15:03:19 <sjmc7_> #startmeeting openstack search
15:03:20 <openstack> Meeting started Thu Jun 22 15:03:19 2017 UTC and is due to finish in 60 minutes.  The chair is sjmc7_. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:03:21 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:03:23 <openstack> The meeting name has been set to 'openstack_search'
15:03:26 <sjmc7_> hi folks, sorry i’m late
15:03:31 <lei-zh> o/
15:04:06 <lei-zh> are you back from holiday
15:04:14 <sjmc7_> i am
15:04:23 <sjmc7_> back to reality
15:04:34 <lei-zh> welcome to the real world!
15:04:43 <sjmc7_> thanks!
15:05:27 <Kevin_Zheng> o/
15:05:31 <sjmc7_> i merged your uwsgi patch, looked like that was all finished
15:05:51 <lei-zh> yeah, thanks
15:07:10 <sjmc7_> i wondered afterwards whether we need to make any changes to support systemd or whether devstack hides all that from us
15:09:43 <lei-zh> better to have a try, my college sets up devstack via systemd and meet some problems, not know what they are related to
15:09:45 <sjmc7_> it looks like run_process hides all that functionality, which is nice
15:10:34 <sjmc7_> ok, so i think all that work is done
15:10:56 <sjmc7_> i think i’ve also fixed the CI problems with searchlight-ui
15:10:59 <Kevin_Zheng> lei-zh: you mean alex？
15:11:11 <sjmc7_> https://review.openstack.org/471922 passes tests
15:11:18 <lei-zh> Kevin_Zheng, yes
15:11:52 <lei-zh> that's great
15:12:21 <lei-zh> how do you find it out
15:12:32 <sjmc7_> rob pointed out that we had duplicated a file that’s in horizon
15:13:56 <lei-zh> cool
15:14:11 <sjmc7_> so if we can get that merged we can pull in the translations and requirements changes
15:14:32 <sjmc7_> i also am going to check that everything still works because horizon’s undergone a lot of changes since i did any work on it
15:17:21 <sjmc7_> other than that, the only big outstanding patch is the nova notifications one
15:18:21 <lei-zh> will have a look this week, looks very close
15:18:34 <sjmc7_> your comment about the state notifications, lei-zh - they are only ever set on state transition events, but i set them to None for other events so that if the document is used to update it will properly overwrite them
15:20:35 <Kevin_Zheng> Good
15:21:38 <lei-zh> ok
15:21:52 <sjmc7_> i don’t think i had much more on the agenda
15:22:14 <sjmc7_> there may be some potential race conditions with nova - i think i saw yesterday some events appearing out of order because of a small clock skew between nova api and scheduler
15:22:19 <sjmc7_> so i want to try to take a look at that
15:24:27 <sjmc7_> if there’s nothing else i guess i’ll call it here. i urgently need coffee
15:24:43 <lei-zh> one thing about the publisher thing
15:24:57 <sjmc7_> ah yeah, sorry, i meant to ask about that
15:24:57 <lei-zh> have you read the thread recently
15:26:03 <sjmc7_> yeah, i did
15:27:04 <lei-zh> seems first step is that horizon uses the messages to get notified about resource changes
15:27:38 <lei-zh> and they do api calls to get resource info
15:28:04 <sjmc7_> ok
15:28:09 <sjmc7_> the concern was around hiding admin info?
15:28:24 <sjmc7_> i think it would be reasonable to always scrub the data as though the user was not an admin
15:29:39 <lei-zh> yeah, that's what I'm going to do, no admin info exposed
15:30:01 <lei-zh> and we'll come back to it if there's a solution for this case
15:30:57 <sjmc7_> i don’t really think there’d ever be a case to want to send the admin info
15:31:05 <sjmc7_> at most, the UI would present a few fields
15:32:08 <lei-zh> in horizon's case, yes
15:33:01 <lei-zh> I'll update the publisher patch
15:33:32 <lei-zh> zaqar guy asks me if this could be merged in pike 3
15:34:14 <sjmc7_> that seems like a pretty small change
15:34:19 <sjmc7_> (famous last words)
15:35:56 <lei-zh> I don't see there are many things left, but the service queue feature should be merged before
15:36:16 <sjmc7_> service queue?
15:36:50 <lei-zh> yeah, that's a zaqar feature
15:37:09 <sjmc7_> ok
15:37:12 <lei-zh> they have a poc I just tested
15:37:27 <sjmc7_> i’d be fine merging changes to the publisher for p-3
15:38:01 <lei-zh> yeah, I'll update it soon
15:39:07 <sjmc7_> ok. let us know
15:39:28 <lei-zh> sure
15:39:54 <lei-zh> I guess I have nothing else
15:39:58 <sjmc7_> ok. anyone got anything else?
15:40:03 <sjmc7_> quiet today
15:41:07 <sjmc7_> alrighty. thanks everyone! have a good weekend
15:41:46 <sjmc7_> #endmeeting