15:01:02 <sjmc7> #startmeeting openstack search
15:01:03 <openstack> Meeting started Thu May 26 15:01:02 2016 UTC and is due to finish in 60 minutes.  The chair is sjmc7. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:06 <openstack> The meeting name has been set to 'openstack_search'
15:01:14 <sjmc7> morning all
15:01:43 <lei-zh> o/
15:02:09 <sjmc7> #link https://etherpad.openstack.org/p/search-team-meeting-agenda
15:02:35 <rosmaita> sjmc7: i am on vacation this week, will not be attending
15:02:42 <sjmc7> but yet..!
15:02:57 <sjmc7> thanks for letting me know, have a nice break!
15:03:14 <sjmc7> travis is also out, so i’ve got the wheel this week
15:03:18 <sjmc7> we’ll give it a few minutes
15:03:36 <yingjun> :)
15:04:15 <sjmc7> ok, will get started. others may drift in
15:04:27 <sjmc7> #topic newton-1 milestone
15:04:43 <sjmc7> Sometime next week is the newton-1 milestone
15:05:05 <sjmc7> i think our aim is to have any blueprints/specs that we expect to get implemented approved by then or soon after
15:05:37 <sjmc7> so if anyone’s got anything they would like to see, it’d be good to get something written up in the next week
15:05:57 <sjmc7> lei-zh: the pipeline one is the only big architectural one taht came to mind
15:06:13 <lei-zh> I'll update pipeline spec soon
15:06:21 <lei-zh> and we can have further discussion
15:06:35 <RickA-HP> Are there any others that folks are thinking about? Or need help with?
15:06:38 <sjmc7> ok. yeah, hopefully we can get that moving
15:07:23 <sjmc7> the other high priority BPs we have are mostly plugins, but we’ll review next week
15:07:53 <sjmc7> ok, moving on
15:07:58 <sjmc7> #topic Request for mentoring
15:08:15 <sjmc7> there was some traffic on the mailing list over the last week
15:08:17 <sjmc7> #link http://lists.openstack.org/pipermail/openstack-dev/2016-May/095921.html
15:08:56 <sjmc7> we do already have some contributors that we’ve been semi-officially mentoring, but if anyone would like to officially put themself forward for it, that’s the information
15:09:37 <sjmc7> as i understand it it’s intended to be a few hours a month of answering questions and being a contact for the person in the program; it’s not meant to be a full-time internship kind of thing
15:09:52 <Kevin_Zheng> o/
15:10:03 <sjmc7> ah, hi Kevin_Zheng
15:10:15 <Kevin_Zheng> Sorry
15:10:20 <Kevin_Zheng> To be late
15:10:40 <sjmc7> ah, no worries. i think most of the people here are in china so thanks for attending at a late hour
15:10:52 <sjmc7> #topic Lei Zhang is now core
15:11:00 <sjmc7> thanks lei-zh !
15:11:17 <lei-zh> it's my honor : )
15:11:17 <RickA-HP> Congrats!
15:11:33 <yingjun> Congrats!
15:11:33 <sjmc7> you and yingjun don’t have to wait til the next day for +2s now :)
15:11:38 <Kevin_Zheng> Cong
15:11:49 <lei-zh> haha, indeed
15:12:44 <sjmc7> #topic Nova notification updates
15:12:48 <sjmc7> couple of things here
15:13:02 <sjmc7> first, I put a review up that reduces the number of calls to nova’s API from about 9 to 3 during boot
15:13:07 <sjmc7> https://review.openstack.org/#/c/317100/
15:13:19 <sjmc7> it’s a bit complicated because of some race conditions, so if people can test it that would be great
15:14:34 <sjmc7> the other is that we’re still tracking nova’s move to versioned notifications
15:14:59 <sjmc7> that spec has been approved (https://review.openstack.org/#/c/286675/) so we’ll keep an eye on it
15:15:19 <sjmc7> it looks like initially they’ll reimplment the current notifications as versioned ones, and then consider additions/changes
15:15:43 <sjmc7> anyone else have any information or questions about that?
15:15:44 <Kevin_Zheng> Sure
15:16:15 <Kevin_Zheng> I'm working with them on that in nova
15:16:15 <sjmc7> Kevin_Zheng actually works on nova so has real information on it :)
15:16:42 <sjmc7> one thing that travis and i have requested once that initial work is done is making the notification payload look more like the API response since that helps us a lot with indexing
15:16:58 <sjmc7> the nova team doesn’t seem that keen on it
15:17:07 <Kevin_Zheng> Yes, have just started
15:17:08 <sjmc7> but we’ll wait until the first implementation is done
15:17:18 <yingjun_> FYI, i uploaded the flavour notification spec today, https://review.openstack.org/#/c/321336/
15:17:37 <Kevin_Zheng> The first one as example, so long way to go
15:17:57 <sjmc7> ok. we’ll keep an eye on it
15:18:03 <sjmc7> ah, excellent yingjun_. GB21 put up a patch for indexing flavors without notifications that i think you’ve been reviewing
15:18:06 <Kevin_Zheng> Sure
15:18:23 <yingjun_> yeah
15:18:45 <yingjun_> and the hypervisor one needs one more +2 https://review.openstack.org/299807
15:19:28 <sjmc7> ok. i’ll ask around and see if i can find anyone available to look at it
15:19:45 <sjmc7> it’s always a long process getting specs reviewed for the established projects :)
15:19:52 <lei-zh> I can find some nova people to look at that spec
15:19:59 <sjmc7> that would be great too
15:20:02 <yingjun_> great!
15:20:40 <Kevin_Zheng> johnthetubaguy will be a good choice
15:21:29 <sjmc7> ok
15:21:40 <sjmc7> next up, and quite related...
15:21:42 <sjmc7> #topic Nova microversioning
15:21:53 <sjmc7> this is the reason Kevin_Zheng is here
15:22:28 <sjmc7> correct me if i’m wrong - it sounds like nova’s removing the non-microversioned API at some point
15:22:47 <sjmc7> so we need to support it sooner rather than later. i see you put a patch up today which i’ll take a look at later
15:23:18 <sjmc7> this’ll also let us support some additional fields from later versions of the API
15:23:56 <sjmc7> Kevin_Zheng: do you have a link for the nova work on that?
15:25:04 <Kevin_Zheng> I have
15:25:23 <Kevin_Zheng> But I'm now using mobile device
15:25:31 <sjmc7> ah :)
15:25:47 <sjmc7> then no worries - if you could drop it in the searchlight irc room tomorrow that would be really useful
15:25:52 <yingjun_> https://review.openstack.org/#/c/321375/1/specs/newton/support-nova-microversions.rst
15:26:01 <Kevin_Zheng> And it seems I forgot to include the links
15:26:12 <Kevin_Zheng> I will update the spec
15:26:19 <sjmc7> ok, that’d be good as well, thanks :)
15:26:20 <yingjun_> is this one right? Kevin_Zheng
15:26:21 <Kevin_Zheng> In a minute
15:26:26 <Kevin_Zheng> Yes
15:26:30 <sjmc7> it can wait til tomorrow, i know it’s late there
15:26:55 <Kevin_Zheng> No warriors
15:27:00 <Kevin_Zheng> Worries
15:27:04 <sjmc7> thanks for jumping in on this - it’s hard for us to keep up with changes like this across so many projects
15:27:27 <sjmc7> ok, moving on unless anyone has any more comments on that...
15:27:35 <Kevin_Zheng> I was planning to add it but I was distracted by something
15:27:41 <sjmc7> i know how that feels :)
15:28:07 <sjmc7> #topic Open reviews
15:28:14 <sjmc7> #link https://review.openstack.org/#/q/status:open+AND+(project:openstack/searchlight+OR+project:openstack/searchlight-specs+OR+project:openstack/python-searchlightclient+OR+openstack/searchlight-ui)
15:28:31 <sjmc7> if anyone’s got any experience with angular, reviews for the UI patches would be good
15:28:56 <sjmc7> we do have some horizon cores that spend a little time on them, and i know yingjun_ has put some patches up
15:29:09 <sjmc7> it’s not really my area of expertise
15:29:59 <sjmc7> so i guess i should get better :)  thanks to everyone for being on top of reviews though; we’ve done a good job of not letting things sit for too long
15:30:53 <sjmc7> ok.. that was all i had on the agenda
15:30:56 <sjmc7> #topic Open discussion
15:31:05 <sjmc7> if there’s anything else people want to discuss, feel free to bring it up
15:31:29 <sjmc7> otherwise i’ll give it a few minutes and let everyone get to bed. meetings are quick when people are away on vacation
15:31:51 <david-lyle> sjmc7: or hiding in plain sight
15:31:56 <sjmc7> :O
15:32:22 <sjmc7> you’re just like my driving instructor; silent for 30 minutes then scaring me at the end
15:32:35 <david-lyle> that was all wrong
15:32:40 <david-lyle> ;)
15:32:53 <sjmc7> :)
15:33:06 <sjmc7> i can already feel travis reading the transcript and shaking his head
15:33:25 <david-lyle> in your test environments, are you doing any rate limiting on nova API?
15:33:45 <sjmc7> no, because i think we’d hit it :)  that’s why we’re desperately trying to reduce the number of calls we make
15:33:51 <david-lyle> right
15:33:59 <david-lyle> the move from 9 -> 3 is great
15:34:04 <sjmc7> it’s still too many
15:34:08 <david-lyle> but 3  is still a lot
15:34:11 <david-lyle> yes
15:34:13 <sjmc7> yeah. we could remove another one
15:34:19 <sjmc7> the pre-scheduling one
15:34:39 <sjmc7> or do a partial document for that first one
15:34:44 <sjmc7> maybe i’ll do that as well
15:34:59 <sjmc7> i’ve always said though that we need to aim for no API calls at all
15:35:06 <david-lyle> I'm glad progress is being made
15:35:31 <sjmc7> yeah. it’s definitely a very valid concern
15:35:43 <david-lyle> that certainly should be the goal and if nova is going to point people to searchlight, seems like they should help with the notifications to make sure it works
15:35:57 <david-lyle> not have two hands working in opposite directions
15:36:07 <sjmc7> yeah, we’re hoping that is the case, although i’m not sure the two groups are working closely
15:36:25 <david-lyle> well not to be solved here
15:36:37 <sjmc7> no. we’ll keep working on them
15:36:43 <sjmc7> ok, unless anyone had anything else, i’ll call it
15:36:50 * david-lyle is done
15:36:57 <sjmc7> thanks everyone, especially where it’s 11pm!
15:37:28 <david-lyle> congrats lei-zh
15:37:28 <sjmc7> #endmeeting