40:55 knikolla ++ 18:41:13 yeah - that's going to be a big part of how we manage things going forward 18:42:22 that was a big topic in sydney during the forum 18:44:49 before we get into office hours, do we want to discuss breaking into review teams here? 18:44:59 +1 18:45:15 ++ 18:45:19 fwiw I have the next 105 minutes free 18:45:32 awesome 18:45:53 i already planned on reviewing the application credential stuff, so we can get that unblocked 18:46:07 and go through the entire stack again 18:46:27 last week the app cred stuff worked for me 18:46:35 if someone wants to join me, i'd appreciate the extra set of eyes 18:46:49 I haven't had a chance to look at that stuff, I'd be happy to join 18:46:55 cool 18:47:01 i've already done a fair chunk of system scope reviewing, so i'll keep on that. 18:47:09 knikolla: perfect 18:47:23 unified limits def needs more eyes 18:47:26 ++ 18:47:37 I can look at whatever, I should be available all afternoon 18:47:47 as soon as i'm done with the app creds stuff, i'll transition to unified limts 18:47:51 limits* 18:48:05 * hrybacki fetches coffee before we begin 18:48:05 (getting the app cred stuff gating would be good since it's close) 18:48:40 gagehugo: do you want to pick through the unified limits stuff first? 18:48:43 * knikolla wishes the hazelnut coffee k-cups would arrive faster 18:48:52 lbragstad yeah I can take a look at that 18:49:03 gagehugo: thanks 18:49:26 there is all the scope_types stuff, too (which cmurphy has already reviewed most of) 18:49:47 I think I looked at them last week, will take another look 18:49:51 (probably the lowest priority of the efforts though) 18:49:54 ok 18:50:19 that should be easy reviews for the most part. and will block the gate for an entire day if we approve them all. 18:50:39 we might be able to send a few through in batches 18:50:50 * lbragstad isn't a CI wizard though 18:51:04 or push them through later in the day 18:51:04 * knikolla shrugs 18:51:16 gagehugo: yeah - that might work, too 18:51:19 things definitely run faster at 2am :) 18:51:56 good thinking. we can +2, and delay the +A for late night hours 18:52:11 so long as two other cores +2 i' 18:52:24 i'll +A at night if no one is around and the queue is clear 18:52:39 (as long as people are OK with the process there) 18:52:44 no one should be afraid to +A things, if it waits it waits but it's not going to knock over infra 18:54:02 that works too, i'm going to be watching patches closely for the next few days anyway 18:54:11 cmurphy: you're right. I just feel like those are lower priority, so causing something higher priority which might need a respin to wait might be suboptimal. but whatever works. 18:54:27 back 18:56:04 #action lbragstad and hrybacki to review application credentials during office hours 18:56:13 #action knikolla to review system scope during office hours 18:56:29 #action gagehugo and lbragstad to review unified limits during office hours 18:57:56 anything else people want to go through for reviews? 18:58:19 that's a pretty solid start :) 18:58:26 i'm happy with that plan 18:58:43 give folks a minute to stretch their legs before getting started 18:58:46 thanks all! 18:58:49 #endmeeting