15:00:42 <TravT> #startmeeting openstack search
15:00:44 <openstack> Meeting started Thu Feb  4 15:00:42 2016 UTC and is due to finish in 60 minutes.  The chair is TravT. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:45 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:47 <openstack> The meeting name has been set to 'openstack_search'
15:01:21 <TravT> o/
15:01:26 <RickA-HP> o/
15:01:41 <TravT> i saw sjmc7 and yingjun, i think.
15:02:07 <yingjun> Hi
15:02:13 <TravT> hello
15:02:19 <sjmc7> hi
15:02:40 <yingjun> I'm using my phone, slow typing
15:02:46 <TravT> well, i don't see lakshmiS or david-lyle online.
15:03:00 <TravT> i think we have kind of a light agenda today.
15:03:07 <RickA-HP> Lakshmi was having issues with chat yesterday.
15:03:23 <TravT> agenda: https://etherpad.openstack.org/p/search-team-meeting-agenda
15:03:31 <TravT> add anything there you want.
15:03:46 <TravT> #Cross Project Specs
15:03:52 <TravT> #topic Cross Project Specs
15:04:01 <TravT> Just FYI on a couple that will affect us
15:04:18 <TravT> The policy and roles one     https://review.openstack.org/#/c/245629/
15:04:44 <TravT> I've had a blueprint open for awhile and have added a link to this into that blueprint.
15:05:06 <sjmc7> that one’s for the N cycle?
15:05:22 <TravT> https://blueprints.launchpad.net/searchlight/+spec/per-resource-type-policy-control
15:05:49 <TravT> sjmc7: the cross project spec is for the N cycle
15:06:05 <sjmc7> you and i ahve spoken about this a bit. i think the rbac change that’s under review now can be expanded to cope with that
15:06:17 <TravT> agreed on that.
15:06:18 <sjmc7> since i didn’t go with a boolean flag
15:06:29 <sjmc7> i think i’d like to wait and see how it pans out
15:06:36 <sjmc7> how many priojects actually adopt it
15:06:51 <TravT> i think we still might need to do a "light" version of that spec for us sooner than later.
15:07:26 <TravT> with the role separation for field work by you there is a clumsy workaround
15:07:27 <sjmc7> yeah, let’s keep an eye on it
15:07:37 <sjmc7> all my workarounds are clumsy!
15:07:39 <TravT> set * for admin_only_fields
15:08:36 <TravT> but if, for example, yingjun gets hypervisor plugin in, then the UI and CLI will show non admins that they can search on hypervisors, but just won't get results, I believe
15:08:44 <sjmc7> ah, yeah. there’s an optimization we could do to not index docs under that circumstance at al
15:08:54 <sjmc7> yeah, that’s a stretch goal for this cycle :)
15:09:23 <TravT> yeah, so, doing something there might be something that becomes freature freeze worthy.
15:10:05 <TravT> but adopting that other cross project spec with all those roles is very unlikely to happen for much of openstack in M
15:10:47 <TravT> There is also one on client backwards compatibility
15:10:49 <TravT> https://review.openstack.org/226157
15:10:57 <sjmc7> i’m sure there’ll be a lot of discussion at the summit
15:11:53 <RickA-HP> TravT: Sorry, back to the Policy Spec.
15:12:13 <RickA-HP> WIll we need to make changes to our user role code to handle the different observers?
15:12:28 <TravT> yeah, basically, this is a big reminder to us that if we go 1.0 (like I'd love to do) at the end of mitaka that we'll have to really start dealing with bakcwards compatibility
15:12:57 <RickA-HP> Something we should be thinking about before a 1.0 release!
15:13:24 <TravT> RickA-HP: I'm not sure.  sjmc7 and I talked through it in IRC the other day
15:13:44 <TravT> and basically determined that we need to put more thought in on it
15:13:55 <TravT> there is a good chance there will be an impact.
15:14:15 <TravT> but we're hoping it can be an iteration on top of current progress
15:14:16 <RickA-HP> Ok. I agree about doing nothing for M and allow it to settle.
15:16:10 <TravT> #topic Virtual Sprints
15:17:30 <nikhil> o/
15:17:33 <TravT> yingjun: rosmaita:   the other day sjmc7, RickA-HP, lakshmiS and I were talking about having an hour hangout on Tuesdays to just talk and code share work
15:17:42 <TravT> through end of mitaka.
15:17:45 <TravT> to help us finish out.
15:17:54 <TravT> o/ nikhil
15:18:17 <yingjun> Okay
15:18:23 <nikhil> I saw the virtual sprints and went hooray
15:18:26 <TravT> what city are you in?
15:19:25 <nikhil> The Openstack bug smash/hackathon days has been sorta finalized for Mar 7-9. and if people are interested in joining (remotely or otherwise) I think we can use a day there for this purpose.
15:20:10 <nikhil> we are likely going to have fresh developers though, but simple bugs can be taken by them with us focusing on reviews
15:21:06 <TravT> yingjun: i'm not sure your time, but I was wondering if something like UTC: 23:00 or 24:00 would work better for you on Tuesdays.
15:21:29 <TravT> rather than late you time
15:21:36 <TravT> e.g. http://www.timeanddate.com/worldclock/meetingdetails.html?year=2016&month=2&day=9&hour=23&min=0&sec=0&p1=1947&p2=33
15:22:00 <TravT> or http://www.timeanddate.com/worldclock/meetingdetails.html?year=2016&month=2&day=10&hour=0&min=0&sec=0&p1=1947&p2=33
15:22:06 <yingjun> UTC 24 is better for me
15:22:12 <sjmc7> either’d be fine for me
15:22:59 <RickA-HP> If UTC 24 is better for yingjun, that's fine with me.
15:23:03 <TravT> so, http://www.timeanddate.com/worldclock/meetingdetails.html?year=2016&month=2&day=10&hour=0&min=0&sec=0&p1=1947&p2=33
15:23:03 <TravT> ?
15:23:16 <sjmc7> yep, that works
15:23:36 <TravT> ok, i'll send out an invite for next week and see how it goes.
15:23:47 <TravT> nikhil: re: hacakathon
15:24:06 <TravT> how does that work?
15:24:18 <TravT> is is just people agree to work on bugs that day on various projects?
15:25:19 <yingjun> But not sure i can get time next week, it is spring festival in China here
15:25:25 <TravT> oh, ok
15:25:37 <TravT> yingjun: don't ruin your holiday for us
15:25:47 <TravT> he can try after next week
15:26:38 <TravT> ok, well, let's move on.
15:26:55 <TravT> #topic reviews
15:27:08 <TravT> as usual, let's keep working on reviews.
15:27:35 <TravT> Separate documents by role is top priority
15:27:37 <TravT> https://review.openstack.org/#/c/257516/
15:27:40 <nikhil> oops, sorry I got pulled away at the last min. sorry about that. I will ping again in open discussion.
15:27:45 <TravT> ok
15:28:12 <TravT> sjmc7: jenkins +1'd and my devstack is restacked, I will do a final lookover after this meeting.
15:28:23 <sjmc7> yaaay
15:28:25 <TravT> probably good to go for now
15:29:11 <TravT> I've just copy pasted the neutron review into the agenda for the last few weeks.
15:29:20 <TravT> i haven't had time to look at the data mapping yet
15:29:35 <sjmc7> i think i have a good idea on that
15:29:51 <sjmc7> been holding off til some of the reviews go through to avoid spending all mytime rebasing
15:30:07 <TravT> yeah.
15:30:12 <TravT> #link https://review.openstack.org/#/q/project:%255E.*searchlight.*+status:open,n,z
15:30:28 <TravT> just a reminder of all the reviews
15:30:53 <TravT> any specific questions anybody has on their reviews?
15:31:10 <RickA-HP> Which ones are higher priority?
15:31:47 <sjmc7> RickA-HP: some of them are really small
15:31:51 <sjmc7> and some are in dependency chains
15:32:57 <sjmc7> the bigger ones are obviously a bit more time consuming but are needed for other work to land, so the ‘documents by role’ and ‘index child plugins as part of parent’ are ones i’d really like landed
15:33:08 <TravT> from pure priority, the most important I think are Separate documents by role & Index child plugins as part of parent
15:33:47 <TravT> sjmc7: Index child plugins as part of parent conflicts with separate documents by role
15:34:14 <TravT> so, i didn't want to look at it
15:34:16 <sjmc7> yeah, i know
15:34:23 <sjmc7> i’ll rebase it again
15:34:31 <TravT> k
15:34:36 <RickA-HP> Ok, I'll take a look at Index child plugins. I'm getting tired of Separate Documents :)
15:35:06 <TravT> enough bleeding eyes on that one, RickA-HP?  ;)
15:35:44 <TravT> okay, let's move on to open discussion.
15:35:48 <TravT> #open discussion
15:35:54 <TravT> #topic open discussion
15:36:40 <TravT> nikhil?
15:36:53 <rosmaita> (we are in glance midcycle mtg)
15:37:16 <TravT> rosmaita: ahh, thanks!
15:37:31 <rosmaita> (nikhil is speaking about glanceclient atm)
15:37:37 <nikhil> thanks rosmaita
15:37:43 <nikhil> I will multi task
15:37:53 <TravT> nikhil, we can discuss hackathon next week
15:37:58 <TravT> since it is 3 weeks out
15:38:01 <TravT> 4 weeks
15:38:01 <nikhil> So, my idea is to have one day dedicated to glance folks and one to searchlight during the hackathon
15:38:12 <nikhil> just wanted to see if someone can fly in there
15:38:18 <TravT> what does that mean?
15:38:21 <nikhil> I need a repr from horizon if possible
15:38:28 <nikhil> someone be there in NYC
15:38:32 <nikhil> for the hackathon
15:38:40 <TravT> is there a link on this?
15:38:54 <nikhil> sure
15:39:06 <rosmaita> nikhil: what are the remote attendance possibilities?
15:39:25 <nikhil> https://etherpad.openstack.org/p/OpenStack-Bug-Smash-Mitaka
15:39:36 <nikhil> I have been promised a projector
15:39:49 <nikhil> and right now am thinking a uberconference bridge with hangouts
15:40:20 <nikhil> tl;dr; the sponsors are positive on remote conn
15:40:44 <nikhil> TravT: sorry, I am a bit raw in my approach here.
15:41:00 <nikhil> TravT: so, I wanted to bring these 2 things up explicitly
15:41:23 <nikhil> one is increasing SL interest and use a day during hackathon for bug smashing
15:42:11 <TravT> well, i certainly think it could help
15:42:15 <nikhil> second is get help from someone here to get in touch with horizon core who may be able to come to NYC chapter of hackathon to give a presentation or two and share some knowledge with new devs
15:42:36 <TravT> well, i guess i'm a horizon core...
15:42:52 <TravT> yesterday was the horizon meeting.
15:43:01 <TravT> but i could bring up in the next one.
15:43:11 <nikhil> Thanks TravT!
15:43:18 <TravT> does it have to be NYC?
15:43:25 <TravT> I see a bay area one
15:43:29 <nikhil> if someone can be present in New York on Mar 7-9 that would be excellent!
15:43:34 <TravT> and there are a couple horizon cores in the bay area
15:43:54 <nikhil> both would be great, I am focusing on NYC. there is some interest in learning django stuff
15:44:17 <TravT> if django is the focus, it definitely impacts who i'd try to get to go
15:44:27 <nikhil> ah
15:44:39 <nikhil> I guess any kind of application development will help
15:44:44 <TravT> ok.
15:44:57 <nikhil> I think we may have a lot more of control plane folks there
15:45:15 <nikhil> data plane folks may be less and having some will def help
15:46:01 <nikhil> and I am thinking of metadata control plane here which may be contradictory to binary data (confusing)
15:46:17 <nikhil> Thanks.
15:47:04 <TravT> nikhil: thanks for bringing this up
15:47:05 <TravT> !
15:47:06 <TravT> wi
15:47:13 <nikhil> my pleasure
15:47:31 <TravT> we'll have to consider it, but we definitely will want to do something.
15:47:48 <TravT> Any other topics today?
15:48:09 <TravT> yingjun: did you have any questions regarding the comments i left on your CLI patch?
15:49:51 <yingjun> I didn't take a close look at that yet, but I'm ok to add --json
15:50:41 <TravT> ok
15:50:45 <TravT> well thanks everybody
15:50:53 <TravT> rosmaita: nikhil: have fun in the glance mid-cycle!
15:51:01 <rosmaita> TravT: ty
15:51:03 <nikhil> Thanks TravT
15:51:34 <TravT> o/
15:51:37 <TravT> #endmeeting