20:59:25 <flwang> #startmeeting zaqar 20:59:25 <openstack> Meeting started Mon Dec 7 20:59:25 2015 UTC and is due to finish in 60 minutes. The chair is flwang. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:59:26 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:59:29 <openstack> The meeting name has been set to 'zaqar' 20:59:35 <flwang> #topic roll call :) 20:59:38 <Eva-i> Hello 20:59:44 <flwang> Eva-i: hi 20:59:49 <flwang> what's your time now? 21:00:40 <flwang> Eva-i: if there is only you and me, this could be a short meeting :) 21:00:59 <Eva-i> my time is 00:00 21:01:02 <flwang> ryansb: o/ 21:01:05 <ryansb> hey folks 21:01:12 <flwang> Eva-i: oh, thanks for joining 21:01:16 <Eva-i> But I'm awake till 06:00 21:01:17 <flwang> ryansb: thanks for joining 21:01:18 <Eva-i> =) 21:02:01 <flwang> Eva-i: you must be young, i can't do that after 30 :) 21:03:11 <Eva-i> flwang: yep 21:03:17 <flwang> ok, let's start, i will miss vkmc and kick flaper87 21:03:37 <flwang> #topic code review 21:04:02 <flwang> we made a great progress in Mitaka-1 21:04:29 <flwang> thank you very much for everyone, especiall ryansb, vkmc, Eva-i and MD 21:04:30 <ryansb> not quite code review, but we have lots of patches in merge conflict 21:04:37 <flwang> ryansb: haha 21:04:52 <flwang> ryansb: and gate failure :) 21:05:11 <ryansb> yeah, I haven't had a chance to look at that yet 21:05:17 <flwang> but we made it, though we still have some patches in the air 21:06:07 <flwang> now the v1 client support is almost 'done', still missing the flavor list and create, pls blame /me 21:06:11 <ryansb> my point was, if you have patches out, please rebase them 21:06:24 <flwang> ryansb: sorry, yep 21:06:43 <flwang> ryansb: yep, i know you did a lot rebase for MD's patches 21:06:56 <ryansb> hey, no worries. You made *tons* of progress anyways 21:07:31 <Eva-i> flwang: thank you too, you are MVP 21:07:38 <Eva-i> IMO 21:07:39 <flwang> ryansb: i hope we can complete the client work in this week 21:08:26 <flwang> i will upload new patchset for flavor list/create and v2 subscription/claim 21:08:33 <flwang> pls review them :) 21:09:01 * vkmc is attending another meeting 21:09:04 <ryansb> will do 21:09:05 * vkmc waves to the zaqar team 21:09:12 <flwang> vkmc: i miss you 21:10:09 <flwang> anything we need to highlight for code review? 21:10:35 <flwang> #topic docs 21:10:42 <ryansb> I'd like to highlight https://review.openstack.org/#/c/253731/ 21:11:06 <ryansb> basically want feedback on how folks expect the CLI to be used for signedurls 21:11:19 <ryansb> (flwang and Eva-i, thanks for your feedback so far) 21:11:38 * ryansb stops trampling over the docs topic with code review stuff 21:11:40 <flwang> ryansb: i think for now 21:11:54 <flwang> what we can do is print all the information to end user 21:12:07 <ryansb> makes sense 21:12:13 <flwang> and we can improve it if we can any feedback 21:12:19 <ryansb> cool 21:12:21 <Eva-i> okay 21:12:47 <flwang> ryansb: if user has the permission to see it, show it 21:14:07 <flwang> ok, cool 21:14:12 <flwang> let's talk about the docs 21:14:21 <flwang> now the spec of config ref has been merged 21:14:35 <flwang> so we can start to contribute the config ref for zaqar 21:14:48 <flwang> it's great 21:15:03 <ryansb> yeah, so what's the format for the API ref, will we be able to use swagger? 21:16:15 <flwang> for the api ref, we need to contribute to here https://github.com/openstack/api-site/blob/master/api-ref 21:16:44 <flwang> anne is proposing to use swagger, but i think it's not now 21:16:56 <flwang> for now, we still need to follow the common way, like ^ 21:17:15 <flwang> if anybody can take it, it would be super awesome 21:17:26 <flwang> since the rest api of zaqar is the critical part 21:17:57 <Eva-i> I think I can take it as well 21:18:10 <flwang> Eva-i: are you sure you have enough bandwidth? 21:19:05 <flwang> Eva-i: based on my discussion with lana, we(zaqar) are allowed only focus on one doc 21:19:19 <flwang> means one of config ref, api ref and user guide 21:19:26 <Eva-i> flwang: yes, I'm sure. I have a proposal 21:19:55 <flwang> but i'm always greedy, i want to do 2 or more 21:20:18 <flwang> i would like to see config doc for sure, and i will do that 21:20:18 <Eva-i> flwang: only on one doc? Let me do config ref then. 21:20:27 <flwang> haha 21:20:35 <flwang> we can argue with docs team 21:20:48 <flwang> i think they just worry about the reviewing 21:20:51 <Eva-i> flwang: also https://blueprints.launchpad.net/openstack-manuals/+spec/zaqar-user-guide is too wide. We should narrow it to one type of all guide types. 21:21:15 <flwang> if we prepare everything and just need a +2 from them, then i think it shouldn't be too hard for them 21:21:52 <flwang> Eva-i: then how about let's focus on ap-ref in M-2 21:22:05 <flwang> and we can work together in M-3 for user guide 21:22:22 <flwang> IMO, ap-ref needs a big effort 21:22:26 <Eva-i> M-2 just started, right? 21:22:28 <ryansb> yeah 21:22:32 <ryansb> well, last week 21:23:52 <Eva-i> flwang: what about config ref that is already created? 21:24:09 <Eva-i> *spec approved 21:24:50 <flwang> Eva-i: what did you mean? 21:25:09 <flwang> Eva-i: i mean i will take care the config ref 21:25:29 <flwang> and if you want, i think ap-ref has a higher priority than user guide 21:25:58 <Eva-i> flwang: alright, lets argue with doc team and based on the result we'll be working on config ref only or on both config ref and api ref. 21:26:02 <flwang> ryansb: how do you think? 21:26:29 <flwang> Eva-i: i would suggest, write a spec and submit it, and meanwhile, start to coding/writing the api-ref 21:26:41 <Eva-i> flwang: okay 21:26:43 <flwang> don't wait for the result from doc team 21:27:04 <ryansb> yeah, you can always write it and submit it later 21:27:21 <flwang> so even we can't make it happen in Mitaka, at least it will be merged at early of N 21:27:26 <Eva-i> flwang: I'll start coding api-ref then. I'll convert existing api-ref into appropriate format and add new things. 21:28:12 <flwang> Eva-i: awesome, before that, i think we just need to confirm with doc team if we still need to follow current way, like https://github.com/openstack/api-site/blob/master/api-ref 21:30:15 <Eva-i> flwang: yes 21:31:16 <flwang> Eva-i: awesome, thanks 21:31:32 <flwang> anything else under this topic? 21:32:04 <flwang> #topic UI 21:32:06 <Eva-i> flwang: so what to do with https://blueprints.launchpad.net/openstack-manuals/+spec/zaqar-user-guide ? Doc team doesn't understand what type of user guide this mean 21:32:23 <flwang> Eva-i: that's the feedback from doc team? 21:32:34 <Eva-i> flwang: yes. 21:32:37 <flwang> then i will leave a comments on that blueprint 21:32:48 <flwang> and figure out 21:33:18 <Eva-i> flwang: it should be either api-ref or something. Or "admin-guide-cloud". We have to use their terminology. 21:33:37 <flwang> personally, i would like to see the most important one, but i'm not sure if we have the resource, and i'm not a native english speaker :( 21:33:49 <flwang> admin-guide-cloud 21:34:14 <flwang> but i need to figure out the workload and our bandwidth 21:34:20 <flwang> we can discuss it offline 21:34:29 <Eva-i> okay 21:34:31 <flwang> i know it's complex 21:34:44 <flwang> really appreciate for your effort on this domain 21:35:14 <flwang> as for UI, now we have a repo and my goal is having a queue panel in M-2 21:35:46 <flwang> and complete the pool (and flavor?) panel in M-3 21:36:28 <flwang> or the subscription/notification panel for M-3 21:37:04 <flwang> given the pool and flavor are admin only actions 21:37:14 <flwang> not used too much 21:37:16 <flwang> any comments? 21:37:33 <Eva-i> I remember exploreshafali was working on pool and flavor design 21:37:51 <Eva-i> Did she worked on queue desing? 21:37:57 <Eva-i> *design 21:38:20 <flwang> Eva-i: shafali is busy now, she won't work on the UI 21:39:05 <flwang> i will try to make a basic shape of UI project and any help is welcome 21:39:58 <flwang> that's basically my plan 21:40:33 <Eva-i> flwang: I know she's busy, I was just asking if she might already invented something for queue panel 21:41:02 <flwang> Eva-i: i'm afraid no 21:41:14 <flwang> given there is no repo :) 21:41:20 <flwang> there was 21:41:37 <Eva-i> flwang: she made some sketches 21:41:49 <flwang> anyway, when I propose the first commit for zaqar ui, i will add you guys as reviewer 21:42:02 <Eva-i> flwang: okay 21:42:09 <flwang> Eva-i: the sketches is some graph design, not code 21:42:17 <flwang> we have reviewed it on tokyo summit 21:43:05 <Eva-i> flwang: so we already have graph designs? 21:43:41 <flwang> Eva-i: you can say that :) 21:43:48 <flwang> it's very simple i would say 21:44:16 <Eva-i> okay =) 21:45:38 <flwang> Eva-i: but we also don't need a very complex UI 21:45:50 <flwang> it's a normal panel just like images :) 21:46:12 <Eva-i> flwang: yep 21:46:21 <flwang> #topic integration with other projects 21:46:26 <flwang> 1. Sahara 21:47:02 <flwang> i just checked with ethan from sahara team, seems sahara team is not keen to use zaqar for their agent, at least for now 21:47:14 <flwang> so we shouldn't expect to see the integration in Mitaka 21:47:45 <flwang> 2. Searchlights 21:48:01 <flwang> the integration with searchlight is in progress, see https://review.openstack.org/246220 21:48:28 <flwang> 3. Horizon 21:48:50 <flwang> the integration of horizon may depend on the integration with searchlight 21:49:11 <flwang> since we need to get the notifications/messages into zaqar queues, then notify horizon to refresh panels 21:49:58 <flwang> here is the PoC if you're interested in https://review.openstack.org/#/c/246009/ 21:50:18 <flwang> i have talked with mrunge, but haven't get any feedback yet 21:50:23 <flwang> any comments? 21:51:25 <flwang> so my plan is supporting searchlight team to get a good integration, and workable version by the end of M-2 21:52:15 <flwang> meanwhile, interlock with horizon team to see how to leverage that work for horizon 21:52:29 <flwang> not sure if we can make it happen in Mitaka 21:52:45 <flwang> sometimes, something out of control 21:55:03 <Eva-i> your plan is good 21:55:37 <Eva-i> so far I saw no one from searchlight team asking for help in zaqar chat 21:55:43 <flwang> Eva-i: haha, plan is always good, progress is always surprising :) 21:56:01 <flwang> Eva-i: lei-zhi1 is the guys from searchlight team 21:56:12 <flwang> i have some talks with him 21:56:32 <flwang> he is a chinese, so we talked in Chinese in private channel, sorry about that 21:56:53 <flwang> i just want to make it faster 21:57:11 <flwang> i will ask him speak up in zaqar channel next time :) 21:57:18 <flwang> in Chinese 21:57:54 <flwang> ok, 2 mins left 21:58:03 <flwang> anything we need to discuss? 21:58:13 <ryansb> I don't think so 21:58:16 <flwang> otherwise, let's back to our channel and rock on :) 21:58:18 <flwang> cool 21:58:18 <Eva-i> yes, I want 21:58:31 <flwang> Eva-i: ok, listening... 21:59:05 <Eva-i> I'd like this to implement during mitaka https://blueprints.launchpad.net/zaqar/+spec/websocket-binary-support 21:59:20 <Eva-i> It'll be great if you check it out. 21:59:25 <Eva-i> Thanks 21:59:45 <flwang> ok, sure 21:59:49 <flwang> #endmeeting