03:02:40 #startmeeting Large Deployments Team January 2016 Meeting 03:02:41 Meeting started Fri Jan 22 03:02:40 2016 UTC and is due to finish in 60 minutes. The chair is VW. Information about MeetBot at http://wiki.debian.org/MeetBot. 03:02:42 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 03:02:44 The meeting name has been set to 'large_deployments_team_january_2016_meeting' 03:02:53 is it just you and I, mdorman? 03:03:07 possibly. i told klindgren about it but not sure if he got the info 03:03:32 not sure about AU folks 03:04:03 oh snaps I be here 03:04:04 yeah, was hoping Sam was about 03:04:09 hee hee 03:04:42 yeah might just be us three then 03:04:53 #topic Neutron Work 03:04:54 I haven't seen sorinson on irc in a while 03:05:00 hmmm 03:05:40 will have to ask Tom how he'd doing 03:05:48 s/he'd/he's 03:06:10 anyway - how are things going with the Neutron stuff 03:06:51 we are getting close on the usages API extention, i think: https://review.openstack.org/#/c/212955/ dbingham has been working on that a lot. 03:07:03 but the spec got approved and everything, so i still expect we’ll get it into mitaka 03:07:13 iirc someone from rackspace was working on it as well 03:07:38 It's very possible 03:07:45 * VW can't keep up with all the things now 03:07:52 yeah he’s been coordinating with a couple people over there 03:08:34 Brian Stajkowski 03:08:41 ah - yes - Brian 03:08:48 he manages our Neutron team 03:09:07 I'll make a note to catch up with him tomorrow on it 03:10:18 Carl has been working on the SPEC for the routed/segmented networks. There has been a ton of back and forth on that thing. But he messaged me the other day saying that he thinks that its at a pretty solid position now 03:10:33 That's good 03:10:46 I was chatting with some technical leaders from our private cloud side 03:10:48 i reread the current version of that last week and i feel like it’s in a better spot now too. i think our needs/asks are accurately captured 03:11:08 that's a big piece to get Neturon and cells working in a way that is much more upstream :) 03:11:09 though it will still be a long time before anything really useful sees the light of day, i think. 03:11:22 cool 03:12:37 Cool any other Neutron/Network stuff to touch on? 03:13:02 not that i know of 03:13:35 #topic mid-cycle 03:13:49 so, niether of you are making it to the mid-cycle, correct? 03:13:57 right. 03:14:39 yeah, it's still not completely decided for me, but there are some things brewing 03:14:58 they have an LDT slot in the meetup 03:15:04 just not sure whos going to be running it 03:15:05 ? 03:15:24 sounds like belmiro is fairly involved in the planning 03:15:57 I'll have an engineer there. Belmiro will be there 03:15:58 yeah 03:16:00 hee ehe 03:16:12 I think I'll end up going 03:16:18 cool 03:16:40 I thought it was sold out already 03:16:57 hmm 03:17:04 that poses a challenge 03:17:07 will follow up 03:17:27 I assume emails to Tom might smooth things out a bit :-) 03:17:38 let's hope 03:17:51 I've got Rackspace and Intel folks pushing for a fleet managment project and some of the inventory code public by Austin 03:18:06 good deal 03:18:11 unfortunately with two sets of leaders, it takes a little longer to make a decsion 03:18:17 otherwise, I'd be booked already 03:18:18 :) 03:18:36 But everyone wants to do some talking about it at the mid-cycle 03:18:50 hopefully get more traction than the CMDB stuff back in August 03:19:06 i would imagine it would since this is more than just vapor 03:19:32 :) 03:20:02 yeah, my hope is that pieces of inventory (and maybe provisioning) out by Austin with some folks (like your selves) jumping in to work 03:20:15 and then shoot for Resolver/automation services out by Barcelona 03:21:04 There is a big bug smash we are doing with Intel in early march, so I'm working on making some of that inlcude prepping code for public use 03:21:06 we'll see 03:21:13 yeah, something to shoot for at least 03:22:07 any burning topics you guys want to make sure are covered at the mid-cycle? 03:22:26 Side question - are their details of how the intel-rackspace lab is setup? Cells? number of computes per cells? ect ect 03:22:54 There probably are - I'll have to get them. I know they were shooting for 1000 node clusters 03:22:58 I don't believe they have cells 03:23:16 yeah i would not really make sense to deploy that with cells v1 at this point. 03:23:16 I'm hoping in the next couple of weeks, we'll start trying to stand up the fleet management stack with them 03:23:28 cool 03:23:36 to date, they have mainly been run by the Private Cloud folks, so sadly, I don't have a lot of direct knowledge on them 03:23:37 yet... 03:23:38 :) 03:24:06 So the only other thing that I can think about was how the cells patches stuff is going... 03:24:22 I'm behind on that 03:24:27 it seemed like it was on track - then got shot down due to lack of gate testing for cells v1? 03:24:28 is there stuff in motion since the summit 03:24:33 sight 03:24:35 sigh even 03:24:54 then matt R added some gate testing jobs for cells 03:25:07 but it seemed like that got voted down by other nova members 03:25:19 not much. the few that we reviewed with mriedem more or less got rejected by nova 03:25:27 yeah, what kris said 03:25:30 blargh 03:25:53 cuz you know -multi cell v2 is "right around the corner" 03:26:00 i think that even if we were to do the effort to add the tests, it still wouldn’t have much support b/c there is no interest in doing anything on cells v1. 03:26:14 ok - at one point, I had asked Garbutt to try and get to the Ops meetup. Not sure if he did, but if I get over there, I'll follow up with him 03:26:30 and there is potential for other changes not related to v1 could cause those tests to fail, and they don’t want to have to support making those pass. 03:26:33 yeah, but cells v2 is dragging 03:26:40 right. 03:27:06 yeah, i mean i understand both sides. i kind of think a separate repo where we could maintain the patches separately might be a more effective route and a better use of time. 03:27:06 good old fashtioned crap sandwich 03:27:09 haha 03:27:36 i had kind of worked on pushing on some of that stuff during the weeks right after the summit, but since december haven’t really had bandwidth to do much with it. 03:28:46 yeah 03:29:17 may reach out to you for an offline summary mdorman - if I'm able to sit down and visit with John 03:29:44 so, i donno. i wouldn’t be opposed to championing/working on a separte repo thing. but would wnat to get some more input from other cells deployers to see if that would actually be valuable. 03:29:47 sure, that’s cool 03:31:15 #topic Other Business 03:31:20 anything else pressing? 03:32:19 i was goign to ask you if you had any more info in this fleet management sprint idea 03:32:22 Nothing from me - I htink like everyone else since the summit - I have been pretty much slammed 03:32:35 totally acceptable if it’s all still TBD 03:34:14 well, I jsut found out about the bug squash stuff in early march mdorman 03:34:26 so even if we don't actually pull folks together in one spot 03:34:37 then I think we have an oporrtunity to rally folks virtually 03:34:42 but let me dig 03:34:57 folks are on board for some focused pre-Austin work though 03:35:13 just trying to get two leadership teams to agree on things as a priority, etc 03:35:14 :D 03:35:38 alright, yeah that’s cool. just wondering if we should still think about pulling together travel, but sounds like at this point likely not 03:36:03 we just got a new manager, so need to get stuff like this on a new radar 03:36:07 Give me two weeks and I'll know for sure 03:36:18 sorry - it's hurry up and wait 03:36:20 :\ 03:36:54 yeah that’s np at all 03:37:04 nothing else for me 03:37:23 me either 03:37:38 thank you guys for hopping on 03:37:49 I'll try to reach out to Tom and Belmiro in the next couple of days 03:38:11 cool. just reach out if there’s anything else we an do in the meantime 03:38:22 will doo 03:38:25 do even 03:38:51 see you guys in Austin if not before 03:38:58 * gfa late, reading the logs 03:39:10 * VW pauses 03:39:30 we still have time gfa - if you have anything 03:39:55 no, i'm fine 03:39:57 tnx 03:40:11 no problem 03:40:15 all is welll with you? 03:40:52 kinda, i'm not doing openstack for the moment :-/ 03:41:01 well that's no fun 03:41:04 but wanted to follow close what others are doing 03:41:16 and neutron 03:41:18 but we all understand day jobs gettin gin the way 03:42:58 alright, well unless anyone objects, I'll call the meeting good then 03:43:22 np 03:43:40 alrighty then 03:43:44 #endmeeting