12:03:32 #startmeeting Heat 12:03:32 Meeting started Wed Sep 3 12:03:32 2014 UTC and is due to finish in 60 minutes. The chair is asalkeld. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:03:33 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 12:03:36 The meeting name has been set to 'heat' 12:03:53 right haven't done this for ages 12:04:04 shardy, o/ 12:04:09 https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282014-09-03_1200_UTC.29 12:04:16 Hey guys, sorry, bit late 12:04:33 I am just getting the agenda 12:04:34 you are just in time 12:04:37 log from last time: http://eavesdrop.openstack.org/meetings/heat/2014/heat.2014-08-27-20.00.txt 12:05:05 ryansb: thx :) 12:05:06 #topic review actions 12:05:11 i have topic to discuss for the review https://review.openstack.org/#/c/110557/, could some one bring this to today's agenda 12:05:21 Item to add is probably review of J3 status, unless it's already been added 12:05:25 were there any actions? 12:05:41 looks like the only actions were 12:05:51 1) stevebaker raise bugs for rstrip(None) issue from ask.openstack 12:06:09 and 2) review the mission statement https://review.openstack.org/#/c/116703/ 12:06:20 not sure steveb's around 12:06:38 everyone had a look at the mission? 12:06:41 right, this is the week it's at a ridiculous time for him 12:06:56 let's move on 12:07:06 #topic new agenda items 12:07:15 shardy, ... 12:07:47 let's do that now 12:07:48 asalkeld: Yeah we should probably review https://launchpad.net/heat/+milestone/juno-3 12:08:06 #topic review juno-3 12:08:16 all your shardy 12:08:19 your's 12:09:07 So the two I spotted which probably need a status update: 12:09:10 https://blueprints.launchpad.net/heat/+spec/cancel-update-stack 12:09:11 and 12:09:13 pas-ha, 12:09:27 looked quite risky 12:09:27 https://blueprints.launchpad.net/heat/+spec/troubleshooting-low-level-control 12:09:47 I looked at the reviews and neither looked like imminent merge candidates 12:09:47 shardy, yes, i think I just nailed it. will post new patch asap 12:10:06 cool 12:10:08 pas-ha: Ok, cool, if you think it's ready, lets try to get it reviewed and gating asap 12:10:19 asalkeld: I saw you had concerns re the patch update 12:10:27 thing we'll be deferring that one? 12:10:33 s/thing/think 12:10:37 I don't know 12:10:48 I just not a big fan of it 12:10:57 but if someone thinks it's ok 12:11:05 then i don't mind 12:11:46 do we really need a bp for the funct. tests? 12:11:59 not at all production effecting 12:12:17 asalkeld: I think it's just a way of tracking progress 12:12:29 so a BP is appropriate, but a spec is not 12:12:35 (IMO at least) 12:12:52 ok 12:13:33 pas-ha: is anything blocking https://blueprints.launchpad.net/heat/+spec/handle-update-for-security-groups other than reviews? 12:13:49 shardy, no 12:14:31 Same question re https://blueprints.launchpad.net/heat/+spec/implement-ec2eip-updatable but I'm not sure tiantian is around 12:15:12 Ok I guess that's all the BP's, are there any bugs there (or not targetted) which can't wait until RC1? 12:15:37 we have a couple of weeks after J3 to go into test/bugfix mode, so we will defer any bugs not fixed today until RC1 12:16:02 that's what rc's are for ? 12:16:03 So if you find a critical issue, please mark it as such and target it to J3, so we know it's a blocker 12:16:18 shardy, how is your bp? 12:16:26 asalkeld: Yeah, but if possible, we don't want to release J3 with any known critical issues 12:16:38 sure 12:16:48 asalkeld: Better after your help, but I'm still working on the rest of the series 12:16:51 is says implemented, but i know better;) 12:17:02 ok 12:17:04 asalkeld: It does? 12:17:19 yip 12:17:24 https://blueprints.launchpad.net/heat/+spec/decouple-nested 12:17:39 It's Beta Available, deferred to RC1, pending a FFE 12:17:55 ok, i was looking at nested 12:18:04 I'll try to get the patches in shape, then we can have a discussion about whether it makes sense to push for a FFE or just defer to Kilo 12:18:21 asalkeld: Yeah list-nested is done (other than a heatclient patch, which isn't part of J3) 12:18:44 Ok, anyone have anything else re J3 or the release process in general? 12:19:16 maybe a link, i should read up on it 12:19:21 ok then, thanks all :) 12:19:30 #topic Can we have a grenade testing update 12:19:38 i put this in 12:19:45 asalkeld: be PTL for a cycle, you'll soon get the hang of it :D 12:20:00 :-O 12:20:13 asalkeld: Ok, so I posted this patch $months ago: 12:20:23 shardy: right now this bp targets only the nested stack, how about other resources and app-config,etc ? 12:20:24 shardy, so you and EmilienM are working on that grenade pathc 12:20:28 https://review.openstack.org/#/c/86978/ 12:20:50 #link https://review.openstack.org/#/c/86978/ 12:20:52 KanagarajM: You mean the decoupling BP? 12:21:01 shardy: yes 12:21:23 KanagarajM, let's keep on topic 12:21:26 KanagarajM: well the config and deployment resources are already decoupled, because they're addressed via Rest calls to heat 12:21:29 asalkeld: +1 12:21:53 KanagarajM, you wanted to talk about: https://review.openstack.org/#/c/110557/ 12:21:58 yes 12:22:00 i'll change topic 12:22:03 The grenade thing, so we've been drip fed some reviews, and EmilienM has helped get it working 12:22:08 #topic https://review.openstack.org/#/c/110557/ 12:22:13 but basically, I/we need help getting it actually merged 12:22:18 shardy: hey 12:22:28 this patch is summited for the cloumn name change and its been suggested to support 'zero downtime migration' 12:22:45 I've got limited experience with grenade, and don't really know how to push it further, given that so far I've failed to get it working locally at all 12:23:12 right now no service support for 'zero down time migration' for this one. 12:23:42 thanks shardy 12:23:55 KanagarajM, we chatted about this in #heat 12:24:05 i said there: seems like a cosmetic change really 12:24:05 and potentially effecting upgrades 12:24:05 we *really* need to plan how we are going to tell users to upgrade to minimise downtime 12:24:05 I think we need a summit session on how to achieve this 12:24:27 yes 12:24:30 do others feel the same/different 12:24:45 +1 (definitely worth a summit session) 12:24:50 +1 12:24:51 +1 12:24:52 +1 12:24:54 asalkeld: I agree, and in the meantime, limiting disruptive non-functional migrations seems like a reasonable step 12:24:57 +1 12:25:06 ok, cool 12:25:19 ok, shall i summit for design session in K ? 12:25:25 #agreed limiting disruptive non-functional migrations seems like a reasonable step 12:25:32 i will come with initial approach i have identified 12:25:40 KanagarajM, you/me/someone else 12:25:51 we will get a session there 12:26:00 for upgrades in general 12:26:17 KanagarajM, we can move on? 12:26:22 ok. i will update my findings and share you all . 12:26:27 yes sure. THanks 12:26:38 any other topics? 12:26:48 #topic open discussion? 12:26:58 free for all 12:27:14 we need this patch in keystone client https://review.openstack.org/#/c/118383/ 12:27:20 remember to review the mission statement if you haven't already https://review.openstack.org/#/c/116703/ 12:27:31 BillArnold_: Is that a backwards-incompatible regression in the keystone API? 12:27:36 y 12:27:42 it sounded like it from the bug 12:27:44 shardy y 12:27:48 :( 12:28:05 Are they planning to fix the API, or just paper over it in the client? 12:28:36 shardy i don't know, was investigating it enough yesterday to get some tests working again 12:29:21 BillArnold_: Ok cool, I'm glad we've solved the heat issue, just concerned about the wider issue of the API change 12:29:43 shardy: right, this sounds like a serious botch 12:29:53 it looked from the docs like keystone were doing minor version deprecations without any corresponding bump to the actual API version 12:30:15 was looking for clarification as to whether that was actually the case 12:30:20 (sad versioning trombone plays) 12:30:55 anyway, probably a bit OT for here, I'll ask ayoung or jlennox later 12:31:04 https://review.openstack.org/#/c/118383/, keystone work with both region and region_id in juno for backward compatibility 12:31:24 sorry, flaky wifi 12:31:53 and i have patched it keystone for region and region_id. but its worth to make this change in the middleware as well in juno, though we have backward compatibility 12:33:41 all we all done, so fast? nothing else... 12:33:45 I would like to get comments regarding the approach for moving check calls to Observer https://review.openstack.org/118143 12:34:58 please give it a look and leave comment if anything looks incorrect 12:35:18 unmeshg, i think after the releases are done we will all be very keen to look 12:35:37 everyone is maybe a bit push for time atm 12:35:58 ok 12:36:21 shardy, anything else? 12:36:40 asalkeld_: not from me 12:36:50 #endmeeting 12:36:50 shardy: re https://review.openstack.org/#/c/116615/, I thought about our idea with handling error for different versions 12:37:01 ok I will continue in #heat ;) 12:37:15 i think my new nick has no powers:( 12:37:32 #endmeeting