22:02:06 #startmeeting horizon 22:02:07 Meeting started Tue May 14 22:02:06 2013 UTC. The chair is gabrielhurley. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:02:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 22:02:10 The meeting name has been set to 'horizon' 22:02:10 Hi folks! 22:02:18 hey 22:02:25 hey gabriel 22:02:39 #topic overview 22:02:57 Hello 22:03:04 We've got two weeks left to get blueprints into H1, so this is crunch time. The more we can do this week the better. 22:03:20 Thank you to everyone who's been reviewing lately, it's been a big help 22:03:26 hi 22:04:03 Basically, anything that's not up for code review by this time next week is probably going to get deferred to H2, just FYI. 22:04:34 Overall things are looking decent, though, so I still have confidence in what we can do in H1 22:04:38 :-) 22:04:51 let's dive into specifics of the blueprints 22:04:55 #topic blueprints 22:05:10 The two "essential" BPs are committed, so that's great. 22:05:20 Keystone BPs are the next big chunk 22:05:34 I noticed that https://review.openstack.org/#/c/21942/ is currently abandoned 22:05:43 lcheng, david-lyle: care to update us? 22:05:53 gyee is going to pick it up 22:06:05 gotcha 22:06:06 any ETA? 22:06:27 starting early this week 22:06:42 k 22:06:46 is that actually blocking work on the domain, group, etc. CRUD or just a nice-to-have for those? 22:06:54 seems like it's only strictly a blocker for the login w/ domain 22:07:00 not blocking 22:07:09 k 22:07:10 just blocking anything with a v3 token 22:07:23 so are you still feeling confident on the various keystone v2 blueprints for h1? 22:07:37 so we can work against the v3 API using a v2 token 22:07:41 yeah 22:08:03 gabrielhurley: +1 that's all accurate 22:08:08 yep 22:08:10 for basic CRUD yes, not sure the total scope that will be in the initial commits 22:08:24 david-lyle: sounds good. we can always improve over time. 22:08:50 yeah, at least we'll have something to build on 22:09:02 anything myself or dolphm can do to make things smoother? 22:09:43 (since keystone v3 accounts for about 1/3 of the H1 milestone... ;-) ) 22:09:49 no, I think we're set, we do have questions regarding domain scoping 22:10:08 domain scoped tokens? 22:10:09 lcheng put it in the blueprint with screenshots 22:10:17 which blueprint? 22:10:25 no scoping operations to a domain in the UI 22:10:40 https://blueprints.launchpad.net/horizon/+spec/admin-domain-crud 22:12:02 if you are admin and looking at a domain it doesn't necessarily make sense to then go look at other panels and see all 22:12:14 projects/users/groups/etc 22:12:41 david-lyle: my initial thought is that for the case of an uber-admin with cross-domain privileges some sort of "active context" is probably a good idea, as per the suggestions in the whiteboard 22:12:49 so setting a domain scope to filter those panels makes domain adminstration a lot easier 22:12:51 ok 22:12:56 cool 22:13:09 okay, other blueprints 22:13:38 stevebaker: are you around to talk about the Heat UI at all? 22:14:41 anyhow, I messaged Steve about that BP this morning, I'm sure he'll get back to me about it. It may slip to H2, but we'll see. 22:15:01 here! 22:15:04 ah, h! 22:15:09 hi! 22:15:09 * stevebaker reads 22:15:22 basically just asking what the state of it is and what we can do to help 22:16:07 I'd love to let people start trying out Heat in Horizon as early in the cycle as possible (which'll bump adoption for y'all too) 22:16:17 I've been distracted by heat tempest stuff, but will get back onto horizon heat next. Current task is writing a test for the stack create workflow 22:16:29 which I'll need some help with. 22:17:05 yeah, I recall that there were two pieces that needed doing from the summit: the tests, and the "post the template and get back a form" workflow bit. 22:17:37 at least I'll be able to submit to gerrit once this test is done 22:17:39 If that second part is still unresolved then perhaps you could post the current state of your code for collaboration and we can collectively get that doe 22:17:43 cool 22:17:51 sounds great 22:17:56 keep us posted :-) 22:18:06 ok, cool 22:18:27 I don't see amotoki here... can anyone speak to the Quantum Security Groups blueprint? 22:18:37 I haven't had any updates there in a couple weeks. 22:19:05 eric peterson contacted him about that 22:19:13 good to know 22:19:16 he's not made any progress lately 22:19:22 this is the one I actually think is most likely to slip to h2 22:19:37 it's not a simple problem due to the differences in quantum and nova 22:19:40 he did provide a link to his work in progress but it dates back to sometime in grizzly 22:19:48 (per-port vs. per-server) 22:20:08 my guess is a slip, but I can't speak for amotoki 22:20:10 I'll check in with him myself, but don't be surprised if this one is deferred 22:20:12 yep 22:20:17 thanks for the update though 22:20:52 Tatiana Mazur indicated confidence in landing the DRY templates BP in H1, so I'd guess we'll see a review in the next week or so. 22:20:58 https://github.com/amotoki/horizon/tree/quantum-secgroup/openstack_dashboard/api 22:21:03 thanks 22:21:30 I believe that leaves the rest of the BPs as currently in code review (and most of those look pretty good as far as I could see today) 22:21:36 gabrielhurley: when you get the chance, can you perform an initial review https://github.com/gabrielhurley/django_openstack_auth/pull/41 22:22:16 lcheng2: I actually have looked at it and don't see any major issues, but don't wanna commit anything based on non-final code in keystoneclient 22:22:55 #topic bugs 22:23:01 There's one bug I wanna call attention to 22:23:13 and maybe other folks here can help sort out where we need to go with it 22:23:17 https://bugs.launchpad.net/horizon/+bug/1179526 22:23:19 Launchpad bug 1179526 in horizon "source_lang in Horizon repo is overwritten by Transifex" [High,Confirmed] 22:23:27 ok 22:23:52 my understanding is we've basically got the source language codes crossed between Horizon and Transifex and that's causing a lot of woes 22:25:08 going further, it sounds like we probably need to scrap the source language on Transifex and align it to the correct source language in Horizon 22:25:40 doing that may screw up the existing translations, so what we need to do (I think) is back up the actual current translations and be prepared to re-add them to transifex if the whole thing goes haywire 22:25:47 anyone got any input here? 22:26:51 no, sadly not. 22:26:55 heh. okay. 22:27:01 well, I'll post my thoughts on the ticket 22:27:03 No, might be worthwhile talking with those mysterious contacts we have in Transifex to see if they have any input on handling this 22:27:13 jpich: also a good idea 22:27:31 okay then 22:27:34 #topic open discussion 22:27:39 That's all I have 22:27:42 other folks? 22:28:42 I have high level multi-region support implemented and would like to review, but I need openstack-auth changes 22:29:12 I know it's not H1, but our environment needs it 22:29:52 david-lyle: I saw that one too. I think I misread the description the first time. 22:29:57 I'll take another look at it 22:30:30 ok, thanks! https://github.com/gabrielhurley/django_openstack_auth/pull/42 22:30:57 When we get that squared away we need to do some better differentiation between regions in a single service catalog vs. different auth endpoints (which is what Horizon's "regions" setting controls now) 22:31:20 I assume that'll happen in the Horizon review that accompanies the openstack-auth PR, though 22:31:59 yes, I'm not sure providing the current region picker is of much value 22:32:20 as a login is required when switching 22:32:46 It's actually super-useful for dev against multiple environments, and it originally came out of a use case at rackspace, I think. 22:32:53 but I'm not sure it's used very often these days 22:33:37 we use it, but essentially the same number of clicks would log you out and select a different region 22:33:54 I'm don't mind it, but I think it can be confusing 22:34:49 fair enough 22:34:57 we can discuss more later 22:35:03 sure 22:35:05 anybody else got other topics before I wrap things up here? 22:36:23 okay 22:36:31 thanks for the input, y'all 22:36:33 keep up the hard work 22:36:37 same time next week! 22:36:39 #endmeeting