22:00:12 #startmeeting nova_cells 22:00:13 Meeting started Wed Jan 7 22:00:12 2015 UTC and is due to finish in 60 minutes. The chair is alaski. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:00:15 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 22:00:18 The meeting name has been set to 'nova_cells' 22:00:34 Anybody here for the cells meeting? 22:00:38 \o aloha 22:00:39 hi 22:01:03 hi 22:01:08 o/ 22:01:24 cool, let's get started 22:01:44 I will admit up front that I'm a bit disorganized after the break 22:01:50 still catching up on a lot of stuff 22:01:53 eh 22:02:01 join the club :P 22:02:04 :) 22:02:05 :) 22:02:09 #topic testing 22:02:39 so it appears that we're up to about 70 failures on the cells job now 22:03:00 it was at around 40 before the break 22:03:06 ergh 22:03:13 I'm not sure what tests are failing now yet 22:03:35 it's possible that we'll just want to skip them 22:03:42 * edleafe arriving late 22:04:02 but this does make me want to get this job voting asap 22:04:24 #link https://etherpad.openstack.org/p/nova-cells-testing 22:04:32 so as always, any help here is appreciated 22:04:36 can we maybe provide a non-voting job first ? 22:04:52 because atm, it's only experimental pipeline right ? 22:04:52 bauzas: we have one, but it is in experimental 22:04:57 right 22:05:15 moving that to the check queue might be a good step 22:05:18 alaski: yeah, that's why I thought we should maybe put a non-voting in the check pipeline 22:05:42 alaski: in there, just skip the failing ones 22:05:46 I like that 22:05:59 alaski: and leave the full testing in experimental 22:06:14 alaski: so at least we would make sure we wouldn't regress 22:06:25 yeah 22:06:47 bauzas: would you like to look into that? 22:06:55 alaski: sure 22:07:07 cool 22:07:26 alaski: just lemme find how to get 25 hours for a day but I'm fine eh ;) 22:07:29 #action bauzas look into getting a cells tempest job into check pipeline 22:07:40 bauzas: heh. I'm happy to help with that too 22:07:58 alaski: lol, I guess we all are kinda busy 22:08:10 anything else on testing? 22:08:11 that's life, go ahead 22:08:37 #topic table analysis 22:08:57 I was supposed to put up a review to get the uncontroversial tables into the devref 22:09:08 I have not 22:09:10 lol 22:09:31 but I will propose that tomorrow 22:10:07 well... is not really the tables that are controversial 22:10:36 there are questions like the scheduler, aggregates, nova-network that we need to discuss 22:11:03 agreed 22:11:36 maybe having a meeting to discuss these points would be a good start 22:11:44 I think it will be helpful to have some things that are more concrete so we have a common understanding 22:11:59 but we definitely need to discuss those larger topics soon 22:12:00 belmoreira: +1 22:12:57 should we plan to devote these meetings to a particular topic each week for a bit? 22:13:36 we could discuss that during midcycle but I'm not sure that everyone can attend 22:13:40 alaski: I think is a good idea 22:14:01 bauzas: I will not be in the midcycle 22:14:15 belmoreira: ack 22:14:32 belmoreira: okay. we'll pick a topic to focus on each week 22:14:44 belmoreira: we did something last midcycle about a remote participation for the sched, maybe we could do this too ? 22:15:22 bauzas: that would be great 22:16:05 Everyone okay with starting with scheduling next week? 22:16:15 +1 22:16:19 alaski: +1 22:16:36 +1 22:16:47 cool 22:17:16 #action alaski clear out next weeks agenda to discuss scheduling 22:17:33 which leads nicely into the next topic 22:17:40 #topic cells scheduling 22:17:59 mainly I want to point out https://review.openstack.org/#/c/141486/ 22:18:37 I am going to update it as soon as I can, and we can use that to work towards next weeks discussion 22:18:53 cool 22:19:03 great 22:19:30 #topic open discussion 22:19:50 Anyone have anything to discuss? 22:20:33 alaski: now that the first BP has been merged, do you plan to provide the change also ? 22:20:39 I mean soon 22:21:05 and by soon, I mean k2 22:21:08 I would like to work on those soon, but I haven't written any code yet 22:21:20 alaski: ok 22:21:40 and I'm happy to share if others are interested in getting involved 22:21:41 alaski: also, I won't explain to a core that you will need to ask for an exception for all the missing specs :) 22:22:14 bauzas: :) 22:22:28 I'm not sure that I want an exception for them honestly 22:22:40 I don't want to rush things here 22:22:48 but I also don't want work to get held up 22:22:59 alaski: yeah makes sense 22:23:04 so I'm considering looking at a feature branch to use for prototyping 22:23:16 but I don't know what all is involved with that 22:23:31 alaski: warning, big question : do you plan a stackforge fork ? 22:23:55 I am not planning a fork 22:24:22 alaski: oh by branching you mean a Gerrit change 22:24:23 the second BP is also approved 22:24:26 series 22:24:57 bauzas: that would be one way, but I think we can actually have a separate branch to merge to 22:24:57 by branching, I was thinking about git branch... 22:25:05 bauzas: right 22:25:10 alaski: orly? 22:25:26 bauzas: I don't know any details, but I've heard it discussed before 22:25:34 I want to get some details before proposing anything 22:25:34 * bauzas needs to think about the bribe amount 22:26:22 well, I was stupidely thinking that branches were duties for the Project Manager 22:26:42 hence the stackforge iterations 22:26:52 this only matters though if we finish the two approved blueprints with plenty of time left in the cycle 22:26:56 hence Gantt and all the other tiny ugly projects 22:27:04 right 22:27:32 k, sounds a good approach then, to be discussed later then 22:27:58 cool. let's get some code merged first :) 22:28:02 alaski: about getting involved in the implementation I'm interested. Or at least I can point someone. 22:28:18 belmoreira: excellent 22:28:19 alaski: let's discuss later 22:28:31 belmoreira: sounds good 22:29:03 anything else for open discussion? 22:29:42 crickets 22:29:44 closing up early then 22:29:49 Thanks everyone! 22:29:57 #endmeeting