15:01:21 <TravT> #startmeeting openstack search
15:01:23 <openstack> Meeting started Thu Feb 25 15:01:21 2016 UTC and is due to finish in 60 minutes.  The chair is TravT. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:24 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:26 <openstack> The meeting name has been set to 'openstack_search'
15:01:34 <RickA-HP> o/
15:01:36 <yingjun> hi
15:01:38 <TravT> o/
15:01:40 <lakshmiS> o/
15:01:42 <rosmaita> o/
15:01:46 <lei-zh1> o/
15:01:48 <nikhil> o/
15:01:53 <sjmc7> hey
15:01:57 <TravT> how is everybody this fine day?
15:02:02 <rosmaita> brb ... too much coffee during glance meeting
15:02:15 <TravT> and that answers that... :-P
15:02:48 <TravT> ok, agenda here
15:02:50 <TravT> https://etherpad.openstack.org/p/search-team-meeting-agenda
15:03:04 <TravT> #topic Client Release
15:03:18 <TravT> Final release for client libraries: Mar 2 (we will tag no later than March 1 - next Tuesday)
15:03:34 <TravT> There are a couple of minor reviews out.
15:03:39 <TravT> please take a look
15:03:49 <TravT> I've also been wondering if we needed any additional doc for it.
15:04:22 <TravT> https://review.openstack.org/#/q/project:openstack/python-searchlightclient+status:open
15:04:23 <sjmc7> usage type docs?
15:04:26 <TravT> yeah
15:04:58 <sjmc7> i can take a look at what other clients do (and where to put it)
15:05:31 <TravT> steve martinelli put in a request for us to do a few extra things on a patch review i put up awhile back
15:05:32 <TravT> https://review.openstack.org/#/c/270984/
15:05:56 <sjmc7> ah, ok. i can take a look today
15:06:02 <TravT> cool
15:06:20 <TravT> looks like might need to modify the patch i already put up. if so, just go ahead and amend
15:07:06 <sjmc7> will do
15:07:20 <TravT> awesome.
15:07:29 <TravT> i'll look to tag on Tuesday, if not Monday
15:07:31 <TravT> 0.2
15:07:55 <TravT> #topic Searchlight Panel status
15:08:26 <TravT> we presented searchlight concepts with some cool additions on Tuesday at the horizon mid cycle
15:08:29 <TravT> went over very well.
15:08:46 <TravT> we've landed a couple changes already this week that better enable it.
15:08:49 <sjmc7> excellent
15:09:01 <TravT> still have a few dependency patches to go
15:09:09 <sjmc7> you’re good at cracking the whip in person! :)
15:09:21 <TravT> :)
15:09:28 <RickA-HP> TravT: Are there any blockers to completing this work?
15:09:32 <TravT> yesterday and today are basically review-a-thons
15:09:45 <sjmc7> yeah, the midcycles are good for geting crazy amounts of reviews in
15:10:01 <TravT> searchlight enablement is on the list of things (among many) that are on the review priority list.
15:10:12 <TravT> getting the panel itself in will take a FFE on horizon.
15:10:17 <sjmc7> it’s almost better at the end of a cycle; gives some time to avoid the backlog of reviews piling up before release
15:10:18 <TravT> i hadn't formally requested that yet
15:10:23 <TravT> but will probably do that today
15:10:50 <TravT> i have not created the repo (in case it doesn't go in)
15:11:09 <sjmc7> we can do that quickly if need be
15:11:10 <TravT> just been too busy.
15:11:15 <sjmc7> or i can do it
15:11:18 <TravT> yeah, i think we'll know by tomorrow
15:11:22 <sjmc7> i feel like a secretary
15:11:31 <TravT> i'd rather think of you as superman
15:11:38 <TravT> :)
15:11:42 <sjmc7> he was a faster typist than me
15:12:01 <sjmc7> ok, so hold off til the end of the horizon midcycle on that?
15:12:11 <TravT> yeah, can let you know tomorrow morning.
15:12:33 <TravT> there is a swift panel rewrite that will be a heavy focus today.
15:12:49 <TravT> i integrated in the latest swift patches and tested with UI
15:13:06 <TravT> it works, but has caused me to see some shortcomings with navigation
15:14:09 <TravT> but in talking with r1chardj0n3s who is writing it, he definitely wants to add to his dedicated swift panel the ability to do searches with searchlight
15:14:31 <sjmc7> as you know i’ve been talking to the swift team about this too
15:14:31 <TravT> but that would be newton to integrate that
15:14:56 <sjmc7> they seem more open to discussing adding notifications to swift core than was the case in tokyo
15:15:11 <TravT> yes, i mentioned this to him and he said he has a swift core he works with. he said he'd mention it to him as well.
15:15:26 <TravT> i'll try to talk to him more about that
15:15:34 <sjmc7> ok. i’ve spoken to HP’s team so they keep an ear open at the swift meetup next week
15:15:46 <TravT> cool
15:15:51 <sjmc7> on the plus side, the patch i have does seem to work
15:16:10 <TravT> lakshmiS: your patch doesn't seem to actually even try to listen to the notification steve's patch generates
15:16:13 <TravT> right?
15:16:18 <TravT> or did i miss something?
15:16:28 <lakshmiS> yeah its not linked yet. will update the patch today
15:16:47 <sjmc7> ok. let me know if i need to tweak anything with it, it’s not set in stone
15:18:24 <TravT> ok
15:18:51 <TravT> that's all i have for search panel.  the mid cycle has been very productive.  great group of people at it.
15:19:03 <TravT> #topic REVIEWS!!!!
15:19:09 <sjmc7> haha
15:19:19 <TravT> https://review.openstack.org/#/q/project:%255E.*searchlight.*+status:open,n,z
15:19:34 <TravT> Let's just pound these out
15:19:44 <TravT> there are some small easies ones
15:19:50 <rosmaita> ok, which to start wiht?
15:19:57 <RickA-HP> I'm rebasing the Zero Downtime patch and will update it today.
15:20:24 <TravT> This is priority order I propose
15:20:42 <TravT> Can we do a virtual (using hangouts possibly review hackathon on 3/1/2016 - http://everytimezone.com/#2016-3-1,180,cn3)
15:20:42 <TravT> Remove extra infomation from ID field
15:20:42 <TravT> Fix role based indexing for designate
15:20:44 <TravT> Zero Downtime Re-indexing changes.
15:20:46 <TravT> Per resource policy control
15:20:48 <TravT> Neutron
15:20:50 <TravT> Cinder
15:20:52 <TravT> Swift
15:20:54 <TravT> Zaqar notification forwarding
15:20:58 <TravT> First one just because it is small
15:21:00 <TravT> get it out of the way
15:21:21 <TravT> but let's tackle some now
15:21:43 <TravT> and i propose at review-a-thon next Tuesday at the time I listed
15:22:01 <sjmc7> yep, i can do that
15:22:19 <RickA-HP> The review-a-thon works for me.
15:23:14 <rosmaita> i can attend review-a-thon, but probably only intermittently
15:24:46 <TravT> ok, i can do hour later too
15:25:28 <TravT> anyway, i guess just start looking over reviews above and we can chat as needed
15:26:18 <sjmc7> yingjun: you’re right, my suggestion on https://review.openstack.org/#/c/283424 was terrible :)
15:27:33 <yingjun> so i take your second suggestion, ensure _source[‘id’] always there
15:28:08 <sjmc7> yep - given that it should always now be there, could line 117 be shortened?
15:28:24 <sjmc7> i guess it’s a little safer that way if there is no id defined
15:28:39 <sjmc7> either way - this looks good to me, will test it again quickly
15:29:25 <yingjun> it could be
15:29:36 <yingjun> yeah
15:29:57 <sjmc7> not really a big deal
15:31:10 <sjmc7> this works fine; i’m happy to +2 it as it is
15:31:30 <yingjun> ok, thanks
15:32:53 <TravT> i looked at the code this morning as well.
15:33:05 <TravT> was waiting for your input steve
15:33:10 <sjmc7> i’m honoured
15:33:15 <TravT> since you had the concerns.
15:33:53 <sjmc7> yeah, i’m fine with it - it’s the same approach you’re using in the UI?
15:34:04 <sjmc7> i think we can make it part of the contract that there’s always an “id” field
15:34:13 <TravT> well, ui falls back to using the _id
15:34:29 <TravT> but prefers source.id
15:34:49 <TravT> the table row generation needs a unique identifier
15:34:50 <sjmc7> would you be happier doing that in the CLI too?
15:35:29 <TravT> i don't know if CLI needs it for same reason UI needs something
15:35:30 <sjmc7> right now it would make no difference; everything has an id. conceivably we might have something in the future that doesn’t
15:35:36 <sjmc7> no, it’s just for display
15:36:08 <TravT> yeah, see in angular when you ng-repeat, if you have a unique identifier that you can track by, it improves performance when updating search results
15:36:33 <TravT> so, i have to have something in the id field that is unique
15:36:43 <TravT> or else it will just break and leave you angry and depressed.
15:36:54 <TravT> :)
15:37:53 <sjmc7> :)  i think if we can say there will always be a source[“id”] then the cli patch is good
15:38:00 <sjmc7> otherwise maybe we need the fallback logic
15:38:32 <TravT> if it makes sense to add in a safeguard, then we could
15:40:44 <TravT> #chair sjmc7
15:40:45 <openstack> Current chairs: TravT sjmc7
15:40:58 <TravT> sjmc7 i have to leave in a couple minutes
15:41:03 <TravT> i'm the carpool driver to the horizon meeting
15:41:07 <sjmc7> i’m drunk with power
15:41:16 <TravT> use it wisely
15:42:27 <rosmaita> just don't be drunk with driving
15:42:29 <sjmc7> ok. we’ve got 20 minutes of everyone in the same place, so if anyone has questions about any of those reviews this is a great time to ask
15:42:42 <sjmc7> rosmaita: i work from home, i don’t leave the house monday to friday, so no chance of that
15:43:23 <rosmaita> did you ever see the move "Raw Deal"? "You should not drink and bake!"
15:44:02 <sjmc7> hahaha
15:44:11 <sjmc7> no, that cultural touchstone passed me by
15:44:56 <rosmaita> you need to see it, arnold's best movie, lots of casablanca references
15:45:40 <rosmaita> plus, they never explain why this small-town sheriff has an outrageous austrian accent
15:45:43 <sjmc7> i don’t know how i missed it
15:46:03 <rosmaita> ok, so at least i did a movie review
15:49:47 <TravT> thx guys, i'm out of here.  thanks a ton for any reviews you can do!
15:50:01 <rosmaita> TravT: enjoy the midcycle
15:50:02 * TravT off to horizon mid-cycle
15:50:14 <rosmaita> and drive carefully
15:50:31 <TravT> thx
15:52:53 <sjmc7> ok, if nobody’s got anything else they want to discuss i guess we can end here
15:53:10 <rosmaita> ok, see you tuesday, i guess
15:53:14 <RickA-HP> I have nothing else.
15:53:30 <sjmc7> aye. we’ll be in #openstack-searchlight if there’s anything else
15:53:55 <sjmc7> thanks all!
15:54:12 <sjmc7> #endmeeting