15:01:08 <sjmc7> #startmeeting openstack search
15:01:08 <openstack> Meeting started Thu Mar  2 15:01:08 2017 UTC and is due to finish in 60 minutes.  The chair is sjmc7. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:11 <openstack> The meeting name has been set to 'openstack_search'
15:01:30 <sjmc7> Morning all
15:01:34 <TravT> o/
15:01:38 <Kevin_Zheng_> ：）
15:01:45 <rosmaita> o/
15:01:59 <lei-zh> o/
15:02:11 <sjmc7> I trust everyone is well
15:03:17 <sjmc7> ok, let’s get started
15:03:20 <sjmc7> #topic nova cells
15:03:32 <sjmc7> I saw http://lists.openstack.org/pipermail/openstack-dev/2017-February/112996.html on the mailing list last week
15:03:47 <sjmc7> It looks like nova is still pretty keen to use searchlight
15:03:54 <yingjun> good news
15:04:08 <sjmc7> Kevin_Zheng: i saw you’d replied to that thread
15:04:15 <Kevin_Zheng_> Yeah
15:04:26 <Kevin_Zheng_> I will be happy to do it
15:05:04 <sjmc7> Excellent, thanks
15:05:13 <Kevin_Zheng_> NP
15:05:33 <sjmc7> I suspect that the issues with notifications requiring hits on the nova server API may become apparent
15:05:46 <sjmc7> But maybe that’ll be the impetus needed to make some changes there
15:07:42 <sjmc7> Did you talk to anyone aside from that thread?
15:08:22 <Kevin_Zheng_> Not too much
15:08:24 <Kevin_Zheng_> I
15:08:58 <Kevin_Zheng_> Have a spec about tag notification we talked a little there
15:09:14 <sjmc7> Yes, I remember that one
15:09:29 <Kevin_Zheng_> I will start to check the details next week
15:09:38 <sjmc7> I lost touch with the current state of notifications (versioned or otherwise) so i will try ot catch up
15:09:46 <Kevin_Zheng_> I mean for the whole thing
15:09:54 <sjmc7> Ok, great
15:10:12 <sjmc7> It’ll be good to see how that works out, though I’m also a little nervous :)
15:10:21 <sjmc7> Thanks again for taking that on
15:10:32 <Kevin_Zheng_> I think people are just sum up the ptg
15:11:11 <Kevin_Zheng_> So maybe next week will be better
15:11:11 <Kevin_Zheng_> :)
15:11:29 <sjmc7> Yep :)  They’re always tiring, especially with the travel
15:11:35 <sjmc7> And a lot of people take holidays afterwards
15:11:39 <Kevin_Zheng_> Yeah me too, I will ask for help here if needed
15:11:53 <sjmc7> #topic Other PTG news
15:12:14 <sjmc7> I think only rosmaita was at the PTG from the regular contributors here
15:12:22 <sjmc7> I know you were busy with glance work, which i hope went well
15:12:28 <sjmc7> Or at least reached some conclusions
15:12:33 <rosmaita> possibly
15:13:09 <sjmc7> Did you overhear anything that might be useful for searchlight?
15:13:29 <rosmaita> no, just that nova is still talking about using it for cells
15:13:40 <sjmc7> OK. Yeah, the mailing list summaries are very helpful
15:13:43 <rosmaita> but nothing specific
15:13:58 <sjmc7> How was the PTG?
15:14:27 <rosmaita> the first 2 days were very disorganized
15:14:43 <rosmaita> it was hard to figure out who was discussing what when
15:14:45 <sjmc7> I guess that’s to be expected with a new format
15:14:57 <rosmaita> yes, hopefully they'll address that next time
15:15:08 <rosmaita> i think the idea was that the format will evolve organically
15:15:19 <rosmaita> but, overall it was good
15:15:20 <sjmc7> Oh dear :)
15:15:29 <rosmaita> about 600 people, i think
15:15:34 <sjmc7> Oh, that’s a good turnout
15:15:36 <TravT> wow, that's pretty good
15:15:37 <rosmaita> but not a lot of distractions
15:15:46 <sjmc7> distractions == big drunken parties?
15:15:52 <rosmaita> had a lot more time for discussions than at the typical summit
15:16:05 <rosmaita> sjmc7: there were zero drunken parties
15:16:23 <rosmaita> but i was just thinking of all the random people wandering around
15:16:27 <sjmc7> Yeah, that seemed to be the goal, to allow the dev teams to discuss things without feeling they had to go to talks and do business activities
15:16:43 <sjmc7> Well, you are a popular guy
15:16:43 <rosmaita> yes, and it definitely succeeded in that regard
15:17:04 <rosmaita> ?
15:17:14 <sjmc7> all the people wandering around wanting to talk to you
15:17:39 <sjmc7> Has there been anything going on with the OSIC? I’ve not heard much really since Tokyo
15:18:55 <lei-zh> you mean what osic people are doing?
15:18:58 <sjmc7> yeah
15:19:23 <sjmc7> At one time we were talking about getting in there but lost steam a bit
15:19:41 <rosmaita> yeah, they've pulled back a bit on how many projects they pay attention to
15:20:50 <rosmaita> i think the big think they
15:21:02 <rosmaita> re pushing are the rolling upgrades stuff
15:21:09 <rosmaita> and some operational improvements
15:21:17 <sjmc7> ah, yes. that’s been a problem for many years
15:21:28 <sjmc7> ok. moving on then
15:21:35 <sjmc7> #topic pike goals
15:21:42 <sjmc7> There are two technical committee goals for pike
15:22:16 <sjmc7> One is that all services should run in apache via WSGI
15:22:23 <sjmc7> The other is python 3.5 compliance
15:22:54 <sjmc7> I think we’re ok on python3.5 and there’s a blueprint for WSGI https://blueprints.launchpad.net/searchlight/+spec/deploy-via-wsgi which i will take for pike
15:23:35 <sjmc7> take = tag
15:23:49 <Kevin_Zheng_> cool
15:23:51 <sjmc7> i think it should be fairly trivial (at which point it will become a massive unpleasant burden)
15:24:24 <sjmc7> #topic Reviews
15:24:27 <sjmc7> Last thing i had was https://review.openstack.org/#/c/423884/
15:24:39 <sjmc7> i had meant to merge this for Ocata and i’m a bit annoyed at myself that i missed it
15:25:36 <sjmc7> There was a comment (from lei-zh i think?) about continuing to support elasticsearch 1
15:25:53 <sjmc7> given that ES 5 is now out i’m ok dropping support for 1.x in the interests of simplifying code
15:26:31 <TravT> +1 on dropping ES 5 support
15:26:36 <TravT> i mean ES 1
15:26:38 <sjmc7> hahaha
15:26:46 <TravT> Let's also drop ES 2.x support
15:26:48 <TravT> ;)
15:26:49 <sjmc7> drop them all!
15:26:54 <yingjun> ;;
15:26:58 <lei-zh> yeah, ES1 error string really looks weird
15:27:05 <yingjun> i can't do that
15:27:10 <sjmc7> that’s why they fixed it in ES 2 :)
15:27:12 <clarkb> fwiw es 1 appears to still be supported by upstream. we got a new release not too long ago
15:27:15 <TravT> that bug could be backported
15:27:31 <sjmc7> clarkb: yeah, i think Elastic will support it for some time still
15:27:50 <sjmc7> it was hard to migrate to 2.x for a lot of their users
15:28:10 <sjmc7> we don’t have any deployments i’m aware of using 1.x (our v1 was post-elasticsearch 1)
15:28:29 <sjmc7> i guess it doesn’t do any harm to keep the support if that’s the extent of it
15:28:42 <sjmc7> so maybe it’s ok merging that patch as is
15:28:48 <sjmc7> drop es 1 next release
15:28:59 <sjmc7> But i would like ot get that one in so i can backport it
15:29:39 <sjmc7> And on that note, I will also endeavour to get an elasticsearch 5.x test job running
15:29:43 <sjmc7> Which i seem to say every week
15:30:03 <sjmc7> Ok, I didn’t have anything else on the agenda
15:30:06 <sjmc7> #topic open discussion
15:30:12 <sjmc7> I shall open the floor
15:31:07 <sjmc7> And if nobody stands on it in the next few minutes i’ll call it there and let people get to bed
15:33:12 <lei-zh> nope
15:33:22 <sjmc7> Ok then. Thanks everyone!
15:33:25 <sjmc7> Have a good weekend
15:33:41 <rosmaita> you too
15:33:44 <sjmc7> #endmeeting