16:00:43 <etoews> #startmeeting api wg
16:00:44 <openstack> Meeting started Thu Aug 27 16:00:43 2015 UTC and is due to finish in 60 minutes.  The chair is etoews. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:45 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:47 <openstack> The meeting name has been set to 'api_wg'
16:00:50 <Nikolay_St> hi
16:00:57 <elmiko> o/
16:00:59 <cdent> hola
16:02:25 <Juno> tidwellr: you mean talk about BGP in this channel now?
16:02:54 <etoews> Juno: wrong timing. i think your people have moved to the neutron channel.
16:03:04 <etoews> #topic agenda
16:03:11 <etoews> #link https://wiki.openstack.org/wiki/Meetings/API-WG#Agenda
16:03:26 <Juno> etoews: what is the channel name?
16:03:53 <etoews> i'm going to guess #openstack-neutron but i have no idea really
16:03:56 <etoews> #topic previous meeting action items
16:04:40 <etoews> #link http://eavesdrop.openstack.org/meetings/api_wg/2015/api_wg.2015-08-13-16.01.html
16:04:55 <Juno> etoews: thanks
16:05:04 <etoews> np
16:05:34 <etoews> looks like there wasn't a meeting from last week so those action items are from 2 weeks ago.
16:05:36 <elmiko> well, i put up the etherpad, and started to fill it in. but not too much there yet
16:05:45 <edleafe> o/
16:05:48 <elmiko> #link https://etherpad.openstack.org/p/mitaka-api-wg-session-plans
16:06:09 <elmiko> also, i will be in tokyo, so i can help moderate those sessions. assuming no collisions
16:06:19 <etoews> awesome. thanks elmiko.
16:06:37 <elmiko> do we know if we'll have 2 sessions, or 1 big one, or?
16:08:04 <etoews> are you asking cdent?
16:08:19 <cdent> sorry
16:08:22 <etoews> (or anyone else who will be at the summit for that matter)
16:08:23 <cdent> was in a different window
16:08:31 <elmiko> yea, i guess so. i can't remember, did you request 2 sessions cdent ?
16:08:34 <cdent> I signed us up on the spreadsheet for a single long session
16:08:39 <elmiko> ok, cool
16:08:59 <elmiko> i'll adjust the pad to be more agenda related then, instead of broken out by session
16:09:01 <cdent> I wrote to the organizer to ask about whether we were supposed to split or could have more and got no response (this was the day after the last time I was at a meeting)
16:09:12 <elmiko> ah ok
16:09:26 <cdent> the precedent on the spreadsheet was that people were doing a single block
16:09:28 <elmiko> also, i could probably use some help filling out the content for the session
16:09:48 <cdent> elmiko: I'll try to help with that
16:09:50 <elmiko> etoews: not sure if you have any items we should address at the summit
16:09:53 <elmiko> cdent: thanks!
16:10:51 <etoews> nothing in particular. a state of the union to kick things off would be good.
16:11:27 <etoews> discussion on getting more people engaged.
16:11:43 <elmiko> that's a biggie
16:11:52 <etoews> here's an important one, discussion on getting people to use the guidelines in reviews
16:12:32 <elmiko> cool, writing this down on the pad
16:12:42 <cdent> yeah, being able to habitiually link to sections of the guidelines wil be handy
16:12:51 <etoews> ultimately we want people to be reviewing based on the guidelines. -1 because you didn't follow guideline x [LINK]. +1 because you did. etc.
16:13:32 <etoews> i noticed some interesting api related sessions in the sched on the main conf side of things.
16:13:42 <etoews> #link https://www.openstack.org/summit/tokyo-2015/schedule/design-summit
16:13:51 <etoews> #link https://mitakadesignsummit.sched.org/event/97bb02c6bbdfa5f5699b82ea1be2a810?iframe=no&w=&sidebar=yes&bg=no
16:14:01 <etoews> #link https://mitakadesignsummit.sched.org/event/f4d9a391b6bdd09d6bcd9a5571fa2739?iframe=no&w=&sidebar=yes&bg=no
16:14:05 <etoews> there's more too
16:15:42 <etoews> the next topic is on Mitaka sessions but i guess we already covered that right elmiko?
16:15:57 <elmiko> yea, i think so
16:17:04 <etoews> #topic guidelines
16:17:11 <etoews> #link https://review.openstack.org/#/q/status:open+project:openstack/api-wg,n,z
16:18:04 <etoews> i want to replace that link with a link to our dashboard. it's almost ready.
16:18:14 <elmiko> sweet
16:18:18 <ryansb> nice
16:18:19 <Nikolay_St> nice
16:18:24 <etoews> it'll make things a bit more clear for where guidelines are at.
16:19:07 <etoews> it's definitely not perfect and still needs eyes on it to make sure things don't slip through the cracks (w.r.t. what the dashboard query system is actually capable of)
16:19:26 <elmiko> i thought sdague was going to remove the dep. from https://review.openstack.org/#/c/189738/ , should we just repost it without the dep?
16:20:18 <sdague> sorry, I lost track of that one
16:20:30 <sdague> I honestly thought we'd get to resolution on the 501 issue well before now
16:20:31 <elmiko> no worries, i figured you had higher prio stuff =)
16:20:44 <elmiko> i wish we woud have...
16:21:06 <etoews> sdague: i put the freeze on the 501 to move it forward.
16:21:19 <sdague> so, given that it looks like the 501 one is schedule for merge next week, I'd just say we move forward assuming that will merge
16:21:26 <etoews> we don't have unanimity but that's okay.
16:21:52 <elmiko> ok, fair
16:21:53 <etoews> ya. at this point it's (hopefully) just a waiting game.
16:22:13 <elmiko> i just felt bad because the stuck one has been there since vancouver =(
16:23:07 <sdague> yeh, that's fine. It's a slow march :)
16:23:13 <elmiko> cool
16:23:22 <etoews> i'd like to highlight this process change about removing the voting restriction #link https://review.openstack.org/#/c/217764/
16:23:47 <etoews> elmiko: can you please move your +2 to +1 since it's not frozen yet?
16:23:48 <elmiko> i think it's a step in the right direction
16:24:19 <elmiko> yea definitely, sorry bout that. i was a little unsure about the process as opposed to guidelines
16:24:41 <etoews> that's a fair point
16:25:38 <elmiko> it seems ok to me if we just stick to the rule that +2 is only for freeze. but process type stuff won't need a freeze, so we can just get some consensus then merge. does that sound fiar?
16:25:39 <etoews> okay. forget what i said. :)
16:25:46 <elmiko> lol, too late ;P
16:25:52 <etoews> ++
16:26:12 <elmiko> so, gather a few more +1 then we can merge it
16:26:33 <etoews> sure
16:26:57 <etoews> there was some interest in versioning on the mailing list.
16:26:59 * ryansb +1'd
16:27:21 <etoews> alex_xu: think you'll have some time to update your microversion guidelines?
16:28:09 <etoews> i need to update #link https://review.openstack.org/#/c/167793/
16:28:37 <etoews> jaypipes: do you have some time to look at https://review.openstack.org/#/c/183599/ ?
16:29:35 <etoews> has anyone seen Steven Kaufer around lately?
16:29:47 <etoews> we need to figure out what to do with https://review.openstack.org/#/c/162716/
16:30:36 <etoews> i'm inclined to abandon it and proceed as i mentioned in my comment there.
16:30:42 <cdent> +1
16:30:43 <elmiko> is there an amount of time, after which, it is not impolite to take over a review?
16:30:58 <etoews> i'd say yes
16:31:17 <elmiko> so, like a month maybe? then it's fair to rewrite/update/etc
16:31:27 <elmiko> or is that too long
16:31:30 <etoews> ya. that sounds about right to me.
16:31:33 <elmiko> cool
16:31:37 <ryansb> 1 month seems legit
16:31:46 <elmiko> maybe we can unstuck some of these older guidelines
16:31:53 <ryansb> I imagine after a month the number of people that would complain would be quite low
16:31:54 <edleafe> if the submitter is around, ping first.
16:32:04 <edleafe> otherwise a month is more than enough
16:32:15 <elmiko> yea, don't want to be rude
16:32:22 <ryansb> +1
16:32:47 <cdent> elmiko: speak for yourself, jerk
16:32:52 <elmiko> lol
16:32:56 <edleafe> cdent: haha
16:33:15 * cdent couldn't resist
16:33:33 <elmiko> fair, i nominate cdent as the api-wg rude-boy
16:33:43 <cdent> \o/
16:33:46 <etoews> seconded
16:33:49 <elmiko> haha
16:34:16 <etoews> annegentle was looking for feedback on https://review.openstack.org/#/c/214817/
16:34:53 <elmiko> ooh, good point. meant to get back to that
16:36:38 <etoews> ya. i've got a few comments to make there too.
16:37:01 <etoews> any other guidelines people want to highlight?
16:38:41 <elmiko> nothing from me
16:39:00 <etoews> #topic APIImpact
16:39:10 <etoews> #link https://review.openstack.org/#/q/status:open+AND+(message:ApiImpact+OR+message:APIImpact),n,z
16:39:41 <etoews> anything anyone wants to highlight here?
16:40:08 <cdent> nothing specific but: it's pretty cool the number of them, people have caught on
16:40:28 <elmiko> +1
16:41:21 <etoews> i see the Artifact Repository API specification has rumbled back to life. https://review.openstack.org/#/c/177397/
16:42:13 <cdent> yeah
16:42:24 <cdent> that's a good way to put it
16:45:49 <etoews> i'm looking at https://review.openstack.org/#/c/155931/4/specs/liberty/api-order-ssl-add-reissue-support.rst,unified and found http://specs.openstack.org/openstack/barbican-specs/specs/kilo/certificate-order-api.html that doesn't follow our metadata guideline
16:46:08 <etoews> i expect it's already baked in but i'll comment anyway
16:47:10 <etoews> any other topics people want to discuss?
16:47:27 <elmiko> the barbican folks have been really good about checking api-wg stuff recently. i think the kilo spec may have been before this time
16:48:36 <cdent> I think I mentioned this last time, but wanted to say again: ceilometer is now gatting on a fully http api driven integration job that integrates ceilo, aodh, gnocchi, heat, and nova
16:48:40 <cdent> (using gabbi)
16:48:50 <elmiko> nice
16:48:57 <cdent> It is rather cool because it is so simple but powerful at the same time
16:49:12 <cdent> and api-only seems like a good validation of how it is supposed to work
16:49:25 <cdent> https://github.com/openstack/ceilometer/blob/master/ceilometer/tests/integration/gabbi/gabbits-live/autoscaling.yaml
16:49:52 <cdent> and there are several other tests pending merge which add to that
16:50:01 <cdent> it's easy to improve and all the usual good things
16:50:11 <jaypipes> etoews: ugh, sorry, I will look at pushing a rev on that. sorry for the delay.
16:50:35 <cdent> jaypipes: you should gander on that link I just posted if you have a moment
16:50:36 <etoews> np. i'm probably the worst offender with my errors guideline.
16:53:05 <etoews> cdent: have you had interest from any other projects in using gabbi?
16:53:25 <cdent> besdies elmiko having some thoughts on it, no
16:53:33 <cdent> I think that's partly a matter of PR
16:53:49 <cdent> and all the usual too much to do that everyone experiences
16:53:55 <elmiko> yea, i'm trying to push for gabbi testing in our v2 api
16:54:09 <elmiko> not sure if it will get acceptance from the group though
16:54:34 <cdent> elmiko: if that happens I'll be curious to hear why
16:54:57 <elmiko> i think mainly it could be due to our dependence on many other services. but we'll see
16:55:11 <cdent> it's working so well in the various places I'm aware of it being used that I'd want to persuade
16:55:53 <cdent> note that the link above is run from the command line, as part of a shell script, not a unittest harness
16:56:19 <cdent> which provides some interesting ways to build up other services or gathering auth credentials or whatever before hand
16:56:48 <elmiko> well, when the time comes i will definitely drop you a line
16:57:10 <cdent> cool, thanks
16:57:40 <etoews> cool. if there's nothing else we can call it a couple of minutes early.
16:57:59 <cdent>16:58:10 <etoews> #endmeeting