18:58:20 <davidlenwell> #startmeeting refstack
18:58:21 <openstack> Meeting started Mon Dec 29 18:58:20 2014 UTC and is due to finish in 60 minutes.  The chair is davidlenwell. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:58:23 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:58:25 <openstack> The meeting name has been set to 'refstack'
18:58:49 <davidlenwell> Agenda: api status, qa meta data spec, pending reviews, open discussion
18:59:34 <hogepodge> o/
19:00:51 <davidlenwell> hopefully its not just the two of us.
19:01:35 <hogepodge> It's holiday time. It wouldn't surprise me if everyone is on extended vacation.
19:01:49 <davidlenwell> well who could blame them.
19:02:12 <davidlenwell> we'll give it a few .. then we can call it and defer to next week.
19:03:10 <zehicle> o/
19:07:22 <rocky> o/
19:07:34 <rocky> sorry I'm late
19:07:47 <davidlenwell> happens to the best of us
19:07:54 <sslypushenko__> o/
19:08:04 <davidlenwell> okay .. I think we have enough to start
19:08:06 <vladiskuz_> o/
19:08:10 <rocky> Yeah.  my android chat client lost this channel
19:08:41 <davidlenwell> okay.. So item one on the agenda is api status..
19:08:47 <davidlenwell> #topic api status
19:09:01 <davidlenwell> sslypushenko__:  I know I owe you a patch to the model fixing the table name
19:09:15 <sslypushenko__> Yeap
19:09:27 <davidlenwell> with the holiday that got swept under the rug.. appologies .. I will post it as soon as this meeting is finished
19:09:43 <sslypushenko__> Sounds great
19:10:19 <davidlenwell> Do you have anything else to add as to the status of the api?
19:10:25 <sslypushenko__> Also we should land api spec
19:10:46 <davidlenwell> oh yes.. a trimmed down version of the api spec.. I will also ad that to my list today
19:11:01 <sslypushenko__> It just needs +2
19:11:16 <sslypushenko__> Vlad have already finished api prototype
19:11:41 <davidlenwell> okay
19:11:57 <sslypushenko__> I hope we will get working api before midcyle meetup started
19:12:02 <davidlenwell> I'll also do a review cycle today
19:12:06 <davidlenwell> I think thats a good goal
19:12:16 <sslypushenko__> Me too)
19:12:27 <davidlenwell> attainable too
19:12:50 <zehicle> is that something that could be hosted somewhere?
19:12:53 <davidlenwell> okay.. I'll look throught he pending reviews around the api today
19:12:54 <sslypushenko__> Please don't forget about db schema
19:12:58 <davidlenwell> I won't
19:13:09 <davidlenwell> zehicle: once its landed we can put it in infra
19:13:15 <zehicle> even if not the official site - it would let people test the client
19:13:29 <davidlenwell> nothing stopping you from running it locally
19:13:47 <zehicle> I think we need to get people to try this outside of refstack team,
19:13:57 <zehicle> so that we know what's likely to happen in the broader field
19:14:04 <sslypushenko__> Sure, we are planning test it localy first
19:14:24 <davidlenwell> we do .. I have hp cloud instances I could stand it up on first .. once they are passed the initial internal testing we can have outside folks give it a try
19:14:33 <zehicle> having outside people (vendors?) start trying it will expose issues that we need to address
19:14:43 <davidlenwell> +1
19:14:52 <rocky> zehicle +1
19:15:09 <zehicle> davidlenwell, anyone else able?  I don't want to add to david's plate
19:15:37 <hogepodge> I can talk to the infra team.
19:15:44 <davidlenwell> I think what we should be doing is pushing to infra for it ..
19:15:57 <rocky> I was gonna try it against our stack, with the help of the rest of our team.  Early Jan.
19:16:00 <zehicle> I don't want a small issue to make infra think that we are not ready
19:16:07 <rocky> I'm on vacation at the moment
19:16:25 <davidlenwell> infra is used to incomplete things ... just look at storyboard ;)
19:16:30 <davidlenwell> I think it will be alright
19:16:35 <hogepodge> I can work with them. I'm local to one of them.
19:16:47 <davidlenwell> I think they'd be more suprised if we got it perfect the first time
19:17:06 <hogepodge> It's easy to sit down in person and work through everything. I've been really productive with Clark working that way.
19:17:09 <rocky> +1 for hogepodge  Let's try to get some parallelism going
19:17:20 <davidlenwell> good plan
19:17:33 <davidlenwell> okay .. I think thats covered.. lets move to the next topic..
19:17:45 <davidlenwell> #topic meta data spec (uuids)
19:18:05 <davidlenwell> https://review.openstack.org/#/c/144329/
19:18:09 <davidlenwell> #link https://review.openstack.org/#/c/144329/
19:18:51 <davidlenwell> more details still need to be added about how specifically gate testing should work.. and what its testing for.
19:19:07 <davidlenwell> but otherwise I think its ready for public scrutany
19:19:15 <sslypushenko__> Do we really need add some metadata except uuid?
19:19:23 <davidlenwell> yes
19:19:42 <sslypushenko__> For example?
19:20:01 <davidlenwell> our needs are only for the uuid .. but durring the session at the summit the request from q/a was for general meta data that uuid was a member of
19:20:06 <zehicle> I'd like to see the capabilities managed from the tests
19:20:13 <zehicle> so that the PTLs could own it
19:20:13 <davidlenwell> for example you could tag capabilities
19:20:17 <davidlenwell> yes .. that
19:20:32 <rocky> ++
19:20:32 <sslypushenko__> Right now there is test attr decorator, what is the main difference?
19:20:39 <hogepodge> I'd take the workflow -1 off of it. People tend to not look otherwise.
19:20:41 <zehicle> if that's going to work , we'd need to push the current tags into the tests
19:20:53 <davidlenwell> I wanted it to be clear it was a work in progress
19:20:57 <zehicle> and then deal w/ the fact that we can't have 1 off capabilities, so it needs some managemdent
19:20:57 <davidlenwell> but I can remove it
19:21:20 <zehicle> hogepodge, _1
19:21:21 <zehicle> +1
19:21:27 <hogepodge> The feedback I got from a workflow -1 review what that I'd get no feedback. :-)
19:21:43 <davidlenwell> its gone
19:22:37 <davidlenwell> outside of asking you all to read it and give feed back there isn't much else to discuss at this point..
19:22:59 <davidlenwell> so we can move on to going over pending reviews unless anyone has anything else to add.
19:23:04 <sslypushenko__> Spec looks fine
19:23:13 <rocky> two ways to do capability tags: when the test is written, its capability is part of the review. Or when Capability is determined core, the tests are labelled
19:23:13 <davidlenwell> okay ...
19:23:33 <sslypushenko__> Lets disscuss details in review
19:23:47 <davidlenwell> I think thats more of an implimentation concern more so than a design concern..
19:23:52 <davidlenwell> rocky:  ^
19:24:16 <zehicle> it's OK for new capabilities to emerge
19:24:24 <zehicle> there's a process for making them core
19:24:46 <davidlenwell> #topic pending reviews
19:24:48 <zehicle> we can trust that people will correctly apply capabilities
19:24:58 <davidlenwell> https://review.openstack.org/#/c/143742/
19:25:11 <davidlenwell> zehicle: merg?
19:25:57 <zehicle> as we get new tests, they will either be in new capabilities or use existing ones
19:26:22 <zehicle> the known & core capabilities are available for people to pick if they want
19:26:35 <zehicle> the PTL will be able to govern if there are new capabilities being represented or if they should be cast back into existing ones
19:26:48 <zehicle> a human solvable problem of manageable scale
19:26:50 <davidlenwell> understood
19:27:04 <davidlenwell> so moving on to this review .. https://review.openstack.org/#/c/143742/
19:28:36 <davidlenwell> zehicle:  the agreement for these files was to refrain from merging until someone in defcore approves.. is defcore still just you or is _0x44 also on the commitee?
19:28:59 <zehicle> _0x44 is officially now a co-chair
19:29:17 <davidlenwell> so if you post a thing you will need him to also +1 before I'll merge it
19:29:22 <davidlenwell> and visa versa
19:29:23 <rocky> So, one of you +2 and one approved?
19:29:36 <davidlenwell> no.. they don't have +2
19:29:46 <zehicle> this reflects board action - so, you can take my word on it
19:29:50 <davidlenwell> but I won't plus 2 unless one of them gives it a +1
19:29:59 <davidlenwell> at least that was the earlier agreement..
19:30:00 <davidlenwell> okay
19:30:05 <zehicle> I can ask Chris to +1 it also
19:30:16 <davidlenwell> thats fine.. I was just explaining why it was being ignored
19:31:48 <zehicle> adding 0x44 to the review
19:32:19 <davidlenwell> okay .. when I checked with zehicle last week you had told me to wait on this next one.. https://review.openstack.org/#/c/131187/
19:32:29 <davidlenwell> now I see a +1 from you.. are we ready to merge?
19:32:54 <zehicle> I'd like to see a +1 from catherine
19:33:05 <davidlenwell> okay .. she's out until next week
19:33:15 <davidlenwell> So we can wait if thats okay?
19:33:36 <zehicle> yes, that can wait
19:34:06 <davidlenwell> sslypushenko__:  and I already talked about merging this one.. was originally waiting for catherine.. but does anyone object to merging the api spec? https://review.openstack.org/#/c/141921/
19:34:57 <sslypushenko__> I'm ok with this spec
19:36:31 <hogepodge> Why no get method?
19:36:50 <hogepodge> For first and third party reporting/rendering.
19:36:57 <hogepodge> Also validation from the client side.
19:37:05 <hogepodge> I can add those comments to the review.
19:37:11 <davidlenwell> yes .. please do
19:37:25 <davidlenwell> altho I did just merge it..
19:38:02 <davidlenwell> to answer .. no get method becuase those were not use cases we were trying to adress with the first version.. they can certainly be added and I agree would be useful
19:38:12 <hogepodge> ok
19:38:46 <davidlenwell> I am going to go ahead and merge this one..
19:38:46 <davidlenwell> https://review.openstack.org/#/c/118839/
19:39:16 <davidlenwell> I had already given it a plus two .. but that was before I knew catherine would be on vacation for so long.
19:39:32 <sslypushenko__> In this case may we should drop py33 support?
19:39:52 <davidlenwell> I think you are right about that
19:40:27 <davidlenwell> py33 is already nonvoting .. so its not a blocker at the moment.. but yes .. we should transition to py34
19:42:08 <davidlenwell> sslypushenko__:  I am going to leave some feed back in https://review.openstack.org/#/c/141857/ .. however I think we should do this development work in tempest tools and not in the refstack codebase.. Also I'd like to see it be more generalized for meta data and not just uuid.
19:42:46 <sslypushenko__> It is just a proto
19:42:50 <davidlenwell> understood
19:43:04 <davidlenwell> and its good work too.. I just had some thoughts when reading the code .. I'll leave in review.
19:43:44 <davidlenwell> have any I missed any reviews that are important?
19:44:39 <sslypushenko__> Nothing from my side
19:44:48 <davidlenwell> okay .. lets move into open discussion
19:44:53 <zehicle> how is the client status?
19:44:55 <davidlenwell> #topic open discussion
19:45:00 <zehicle> since we'll need that to test the API
19:45:06 <davidlenwell> thats a good topic for open discussion
19:45:32 <davidlenwell> I believe the client to be in a stable enough place to test with the api.. however I am sure that doing so will expose some things we had not encountered before.
19:45:36 <hogepodge> Also before we go I'd like to get an update from rocky on the mid-cycle details (just location mostly so we can plan travel)
19:45:50 <davidlenwell> okay
19:46:01 <davidlenwell> lets cover this client question.. then we can move onto that
19:46:39 <davidlenwell> zehicle: did that address your concern about the client?
19:46:55 <zehicle> I think so
19:47:09 <zehicle> I wanted to make sure that people outside refstack can try it
19:47:50 <davidlenwell> lets get though an internal test cycle first then we'll announce and get some others to try it out
19:47:57 <davidlenwell> I know bluebox wants to try it as well
19:48:37 <davidlenwell> okay .. rocky.. do you have an address for us?
19:48:39 <rocky> 2330 Central Expressway, Santa Clara, CA  95050  Lots of hotels nearby.  Easy flight into SJC (SW, Alaska, AA).  Probably either need a car, or someone to pick you up/drop you off (pretty easy)
19:48:45 <davidlenwell> I also have questions about parking
19:48:57 <davidlenwell> I can pick up and drop off hogepodge if you need from your hotel
19:49:03 <rocky> Also, should this be announced either on dev or fits?
19:49:05 <davidlenwell> I have a 5 seat car
19:49:14 <davidlenwell> yes rocky it should be announced in fits
19:49:14 <rocky> Lots of parking.
19:49:19 <davidlenwell> okay good
19:49:40 <davidlenwell> lets start a etherpad with the details and attendance
19:49:48 <rocky> I'll send a link or a jpeg for the site.
19:50:03 <davidlenwell> maybe we can list off a few of the closest hotels for those booking travel
19:50:10 <rocky> I'll put that on the etherpad;-)
19:50:18 <rocky> Will do.
19:50:32 <davidlenwell> awesome.. thank you .. please post a link to the etherpad you create in #redstack
19:50:47 <rocky> OK.
19:51:32 <davidlenwell> okay.. anything else to cover?
19:53:42 <davidlenwell> okay .. sslypushenko__ I am going to get you the new model now..
19:53:56 <davidlenwell> thanks everyone.. despite the holiday attendance I think that was pretty productive.
19:53:59 <sslypushenko__> Thx!
19:54:03 <zehicle> +1
19:54:08 <zehicle> c u in 2015
19:54:19 <davidlenwell> #endmeeting