21:03:08 <flwang> #startmeeting Zaqar
21:03:08 <openstack> Meeting started Mon Mar 14 21:03:08 2016 UTC and is due to finish in 60 minutes.  The chair is flwang. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:03:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:03:11 <openstack> The meeting name has been set to 'zaqar'
21:03:22 <flwang> #topic roll call
21:03:24 <flwang> o/
21:04:06 <Eva-i> o/
21:04:15 <ryansb> o/
21:04:38 <flwang> flaper87: vkmc: around?
21:05:37 <flwang> ok, let's tart
21:05:50 <flwang> #topic code review
21:06:09 <flwang> yeah! we don't have much patches on the list
21:06:16 <ryansb> hurrah
21:06:19 <Eva-i> =)
21:06:44 <flwang> basically, all the patches targeted for RC1 have been merged
21:06:56 <flwang> unless i missed anything
21:07:48 <flwang> but we do have two patches for client
21:08:11 <flwang> 1. queue metadata set/get
21:08:30 <flwang> 2. ping health support ( i hate myself missed this one)
21:08:41 <Eva-i> Alright, I'll review these.
21:08:43 <flwang> #link https://review.openstack.org/290930
21:08:55 <flwang> #link https://review.openstack.org/289179
21:09:09 <flwang> btw, we have removed the pypy gate job
21:09:18 <flwang> so this one is ready to go https://review.openstack.org/290830
21:09:59 <ryansb> cool, thx for linking
21:10:29 <flwang> besides, i would like to make v2 as the default version for zaqar client if you don't mind :) https://review.openstack.org/290933
21:11:13 <Eva-i> vkmc said about patch: https://review.openstack.org/#/c/265723/ "We are going to tackle the comments pointed out by Eva during RC1." If you want, we still can tackle them. But as I know, you don't want it, right?
21:11:58 <flwang> verify the metadata for v1 and v1.1?
21:12:08 <Eva-i> flwang: yep
21:12:11 <Eva-i> Oki, let's make v2 as the default version for zaqar client.
21:12:24 <Eva-i> Hope no one will complain.
21:12:38 <ryansb> ooh, +1
21:12:41 <ryansb> that'd be great
21:12:42 <flwang> Eva-i: no body will complain because
21:12:59 <flwang> v1.1 and v2 doesn't really work before mitaka, you know
21:13:17 <flwang> we fixed so many functions gaps in this release for client side
21:13:30 <Eva-i> flwang: aha, that's cool
21:13:55 <flwang> Eva-i: if you can fix the verifications by Thu, we can get it done
21:14:03 <flwang> Fri is the last day for RC1
21:14:21 <Eva-i> flwang: yes, you can count on me.
21:14:38 <Eva-i> flwang: glad you changed your mind
21:14:48 <flwang> Eva-i: ok, you will be pushed :D
21:15:11 <Eva-i> oki =)
21:15:48 <flwang> Eva-i: thank you very much
21:16:08 <flwang> anything else?
21:16:50 <Eva-i> nothing from me. I'll review python-zaqarclient thingies.
21:17:17 <flwang> #topic zaqar UI
21:17:38 <flwang> now we have 3 awesome devs from horizon team for zaqar UI
21:17:50 <flwang> so pls pls pls review and try the zaqar UI
21:18:00 <flwang> and return them your feedback
21:18:18 <flwang> P.S. that's my job for today :)
21:18:21 <Eva-i> Okay, I'll try to review it, thought I don't understand many things there. And I'll run it.
21:19:04 <flwang> Eva-i: you don't have to understand the tech details, just use it :)
21:19:19 <flwang> like what i'm doing, haha :D
21:19:35 <Eva-i> flwang: alright, I'll test it thoroughly
21:20:33 <flwang> Eva-i: thanks
21:22:27 <Eva-i> let's go to the open discussion now
21:23:20 <flwang> Eva-i: ok :)
21:23:32 <ryansb> ok: opening item is that I haven't been able to do a good solution on the config copying, has anyone else had a chance to work on it?
21:23:43 <flwang> #topic open discussion
21:23:56 <flwang> ryansb: do we really have to fix it in RC1?
21:24:00 <Eva-i> ryansb: I think it's pointless to copy it
21:24:26 <Eva-i> flwang: seems like we don't have too. It's only serious under special conditions.
21:24:30 <ryansb> is there a way to find what configs a ConfigOpt came from?
21:25:08 <Eva-i> ryansb: what do you mean?
21:25:20 <ryansb> So an alternative to copying the config object
21:25:28 <ryansb> would be to instantiate a new one from the same files
21:25:30 <ryansb> right?
21:27:46 <Eva-i> ryansb: yes, but some option values are not taken from files, but from cli arguments
21:27:56 <ryansb> hrm
21:28:29 <Eva-i> We don't need to create 'conf' with all same options.
21:29:05 <flwang> btw, now we have created zaqar-ui on launchpad https://launchpad.net/zaqar-ui
21:29:14 <Eva-i> We have to create a conf with options which are needed by our storage drivers. And libs used in storage drivers. We have to find these libs.
21:29:23 <Eva-i> flwang: thank you, Fei Long
21:29:30 <flwang> sorry, guys. let's discuss details offline
21:29:53 <Eva-i> flwang: ryansb: let's go to zaqar chat?
21:30:01 <flwang> one thing i would like to get your opinion is the one i mentioned several hours before
21:30:12 <ryansb> yeah, zaqar channel is fine
21:30:13 <flwang> if we would like to refactor the PATCH?
21:30:36 <Eva-i> flwang: I like the change. I don't like it's backward incompatibility.
21:31:08 <ryansb> yeah, pretty much same here
21:31:16 <flwang> i asked, because if we change current PUT of queue to fix a wrong usage of PATCH, it's ridiculous, right?
21:32:29 <Eva-i> flwang: not very ridiculous. It will solve the problem with much more compatibility.
21:33:27 <flwang> i can't say it's not correct :)
21:33:27 <Eva-i> we can always make proper PATCH method in our next API.
21:34:33 <flwang> ryansb: how do you think reverting the change for https://github.com/openstack/zaqar/blob/master/zaqar/transport/wsgi/v2_0/queues.py#L141
21:34:51 <flwang> so that we don't have to introduce another strange change just for a wrong usage of PATCH?
21:35:18 <ryansb> That would be ok with me - I think the change is recent enough we would be better off reverting it
21:35:22 <flwang> i mean new metadata will be fully used to save into db instead of using dict.update()
21:35:28 <Eva-i> flwang: I'm okay with it.
21:35:39 <ryansb> then we can do PATCH the right way in Newton
21:35:41 <Eva-i> Let's revert it
21:35:51 <flwang> ryansb: yes
21:36:03 <flwang> for queue, pool and flavor :(
21:36:07 <flwang> it's a shame :(
21:37:00 <flwang> ok, another thing is the topics for summit
21:37:11 <flwang> i think we have discussed it before in weekly meeting
21:37:15 <Eva-i> unlike queue metadata, pool and flavor attributes exist always
21:37:23 <flwang> i'm going to create an etherpad to track it
21:37:30 <Eva-i> so maybe we don't need to change pool and flavors?
21:37:40 <flwang> Eva-i: not only metadata, but the usage of PATCH
21:37:57 <flwang> Eva-i: pls read the reference of the bp
21:38:43 <flwang> oh, btw, i will run for PTL for Newton if you'll have me :)
21:38:55 <Eva-i> ah, sorry, I said bad thing
21:39:41 <flwang> Eva-i: it's ok. it needs some background
21:40:08 <flwang> if anybody want to run for PTL as well, it would be nice
21:40:20 <Eva-i> I like you as PTL
21:40:41 <flwang> Eva-i: thanks for the trust
21:40:49 <flwang> ok, anything else?
21:41:03 <flwang> i think we can back to zaqar channel now
21:41:50 <Eva-i> I Just can say that I finally prepared virtual machines and going to test "production" installation instruction. =)
21:41:51 <ryansb> k, see you over there
21:42:00 <Eva-i> oki, let's go and discuss conf
21:42:30 <flwang> thank you, guys
21:42:33 <flwang> #endmeeting