12:00:39 #startmeeting heat 12:00:40 Meeting started Wed Aug 6 12:00:39 2014 UTC and is due to finish in 60 minutes. The chair is zaneb. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:00:41 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 12:00:43 The meeting name has been set to 'heat' 12:01:21 tspatzier: interested in chairing today? 12:01:31 zaneb: sure 12:01:41 zaneb, getting weary of the chair 12:01:47 #chair tspatzier 12:01:47 Current chairs: tspatzier zaneb 12:01:55 hi 12:01:58 hi 12:02:08 zaneb: ok, let's see how I do ;) 12:02:09 o/ 12:02:15 asalkeld: it's 8am, hard enough to just function normally ;) 12:02:15 tspatzier: Error: Can't start another meeting, one is in progress. Use #endmeeting first. 12:02:21 hey :) 12:02:27 o/ 12:02:28 10pm here 12:02:29 o/ 12:02:39 tspatzier: I already started the meeting, but I set you as a chair 12:02:50 ok 12:02:58 #topic rollcall 12:03:10 o/ 12:03:12 hi 12:03:15 again?)) 12:03:50 sorry guys, doing this the first time, so bare with me ;) 12:04:20 I'll keep my clothes on, if you don't mind 12:04:28 #topic Review last week's actions 12:04:34 ugh, I didn't create the agenda for this week 12:04:45 and there were no actions 12:04:55 tspatzier: norm, do not worry :) 12:05:08 there were no actions :) 12:05:23 so moving on 12:05:31 #topic adding items to the agend 12:05:34 #topic adding items to the agenda 12:05:43 hmm, we should have had an action for stevebaker to update the gap closing wiki page 12:05:55 btw, speaking of actions = what happened to Rax CI/CD job? is it on? 12:06:04 according to the log there was none 12:06:05 #action zaneb check with stevebaker about progress on gap coverage plan 12:06:50 ok, regarding the agenda: https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282014-07-30_2000_UTC.29 12:06:54 anything to add? 12:07:08 #link https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282014-07-30_2000_UTC.29 12:07:15 that's last week's 12:07:22 damn 12:07:38 https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282014-08-06_1200_UTC.29 12:07:42 #link: https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282014-08-06_1200_UTC.29 12:07:42 ftfy 12:07:54 I just added it like 1 minute ago 12:07:55 how do we keep up with the convergence meetup? 12:08:03 * zaneb is not very organised this week 12:08:49 asalkeld: you mean the heat meetup, not the tripleo meetup right? 12:08:55 yip 12:09:04 * asalkeld feeling left out 12:09:19 might be hard given the tz difference 12:10:16 ok, let's put that on the agenda 12:10:25 doing it right now... 12:10:33 it's going to be a short meeting anyway, I suspect :) 12:11:27 #topic Keeping up to date on the Heat mid-cycle meet-up 12:11:43 so this is the only topic before critical issues synch 12:11:57 given the large review list, is there a way for us to priortise reviews? 12:12:12 (so I can focus my reviewing ...) 12:12:15 btw, small announcement: if you are going to the meetup, check my mailing list post and make sure you are signed up 12:12:40 otherwise security will bounce you 12:13:22 zaneb: checked + signed in. Is starting at 9am still the plan? 12:13:27 yep 12:13:55 ryansb: in your case, bring your employee badge an you'll have no trouble ;) 12:13:58 zaneb: sound like you have iron man as security :) 12:14:22 zaneb: Yup. It lives in my suitcase so I won't forget it. 12:14:35 ryansb: smart :) 12:14:53 * ryansb has a problem with forgetting identification when going places 12:15:13 asalkeld: so etherpads would be one way for folks to keep up 12:15:28 but as you said, with the time zones it's going to be tricky 12:15:33 zaneb, yip was thinking of that 12:15:45 what is planned? 12:15:48 what parts are you interested in following? 12:15:55 maybe do a bluejeans sync-up either early morning or late evening to accomadate tz's? 12:16:00 in general 12:16:10 but maybe the cont. observer 12:16:19 ok 12:16:27 I'll be on vacation, so would be interested to read things afterwards. etherpads? 12:16:48 I think the idea is to actually start hacking, more than to just talk about stuff again 12:17:17 so maybe daily pushes of working branches to e.g. GitHub might help as well 12:17:32 review is better 12:17:36 wip 12:17:44 so reviews for everyone to stay up to date 12:18:29 #info try to push WIP reviews during meetup so non-attendees can follow along at home 12:18:50 zaneb +1 that would be very helpful 12:18:58 +1 12:18:59 may be make sense to have one common tag for these reviews? it will allow search easier 12:19:22 also a "how to run", so we can give it a go and help out 12:19:30 skraynev: that's a great idea 12:19:31 24hr hacking 12:20:08 asalkeld; sounds awesome :) 12:20:37 asalkeld: and a little crazy ;) 12:20:49 just don't want a massive code dump on the reviews that I know nothing about 12:21:13 it's better to get involved early 12:21:18 #info use branch name "juno-midcycle" to ensure easy filtering of reviews 12:21:28 zaneb: thx 12:21:45 great, sounds like we have a plan. anything else on this topic? 12:22:25 so moving on ... 12:22:33 #topic Critical issues sync 12:23:00 is there anything critical at the moment? 12:23:58 does not seem so 12:24:06 #topic open discussion 12:24:19 does anyone have anything? 12:24:29 I wouldn't mind a little follow-up to last week's beating over ASG health 12:25:00 I have a plan B to discuss 12:26:21 khkhkjhkj 12:26:31 ok then.. 12:26:41 Plan B is in two parts... 12:27:07 one is to get LB pool member health info into Ceilometer samples; that is not Heat work 12:27:19 the other part is a stop-gap while waiting for automatic convergence... 12:27:51 What do you think of this idea: let a scaling group do a check for undermined members and remove them, as part of a resize 12:28:16 undermined = the underlying thing has been deleted, leaving a record in the Resource table whose nova_instance refers to no existing thing 12:28:25 ? 12:28:41 mspreitz: we know we need something to allow the application to choose members to delete on scale-down, so this may tie in with that 12:29:19 zaneb: I think these can be mods to distinct parts of the code... 12:29:42 my mod affects the list of existing members handed to the method that produces the new template; Bill's part modifies that method. 12:29:43 yeah, I saw your definition of 'undermined' after I wrote that 12:30:30 so the 'stack check' reviews are I think close to landing 12:30:45 although, maybe, there is one little interaction.. 12:30:54 it would make sense to me if the autoscaling code used something like stack check to check for member health during scaling operations 12:31:11 the scale down plugin will be given a list of members already decided to remove; maybe that should include ones detected as undermined. 12:31:33 OK, I'll look at those 12:33:12 #link https://review.openstack.org/#/q/status:open+project:openstack/heat+branch:master+topic:bp/stack-check,n,z 12:33:39 thanks for the link 12:33:53 np 12:35:01 I am done 12:35:06 with that topic 12:35:10 mspreitz: I think you and BillArnold had some specs. Are they current? 12:35:23 I mean specs on that topic ... 12:35:31 I think the spec about autoscaling across AZs is current 12:35:40 I will let Bill speak to his 12:36:00 What's the current status of the config change for qpid? Did that go out yesterday? 12:36:05 Yes, scaledown plugpoing is current 12:36:12 My network is flaky right now, can not do much with openstack.org 12:36:54 ok, so that might be worth looking at. 12:37:07 #link https://review.openstack.org/#/c/110379 12:37:09 ryansb: is this upstream, or in RDO? 12:38:15 I think RDO, but I'm not sure, lemme dig up the bug. 12:39:19 tspatzier also stack lifecycle plugpoint spec - https://review.openstack.org/#/c/106468/ 12:39:32 ryansb: this is probably the wrong forum to discuss then 12:39:59 zaneb: right, it is. 12:40:02 Sorry 12:40:04 BillArnold: that one actually looks pretty complete to me. 12:40:14 ryansb: np, we can discuss after 12:40:23 #link https://review.openstack.org/#/c/106468/ 12:40:58 ^^if some cores could look at this if this is good to go? 12:41:14 anything else for this meeting? 12:41:57 no - then thanks everyone! 12:42:03 zaneb, what's up with the plugin stuff? should I dump that 12:42:17 tspatzier, you can end it 12:42:21 ok 12:42:24 we can chat in #heat 12:42:24 #endmeeting