15:01:02 <flwang1> #startmeeting Zaqar 15:01:02 <openstack> Meeting started Mon Dec 14 15:01:02 2015 UTC and is due to finish in 60 minutes. The chair is flwang1. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:07 <openstack> The meeting name has been set to 'zaqar' 15:01:32 <flwang1> #topic roll call 15:01:35 <flwang1> o/ 15:01:38 <vkmc> o/ 15:02:03 <njohnston> o/ 15:02:10 <ryansb> o/ 15:02:18 <mdnadeem1> o/ 15:02:22 <Eva-i> o/ 15:04:32 <flwang1> hi guys, let's start, i don't have much topics today so feel free to contribute :) 15:04:47 <flwang1> #topic code review 15:05:13 * ryansb has been bad and owes reviews 15:05:55 * vkmc too 15:06:24 <flwang1> zaqar client is still our priority in M-2 15:06:43 <flwang1> so let me highlight the gaps currently 15:07:18 <flwang1> we only have some patches for v1 which are in the waiting-for-love status 15:07:24 <flwang1> means just need another +2 15:07:33 <flwang1> then we can complete all the v1 work 15:08:09 <flwang1> vkmc: flaper87: they especially needs yours love :D 15:08:24 <vkmc> flwang1, roger that 15:08:26 <Eva-i> especially this one https://review.openstack.org/#/c/248982/ 15:10:11 <flwang1> Eva-i: yep, that one is important 15:10:35 <flwang1> for v2, we are still missing the supports for claims and subscriptions 15:10:57 <flwang1> the claims support is here https://review.openstack.org/213165 15:11:22 <flwang1> and i'm working on the subscriptions, i will submit some patches for subscriptions today soon 15:11:44 <vkmc> thanks for taking care of that 15:13:26 <flwang1> and another big progress is our UI repo 15:13:28 <flwang1> see https://review.openstack.org/255213 15:13:43 <flwang1> pls try it on your local with devstack :) 15:14:10 <flwang1> with this plugin, now we can have a new panel group named 'messaging' under 'project' dashboard 15:14:20 <mdnadeem1> cool, i ll try it 15:14:34 <flwang1> and a 'queues' panel under 'messaging' 15:15:46 <vkmc> sweet! 15:16:08 <dynarro> that's cool =) 15:16:13 <flwang1> btw, i would suggest we only focus on the queues(subscriptions) panel for Mitaka, and leave the 'pools' and 'flavors' panels to N, since they admin only actions 15:16:21 <Eva-i> mdnadeem1: I see you nicely modified claims patch :o 15:16:53 <mdnadeem1> Eva-i, just give a try :) 15:17:10 <ryansb> yeah, that'd be reasonable 15:17:23 <flwang1> vkmc: ryansb: i have talked with some horizons guys and i would like to get at least one +1 from horizon team before we merge patch for zaqar ui repo 15:17:34 <ryansb> yeah definitely 15:17:40 <flwang1> vkmc: ryansb: flaper87: are you ok with that? 15:18:03 <ryansb> sounds good 15:18:08 <flaper87> ++ 15:18:12 * flaper87 <- in a call 15:18:23 <flwang1> cool 15:19:09 <flwang1> flaper87: thanks for joining us :) 15:19:33 * flaper87 misses Zaqar... A LOT! 15:19:51 <mdnadeem1> flwang1, we need to discuss about https://review.openstack.org/#/c/246147/ 15:20:34 <flwang1> mdnadeem1: clicking... 15:21:16 <flwang1> mdnadeem1: thanks for the heads up, i would like to put it into next topics, new stuff for Mitaka 15:21:33 <mdnadeem1> ok 15:21:48 <flwang1> anything we need to discuss against the code-review topic? 15:22:14 <ryansb> mdnadeem1: I'll put that on my to-review list 15:22:19 <flwang1> #topic new stuff in Mitaka 15:22:27 <mdnadeem1> ryansb, Thanks 15:22:40 <flwang1> HTTP PUT enhancement https://review.openstack.org/#/c/246147/ 15:23:33 <ryansb> #link 15:23:36 <ryansb> #link https://review.openstack.org/#/c/246147/ 15:23:41 <flwang1> mdnadeem1: would you like to give a basic introduction? 15:24:00 <mdnadeem1> flwang1; sure 15:25:17 <mdnadeem1> currently, put operation on existing resources like queue, pool flavor 15:25:34 <mdnadeem1> update the resource attribute 15:26:38 <mdnadeem1> however, we don't want it as there is patch method to update the resources 15:27:16 <mdnadeem1> so, this spec will change the PUT operation behavior incase of existing resource 15:27:46 <mdnadeem1> Is that make sense 15:27:47 <mdnadeem1> ?? 15:28:22 <flwang1> let me explain more 15:28:32 <mdnadeem1> ok 15:28:40 <flwang1> currently, we're using PUT for queue/pool/flavor to create new resources 15:29:08 <flwang1> instead of POST 15:29:35 <flwang1> that makes sense since for those resources, user know the 'name' 15:29:51 <flwang1> instead of creating by zaqar server 15:31:00 <flwang1> however, when user use the same parameter for an existing resource, or let's say user use some different parameters for an existing name 15:31:10 <flwang1> the object will be updated 15:31:32 <flwang1> that also makes sense :D 15:31:46 <flwang1> since for PUT method, it should be idempotent 15:32:19 <flwang1> however, from the user perspective, it's a little bit weird 15:32:38 <flwang1> since they may update an existing object by mistake 15:32:53 <flwang1> am I make it more clear or worse? :) 15:33:21 <dynarro> no, that was actually very clear, thanks flwang1 15:33:58 <flwang1> so after discussed with kgriffs 15:34:08 <flwang1> we proposed the solution is like this 15:35:13 <flwang1> 1. Given the PUT method is reasonable from the HTTP design perspective, so we don't need need it 15:35:29 <flwang1> 2. Add a new 'HEAD' support for those resources 15:36:05 <flwang1> 3. At zaqar client, when user intend to update resources, zaqar client call 'HEAD" firstly 15:37:26 <flwang1> 4. if there is an existing resource, just update it, otherwise, say not found 15:38:26 <flwang1> 5. when user try to 'create' a new resource, zaqar client will call 'HEAD" firstly as well, if there is one, say 'existed', otherwise, create it 15:38:40 <flwang1> mdnadeem1: am i missing anything? 15:38:54 <mdnadeem1> no, perfect :) 15:39:16 <flwang1> flaper87: ryansb: vkmc: how do you think? 15:39:42 <ryansb> that does make sense, since PUT is for new items 15:40:29 <mdnadeem1> exactly ^^ 15:40:39 <vkmc> makes sense to me 15:40:58 <flwang1> awesome 15:40:59 <vkmc> still, I'm a bit hesitant of using HEAD 15:41:07 <flwang1> vkmc: why? 15:41:08 <vkmc> for one of the endpoints we decided to use PATCH 15:41:29 <vkmc> shouldn't use the same for updates? 15:42:02 <flwang1> we should use 'PATCH' for update for sure, if the endpoint support that 15:42:19 <flwang1> we can manage the logic in zaqar client side i think 15:42:23 <flwang1> let me check 15:43:22 <Eva-i> we must create a logic that forces the client implementor to do things right 15:43:26 <flwang1> `queues` doesn't support 'PATCH' 15:43:40 <flwang1> pools and flavors support 'PATCH' 15:44:35 <flwang1> anyway, we can discuss details by adding comments on that patch :) 15:44:42 <flwang1> we only have 15 mins now 15:46:08 <flwang1> ok, the new stuff i would like to do in Mitaka is https://blueprints.launchpad.net/zaqar/+spec/more-reserved-queue-attributes 15:46:36 <flwang1> to support more built-in attributes for queue, i have mentioned it in previous meetings 15:46:49 <flwang1> i will submit a new spec this week, just FYI 15:47:16 <flwang1> vkmc: any progress for the sqlalchemy upgrade? 15:47:58 <vkmc> flwang1, no updates about that, will try to start it soon though 15:48:08 <flwang1> vkmc: nice, thanks 15:48:35 <flwang1> jasondotstar: any updates for the puppet work? 15:48:41 <jasondotstar> yep 15:48:45 <jasondotstar> I owe you guys a patch 15:48:50 <jasondotstar> about to submit it now 15:48:52 <jasondotstar> but.... 15:49:00 <jasondotstar> there's more cleanup work to be done 15:49:08 <jasondotstar> still you can see the progress and try it out 15:49:14 <jasondotstar> I'm adding in the .conf settings 15:49:17 <jasondotstar> that you gave me 15:50:11 <flwang1> jasondotstar: cool, should i expect the patch this week ? ;) 15:50:36 <jasondotstar> absolutely. 15:50:44 <jasondotstar> I'll add you as a reviewer 15:50:53 <jasondotstar> that way you can see each patch i submit 15:51:22 <flwang1> jasondotstar: excellent, thanks 15:51:54 <flwang1> #topic docs 15:52:11 <flwang1> Eva-i: could you pls give a quick summary about the status of our docs? 15:52:33 <Eva-i> flwang1: yep 15:54:07 <flwang1> Eva-i: thanks 15:54:16 <flwang1> you have 6 mins :) 15:57:28 <flwang1> Eva-i: hmm... we're waiting :) 15:57:50 <Eva-i> flwang1: contributor docs are under refactoring as well as WIki. In the mid-january doc team will migrate API reference to a new format and Zaqar will create API reference in the official place (http://developer.openstack.org/api-ref.html). For now we can work on contributor docs, Wiki, configuration ref, and admin guide. 15:58:25 <flwang1> admin guide? 15:58:37 <Eva-i> Docs team will also combine "Administrator User Guide" and "Cloud Administrator Guide" soon. 15:58:49 <flwang1> ok 15:58:51 <Eva-i> And it will be called just "Administrator Guide" 15:59:04 <flwang1> but i think we don't have much bandwidth to work on the admin guide 15:59:05 <Eva-i> https://github.com/openstack/docs-specs/blob/master/specs/mitaka/user-guides-mitaka-reorg.rst here are details 15:59:16 <flwang1> ok, thanks 15:59:43 <flwang1> i was assuming the config ref and api -ref are our top priorities in Mitaka 15:59:49 <flwang1> are we on the same page? 16:00:10 <Eva-i> flwang1: yes 16:00:16 <flwang1> cool 16:00:17 <flwang1> ok 16:00:28 <flwang1> let's back to #zaqar channel :) 16:00:36 <flwang1> thank you, guys 16:00:39 <Eva-i> oki, thanks 16:00:46 <flwang1> #endmeeting