14:00:13 #startmeeting nova_scheduler 14:00:15 #link Agenda for this meeting https://wiki.openstack.org/wiki/Meetings/NovaScheduler#Agenda_for_next_meeting 14:00:20 Meeting started Mon Oct 30 14:00:13 2017 UTC and is due to finish in 60 minutes. The chair is edleafe. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:21 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:21 o/ 14:00:23 The meeting name has been set to 'nova_scheduler' 14:00:25 \o 14:00:25 \o 14:00:36 o/ 14:01:03 Who's around this UGT morning? 14:01:40 * efried googles UGT and is unsatisfied. 14:02:09 o/ 14:02:20 efried: http://www.total-knowledge.com/~ilya/mips/ugt.html 14:02:46 efried: the last line is especially relevant 14:03:12 Let's get started 14:03:13 #topic Specs 14:03:13 Still only one spec remaining spec 14:03:13 #link Add spec for symmetric GET and PUT of allocations https://review.openstack.org/#/c/508164/ 14:03:16 Just needs some lovin' (and some reviews) 14:03:27 Anything else for specs? 14:03:43 * alex_xu lurk at here just for avoiding task assigning 14:03:54 alex_xu: smart man! 14:04:22 #topic Reviews 14:04:28 #link Nested RP series starting with: https://review.openstack.org/#/c/377138/ 14:04:34 jaypipes: any comments? 14:05:09 edleafe: in process of rebasing. should be up in about 15 minutes 14:05:21 edleafe: we've all been competing for microversions :) 14:05:26 jaypipes: cool 14:05:39 I won the first battle. 14:05:41 yeah, I have $50 riding on the outcome 14:05:44 But I shall not win the war. 14:05:45 :) 14:05:56 * cdent is going to lose 14:06:07 * cdent has three horses in this race 14:06:17 Though in the grand scheme of things, not sure "making everyone rebase" counts as a win. 14:06:43 efried: it ain't a loss 14:06:54 oh man, DST bye bye 14:07:05 * bauzas waves now 14:07:14 cdent: you'll probably win the next race... :) 14:07:35 bauzas For even more fun, American DST ends next weekend. 14:07:50 yeah I know 14:07:59 US != EU timeshifts 14:08:02 anyway 14:08:07 DST is stupid. 14:08:08 I'll lose 25 hours next weekend 14:08:13 should be abolished. 14:08:31 jaypipes: now the next patches for the nrp series are about the placement DB, right? 14:08:37 I mean the API DB 14:08:39 efried: I don't think anyone who has coded datetimes would argue 14:09:11 efried: edleafe: well, my laptop gave me the right time today, but I didn't looked at the notification :p 14:09:16 so, PEBKAC :p 14:09:32 bauzas: yep 14:09:40 jaypipes: k, will look 14:09:46 bauzas: though it's the REST API one that continues to get the microversion shaft... :) 14:10:02 jaypipes: gotta be quicker! 14:10:06 edleafe: :) 14:10:17 ok, next up: 14:10:18 #link Add traits to GET /allocation_candidates series starting with https://review.openstack.org/479776/ 14:10:25 bauzas: https://review.openstack.org/#/c/377138/ is the one that is ready for review (but I need to rebase so wait up for 15 mins) 14:10:30 efried: alex_xu: comments? 14:11:08 jaypipes: k 14:11:17 The whole series passed the all the existed functional test, the next around is about refine the patch, add more testcases 14:11:39 I'm wondering whether we should do granular at the same time, rather than trying to stage it. But if it's close, then may as well. 14:12:12 efried: remember, our motto is merge shit early! 14:12:42 next up: 14:12:44 #link POST /allocations to set allocations for >1 consumers - series starting with https://review.openstack.org/#/c/512420/ 14:12:54 Looks like that series is moving along nicely 14:12:57 cdent: comments? 14:13:39 nothing to add really. I earlier this morning fixed up a comment that made what version was being talked about a bit unclear 14:13:47 it’s got the right (for the next few minutes at least) microversion on it 14:13:55 I hope it is good to go 14:15:11 ok, everyone who has bet that cdent wins the next microversion race, go review that series! 14:15:45 Last review on the agenda: 14:15:47 #link Alternate hosts: series starting with https://review.openstack.org/#/c/499239/ 14:15:52 I'm in the process of splitting the RPC flag for the change into two separate flags to indicate 1) do we use the new selection objects and 2) if the first is true, do we want alternates returned? 14:15:58 This is for some actions that don't retry at the cell level. 14:16:10 I should have something posted a little later today 14:16:13 edleafe Can we add the 'granular' series(es) to the list please: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/granular-resource-requests 14:16:44 #link Granular resource requests https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/granular-resource-requests 14:17:03 efried: anything to discuss? 14:17:07 I'm coming at this from both sides: one subseries from the flavor end - starting at https://review.openstack.org/#/c/515151/ 14:17:24 and the other starting on the placement side, starting at https://review.openstack.org/#/c/514091/ 14:18:06 For the former, the bottom change has had some attention. Would love to get some followup and see if I've landed stuff the way jaypipes and gibi like it. 14:18:25 For the latter, the bottom change could use some eyeballs on it. 14:19:01 I want to note that there's a separate series that was working on parsing traits from flavors. 14:19:21 https://review.openstack.org/#/c/492026/ 14:19:35 efried: hope that the result of attacking from both ends doesn't end up like https://tctechcrunch2011.files.wordpress.com/2007/11/train_20tracks.jpg 14:19:59 hah! I hope so too. 14:20:22 Anyway, I believe mine is a superset, and could conceivably supersede the above. 14:20:49 efried: ok, be sure to update the agenda next week with your progress 14:20:52 And from the placement side, I'm going to be somewhat lockstepping with alex_xu 14:20:57 edleafe Roger wilco. 14:21:04 Any other reviews to discuss? 14:21:35 edleafe: yeah, one sec 14:21:42 next week ? 14:21:49 isn't the Summit, folks ? :p 14:21:49 edleafe: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/de-orm-resource-providers 14:22:03 bauzas: you are correct s/week/meeting 14:22:22 that reminds me, I'll be out starting on Wed 14:22:37 and I fsckly need to write 20 slides 14:22:40 so... 14:22:46 not a huge week for me, I guess 14:22:51 jaypipes: anything to note about those reviews? 14:23:05 jaypipes: does that close out the de-orm stuff? 14:23:14 mriedem: it does indeed. 14:23:18 edleafe: nope 14:23:52 ok, then let's move on 14:23:53 #topic Bugs 14:23:53 #link Placement bugs https://bugs.launchpad.net/nova/+bugs?field.tag=placement 14:24:14 Two new bugs this week 14:24:23 Both are being worked on 14:24:42 Anyone have anything else to discuss about bugs? 14:25:11 gibi: is lajos still working on https://review.openstack.org/#/c/497399/? 14:25:45 (Haven't seen gibi around today) 14:25:53 k 14:26:25 #topic Open discussion 14:26:25 Nothing on the agenda 14:26:34 1) as bauzas pointed out, no meeting next week because of the Sydney Forum 14:26:40 2) I'll be out the following week, so if anyone wants to run the meeting, it's all yours! 14:26:54 efried and I will be lonely. 14:27:02 :( 14:27:07 jaypipes we'll have each other 14:27:11 jaypipes: That's how I felt during Boston 14:27:57 Anything else to discuss? 14:28:07 not from me. 14:28:14 jaypipes and efried should be able to gain some advantage in the microversion races 14:28:14 I'm done, need to write slides 14:28:34 cdent: heh :) 14:28:37 cdent Well, maybe I will. Nobody will be around to approve jaypipes's patches. 14:28:51 good point :) 14:29:07 OK, thanks everyone! 14:29:09 #endmeeting