15:01:31 <TravT> #startmeeting openstack search
15:01:32 <openstack> Meeting started Thu Aug 18 15:01:31 2016 UTC and is due to finish in 60 minutes.  The chair is TravT. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:33 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:35 <openstack> The meeting name has been set to 'openstack_search'
15:01:42 <TravT> o/
15:01:49 <yingjun> o/
15:01:51 <sjmc7> morning
15:02:01 <TravT> david-lyle: matt-borland saw you wave in the previous meeting
15:02:13 <lei-zh> o/
15:02:38 <matt-borland> lol o/
15:02:53 <TravT> ok
15:02:55 <rosmaita> o/
15:03:13 <TravT> agenda: https://etherpad.openstack.org/p/search-team-meeting-agenda
15:03:19 <david-lyle> o/
15:03:27 * david-lyle re-waves
15:03:39 <matt-borland> o/ o/ o/ o/ o/ o/
15:03:43 <TravT> #topic Newton Milestone 3 has 2 weeks left
15:03:51 <TravT> matt is very excited today
15:03:56 <david-lyle> matt-borland: that's just excessive
15:04:00 * matt-borland coffee
15:04:09 * TravT needs coffee
15:04:30 <TravT> #link https://blueprints.launchpad.net/searchlight/newton
15:04:48 <sjmc7> wow, that came up fast
15:04:54 <TravT> yep
15:05:06 <TravT> anything missing from that list?
15:05:10 <TravT> anything we can drop now?
15:06:04 <TravT> i'm gonna punt 	nova-keypair-plugin
15:06:09 <TravT> no progress made
15:06:12 <sjmc7> the nova-disble-events one is probably as done as it’s getting
15:06:34 <sjmc7> oh, sorry, it’s already marked implemented
15:06:38 * sjmc7 also needs coffee
15:06:57 <TravT> i put a note on nova keypair asking hiroyuki about progress, but not response
15:07:01 <sjmc7> versioned notifications i guess is the big open one
15:07:11 <sjmc7> it doesn’t look like they made as much progress as hoped
15:07:46 <TravT> did you make it to the meeting this week?
15:08:03 * TravT embarrassed that I again did not
15:08:16 <sjmc7> no
15:08:24 <sjmc7> too many things juggling this week
15:09:00 <TravT> Actually, no logs this week from it
15:09:00 <TravT> maybe they went back to every other week
15:09:12 <TravT> and last week didn't have anything really
15:09:24 <TravT> probably need to reach out to gibi
15:10:22 <TravT> lei-zh: how is it going with pipeline?
15:10:34 <lei-zh> I'm add test cases for the code
15:10:56 <lei-zh> Probably I'll rebase on master at weekend
15:11:26 <TravT> ok
15:11:59 <TravT> well, let's move on
15:12:08 <TravT> #topic Elastic Search 1.x not officialy supported
15:12:13 <TravT> we've discussed this before
15:12:25 <TravT> and we spoke of moving devstack to 2.x
15:12:44 <TravT> well, sdague doesn't want elasticsearch in devstack proper repo
15:13:01 <TravT> so sjmc7 implemented ES 2.x install in our searchlight plugin
15:13:24 <TravT> ES 1.x is deprecated and out of support for elastic.co
15:14:01 <TravT> given that apparently there isn't interest in maintaining any common ES support level
15:14:15 <sjmc7> i think they’re still supporting it for a while, but they’re aggressively encouraging people to upgrade
15:14:49 <TravT> Well, I don't particularly think we need to maintain support for 1.x in SL.
15:14:56 <TravT> for the next two releases
15:15:32 <TravT> Maybe just go ahead and put in the docs that our minimum support level is 2.x
15:15:42 <TravT> thoughts?
15:15:43 <sjmc7> yeah, i’m fine with that
15:15:51 <sjmc7> the gate’s still blocked on the es-1 tests
15:16:12 <david-lyle> since we're just going to 1.0 I don't think deprecation is a large concern
15:16:24 <TravT> yes, we need to get https://review.openstack.org/#/c/356757/
15:16:26 <TravT> throught
15:16:28 <TravT> through
15:16:30 <rosmaita> i agree, no need to support 1.x
15:17:35 <TravT> okay.
15:18:19 <TravT> #decision Newton Searchlight will support ElasticSearch 2.x and greater (not ElasticSearch 1.x)
15:18:47 <TravT> #topic Barcelona Summit
15:19:02 <TravT> i just got an email today already asking us to consider our space needs
15:19:14 <rosmaita> a room with windows!~
15:19:15 <sjmc7> i guess that depends how many people are going...
15:19:22 <TravT> right
15:19:36 <sjmc7> :D  it’ll be a broom closet with 95 degree heat
15:19:36 <david-lyle> there will be at least 1
15:19:47 <david-lyle> not sure who
15:20:08 <TravT> We are supposed to submit request by August 31
15:20:22 <TravT> so, please start discussions with your management
15:20:32 <TravT> and we can check back in on this next week
15:20:53 <TravT> maybe david-lyle can proxy for all of us... ;)
15:21:09 <david-lyle> I'm not sure I'm the one
15:21:18 <david-lyle> just 1 of all of searchlight
15:21:37 <david-lyle> I'm a strong tentative
15:21:38 <TravT> we could get somebody to set up a room full of laptops
15:21:46 <TravT> facing eachother
15:21:54 <TravT> and then log us all into a video conference
15:21:57 <TravT> ;)
15:22:01 <sjmc7> i already a told rob i’m gonna find someone to stick a broom with a phone on the end in a roomba
15:22:07 <TravT> nice
15:22:56 <TravT> FYI, this is from the email
15:22:59 <TravT> The main difference with Austin is that the whole Summit event starts on
15:23:00 <TravT> the Tuesday and is only 4 days. Since we didn't want the Ops sessions to
15:23:01 <TravT> overlap with cross-project workshops, we had to squeeze things a little.
15:23:01 <TravT> The plan is to have:
15:23:06 <TravT> 
15:23:06 <TravT> - Ops summit on Tuesday morning until 4pm
15:23:07 <TravT> - Cross-project workshops from Tuesday 4pm to Wednesday 4pm
15:23:09 <TravT> - Fishbowl/Workroom sessions from Wednesday 4pm to Friday noon
15:23:11 <TravT> - Contributors meetups on Friday afternoon
15:23:13 <TravT> 
15:23:15 <TravT> One consequence of the squeeze is that "contributors meetups" will be
15:23:17 <TravT> maximum half-a-day long and will be on Friday *afternoon*.
15:23:44 <sjmc7> that’s a bit sad if folks aren’t staying the extra night
15:24:24 <rosmaita> that's a pretty tight schedule
15:24:30 <TravT> yes, it is
15:24:33 <sjmc7> are the regular talks running tuesday through wednesday?
15:24:35 <rosmaita> guaranteed to have lots of conflicts
15:25:39 <TravT> sjmc7: i don't know... perhaps through thursday
15:25:49 <TravT> #link https://www.openstack.org/summit/barcelona-2016/
15:26:25 <TravT> The Summit will run for four days, Tuesday - Friday, October 25-28, 2016:
15:26:25 <TravT> Keynotes will be held Tuesday and Wednesday
15:26:25 <TravT> Marketplace Expo Hall is open Tuesday - Thursday
15:26:27 <TravT> Main Conference Breakout sessions run Tuesday – Thursday
15:26:29 <TravT> Design Summit developer & operator working sessions run Tuesday - Friday
15:27:27 <TravT> okay, something to think about
15:27:39 <TravT> #topic Searchlight UI Usability Study
15:27:58 <TravT> We have some goals and tasks filled in
15:28:10 <TravT> #link https://etherpad.openstack.org/p/searchlight-ui-usuability-study
15:29:13 <TravT> If anybody has ideas for realistic tasks that we should add, please suggest
15:31:54 <sjmc7> it’s going to be unsupervised, right?
15:31:59 <TravT> yes.
15:32:02 <sjmc7> we can’t have david screaming instrcutions at them
15:32:23 <david-lyle> why not?
15:32:29 <TravT> "no, don't click that!"
15:32:29 <TravT> "I TOLD you NOT to click that!"
15:32:54 <TravT> we will need to massage out the instructions a bit more
15:33:16 <TravT> i had asked piet for additional instructions on what to do next, but I don't think he got back to me
15:33:24 <TravT> i'll ping him again.
15:36:15 <sjmc7> it’ll be good to see some normal people use it
15:37:57 <sjmc7> i think travis may have disappeared
15:38:35 <sjmc7> #topic Call for reviews
15:38:39 <sjmc7> next up!
15:39:28 <sjmc7> two tricky patches are https://review.openstack.org/#/c/351896/ adding more policy support and https://review.openstack.org/#/c/348701/ add neutron’s sharing RBAC support (similar to glance's)
15:40:41 <TravT> sorry i dropped
15:40:43 <TravT> anybody still here?
15:41:00 <rosmaita> hi
15:41:04 <lei-zh> o/
15:41:06 <david-lyle> o/
15:41:07 <rosmaita> sjmc7 was carrying on
15:41:12 <TravT> cool
15:41:14 <sjmc7> like a tortoise
15:41:18 * david-lyle arm's getting tired
15:41:19 <sjmc7> so yes, any reviews on those welcome
15:41:26 <sjmc7> you need to work out more
15:41:34 <sjmc7> the policy one i may split up; it got a bit big
15:41:47 <david-lyle> in a digital society?
15:41:49 <yingjun> o/
15:42:07 <sjmc7> other than those two i don’t think there are many complicated pieces left, so it’ll mainly be bug fixes and testing
15:42:13 <sjmc7> maybe work out with your fingers
15:42:24 <sjmc7> anyone else have any reviews they wanted to discuss?
15:43:08 <lei-zh> not able to do reviews recently, will take a look at them soon
15:43:27 <sjmc7> yeah, i’ve been busy with other stuff as well recently, hoping to free up soon
15:44:09 <TravT> if needed, i think policy stuff should have FFE
15:44:17 <david-lyle> +1
15:44:28 <sjmc7> i’ll try to at least get stage 2 started by n-3
15:44:33 <sjmc7> so it’s less controversial
15:44:45 <TravT> ok, thanks.
15:44:52 <david-lyle> sjmc7: are you splitting up the first one? or should we just roll it as is?
15:45:14 <sjmc7> functionality wise it’s done, so if we can merge it in one go i’m fine with that
15:45:31 <david-lyle> better than reworking it
15:45:38 <TravT> if we don't have to split it out, then i'd rather not.
15:45:43 <sjmc7> ok, i’ll leave it
15:46:01 <sjmc7> it looks bigger than it is because it changes all the plugin files, but they’re just a coupld of properties
15:46:02 <david-lyle> ok, I'll look again today
15:46:19 <sjmc7> the additions i did yesterday were to properly support the admin_or_owner type rules
15:46:19 <TravT> thanks david-lyle
15:46:20 <david-lyle> last I saw there was a jenkins issue
15:46:35 <sjmc7> yeah, i’m trying to get the gozer admins to approve a patch disabling that test
15:46:36 <david-lyle> I think I saw those
15:46:38 <TravT> we need to get that patch through unblocking the gate.
15:46:42 <sjmc7> got broken by a python-elasticsearch change
15:47:09 <david-lyle> how is gozer effecting upstream?
15:47:31 <sjmc7> need approval from the infra team, and several work for us
15:47:52 <david-lyle> ok
15:47:56 <sjmc7> if you have a friendly infra core, the review is https://review.openstack.org/356757
15:48:00 <sjmc7> just makes a job non-voting
15:48:33 <sjmc7> ok, last up since we’re running out of time
15:48:36 <sjmc7> #topic Health Check Middleware
15:48:49 <TravT> i was hoping Kevin_Zheng would be here for that one
15:48:58 <TravT> just to find out if he's intending that for newton still
15:49:23 <TravT> but he's not, i think
15:49:27 <TravT> so, we can move one
15:49:36 <sjmc7> yeah, but midnight meetings are tough. i need to read up on what that middleware’s intended to do
15:50:16 <sjmc7> ok, that was all the topics i think
15:50:44 <TravT> #topic open discussion
15:50:48 <sjmc7> TravT: IRC thinks you’re still chairing, despite my attempts to set topics
15:51:10 <TravT> :-S
15:51:23 <TravT> well, let's close things off here
15:51:28 <TravT> going once
15:51:45 <david-lyle> sjmc7: broadcast plea raised
15:51:46 <TravT> going twice
15:51:52 <rosmaita> one minute
15:52:03 <rosmaita> https://review.openstack.org/#/c/357287/
15:52:27 <rosmaita> it's WIP, but could use a quick lookover if the gate builds the docs
15:52:35 <rosmaita> (it's the api-ref)
15:52:37 <TravT> awesome, thanks rosmaita
15:52:45 <rosmaita> i expect brutal comments
15:52:49 <rosmaita> quick questions
15:52:54 <rosmaita> well, one
15:53:08 <rosmaita> i have only documented 2 api calls
15:53:31 <rosmaita> i think that's all we have: get plugins and do search?
15:53:39 <sjmc7> there’s /facets too
15:54:04 <rosmaita> ok, i need to add that
15:54:47 <TravT> rosmaita are you going to put in more examples from those postman call i sent you?
15:55:37 <rosmaita> TravT: i could ... completely forgot about those
15:55:50 <rosmaita> were those more examples of different searches?
15:56:04 <TravT> i know we can ref people to ES... but you could show example of bool query, etc.
15:56:17 <TravT> i can pick out a few and send to you
15:56:31 <rosmaita> that's some feedback i need, mainly, how to handle that
15:57:01 <rosmaita> because they all come under /v1/search
15:57:21 <rosmaita> but i could do sub-headings so that they're easy to find
15:58:30 <TravT> yeah, probably
15:58:38 <rosmaita> TravT: tell you what, i will add a few as "top level" calls and see what anne thinks of that
15:58:56 <TravT> there are more examples from doc/source/indexapi.rst
15:59:00 <rosmaita> it's a little non-standard, but this *is* openstack after all
15:59:02 <TravT> in tree
15:59:08 <rosmaita> yes, i saw those
15:59:18 <rosmaita> i will grab one or two and update the patch
15:59:38 <TravT> okay.
15:59:43 <TravT> looks like we are about out of time
15:59:50 <TravT> so need to make room for the next party
15:59:58 <TravT> thanks for getting that started
16:00:15 <TravT> thank everybody
16:00:18 <TravT> thanks!
16:00:22 <TravT> #endmeeting