15:01:32 <sjmc7> #startmeeting openstack search
15:01:33 <openstack> Meeting started Thu Jul  6 15:01:32 2017 UTC and is due to finish in 60 minutes.  The chair is sjmc7. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:34 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:37 <openstack> The meeting name has been set to 'openstack_search'
15:02:54 <lei-zh> o/
15:03:17 <sjmc7> morning lei-zh
15:03:24 <sjmc7> give it a minute or two, see if anyone else is around
15:03:30 <lei-zh> morning : )
15:04:06 <lei-zh> are you guys back from holiday?
15:05:35 <sjmc7> yeah. but i’m british, i don’t celebrate independence day :)
15:05:39 <sjmc7> travis is away this week
15:06:22 <lei-zh> lol, that's kind of interesting
15:08:00 <sjmc7> :)  joking
15:08:13 <sjmc7> mostly an excuse to eat too much and watch fireworks
15:09:12 <sjmc7> i guess it’s just us
15:09:15 <lei-zh> road to be a professional photographer : )
15:09:19 <lei-zh> yep
15:09:29 <sjmc7> can maybe take a quick look at the review board
15:10:06 <sjmc7> kevin did the documentation rework that the TC requested from all projects
15:10:19 <sjmc7> to organize documentation better
15:10:31 <sjmc7> https://review.openstack.org/477069 and https://review.openstack.org/480006
15:10:45 <lei-zh> yeah, I'll will have a look at them tomorrow
15:11:01 <sjmc7> they’re pretty straightforward. i checked that they still render
15:11:21 <sjmc7> ah, you put up a zaqar publisher
15:11:27 <lei-zh> that's cool
15:11:39 <lei-zh> yeah, it's not complete yet
15:11:54 <lei-zh> I just upload some work I've been doing
15:12:18 <lei-zh> there's one thing I find a little tedious
15:12:25 <sjmc7> just one? :)
15:12:45 <lei-zh> lol, perhaps for now
15:13:30 <lei-zh> you have to find out projects the publishers are going to notify
15:14:07 <lei-zh> for example, if a nova server is updated, the owner of the instance should be notified
15:14:12 <sjmc7> right
15:14:25 <lei-zh> but things are not all that simple for some resources
15:14:45 <sjmc7> the admin ones? hypervisors etc?
15:15:18 <lei-zh> if it's a neutron network, I have something more to dig
15:15:55 <lei-zh> basically I have to rewrite the logic of every plugin's rbac filters to find affected projects
15:16:21 <sjmc7> ah, yeah
15:16:34 <sjmc7> the weird owner inheritance
15:17:14 <lei-zh> yeah, like neutron, swift, and so on
15:17:43 <sjmc7> swift i wouldn’t bother with for now; it’s not really usable
15:18:33 <sjmc7> yeah, neutron’s a bit tricky. so if you update a port you’d notify the network owner?
15:20:01 <lei-zh> https://github.com/openstack/searchlight/blob/master/searchlight/elasticsearch/plugins/neutron/ports.py#L144
15:20:25 <lei-zh> it seems port is simple because we assume users are not admin
15:20:50 <sjmc7> yeah
15:20:53 <lei-zh> https://github.com/openstack/searchlight/blob/master/searchlight/elasticsearch/plugins/neutron/subnets.py#L140
15:20:58 <sjmc7> but the rule about who to notify would be different i guess
15:21:18 <sjmc7> if i’m logged in and someone adds a port to my network, tell me
15:22:29 <lei-zh> even if that port is not belonged to my project?
15:22:41 <sjmc7> dunno. that’s a tough one i suppose
15:22:46 <sjmc7> maybe just keep it simple for now
15:23:46 <lei-zh> yeah, the way I could figure out is write rules according to rbac filters first
15:24:02 <sjmc7> yeah, see what that looks like
15:24:16 <lei-zh> ok
15:24:45 <lei-zh> sad thing is this leaves two similar rbac logic in the plugin
15:25:04 <sjmc7> yeah.. you could, for now, just use the project id of the thing that changed
15:25:46 <sjmc7> that’ll cover everything except some networking operations
15:26:49 <lei-zh> for some public resources like flavor it may need to ask keystone to get all the projects
15:27:14 <lei-zh> if it's exposed to everyone
15:27:17 <sjmc7> yeah, i’d miss those for now
15:27:26 <sjmc7> the most interesting ones will be servers, volumes, networks
15:27:50 <lei-zh> anyway, I'll try to cover things as much as I can
15:28:34 <sjmc7> ok. only other thing was the nova notifications patch
15:28:50 <sjmc7> i fixed the status thing as well as some other mapping issues - the API fields differ from the notifications
15:28:59 <lei-zh> cool
15:29:25 <lei-zh> I'll look at it along with kevin's documentation patches
15:32:02 <sjmc7> ok. i didn’t have anything else to talk about
15:32:16 <lei-zh> me neither : )
15:32:34 <sjmc7> ok, let’s end it here
15:32:37 <sjmc7> thanks! have a good weekend
15:32:45 <lei-zh> yeah, you too
15:32:46 <sjmc7> #endmeeting