02:31:22 <ekcs> #startmeeting congressteammeeting
02:31:23 <openstack> Meeting started Fri Jan 12 02:31:22 2018 UTC and is due to finish in 60 minutes.  The chair is ekcs. Information about MeetBot at http://wiki.debian.org/MeetBot.
02:31:24 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
02:31:26 <openstack> The meeting name has been set to 'congressteammeeting'
02:32:08 <ekcs> hello al! topics are kept here as usual. https://etherpad.openstack.org/p/congress-meeting-topics please read/comment/add =)
02:32:35 <masahito> hi
02:32:40 <ekcs> hi masahito !
02:33:15 <ekcs> how have you been? hope the new year is going well so far!
02:34:05 <masahito> ekcs: Happy New Year! It was nice off. How have you been, too?
02:35:30 <ekcs> i’ve been good for the most part. just getting over a very bad flu though. bad fever for two straight days. finally better now.
02:37:35 <masahito> That sounds bad. I hope you're well now.
02:39:02 <ekcs> haha yes much better. thanks!
02:39:06 <ekcs> well ramineni isn’t here right now maybe we’ll start off with something she and I already talked about.
02:39:16 <ekcs> #topic congress data type
02:39:21 <ekcs> right now congress converts most things to strings
02:39:27 <ekcs> for example, {'id': '12-34', value: 1} gets converted to row ('12-34', '1')
02:39:42 <ekcs> seems like an old issue that never got fixed: https://github.com/openstack/congress/blob/master/congress_tempest_tests/tests/scenario/congress_datasources/test_glancev2.py#L86
02:40:17 <ekcs> I’d like to preserve the data types. should be a fairly simple change in value translator.
02:40:33 <ekcs> it would enable things like p(value=x), p(value=y), builtin:gt(x,y)
02:41:19 <ekcs> but it’s not a backward compatible change. if someone is relying on the conversion to string behavior, things get broken.
02:41:22 <masahito> sounds nice.
02:41:50 <ekcs> so I’m thinking we could change default but add a flag that can be set do preserve old behavior.
02:42:08 <ekcs> any opinions there?
02:42:51 <masahito> What type do you plan to support?
02:44:08 <masahito> What makes it hard to implement is how we store rules in DB.
02:44:44 <ekcs> thinking at least string, int, and float.
02:45:06 <ekcs> good point about DB. I’ll look deeper but I think it’ll be ok because we already support those types in the rule language.
02:45:23 <ekcs> https://github.com/openstack/congress/blob/master/congress/datalog/Congress.g#L183
02:45:48 <ekcs> in DB it would be stored as original rule string which when parsed by congress becomes float/int/string etc.
02:46:40 <masahito> oh, right.
02:47:15 <ekcs> the types are already supported by the rules, just that the datasource translator is not producing the right types.
02:50:16 <masahito> So what we'll do is just keeping data type in datasource translator, isn't it?
02:50:32 <ekcs> that’s what I’m thinking
02:54:46 <ekcs> hi ramineni_ ! we’re just talking about preserving data types some more.
02:55:56 <ramineni_> ekcs: ok
02:58:57 <ekcs> well why don’t we come back to it later.
02:59:01 <ekcs> #topic gate
02:59:23 <ekcs> we are having a gate failure because upping oslo config to 5.2 broke some config validator tests.
03:00:08 <ramineni_> Looks like it's resolved now
03:00:46 <ekcs> I have a patch here that fixes the unit tests and disables temporarily the dsvm tests so that we can get other patches in for queens3
03:00:47 <ekcs> https://review.openstack.org/#/c/532593/
03:00:57 <ekcs> it is? doesn’t look like it.
03:01:56 <ramineni_> Oh
03:02:07 <ekcs> the only thing that passed gate since the break is this: https://review.openstack.org/#/c/532000/
03:02:13 <ekcs> and only because it didn’t run the tests.
03:02:37 <ramineni_> Ohok ..then I have mislooked it
03:02:47 <ekcs> anyway so I’m hoping the patch works and we can merge it asap to unblock the gate so others things and proceed.
03:03:43 <ekcs> any more thoughts/comments about gate?
03:05:31 <ramineni_> We got a fix for skipping tests also
03:05:54 <ekcs> yes that’s great!
03:06:04 <ramineni_> All splitted tempest repos needs to be added .. after that all the tests are enabled
03:06:34 <ramineni_> Aodh works fine now ..I'm in proccess of adding others too
03:07:10 <ramineni_> Zuul looks very slow for past couple of days .. so taking time to get validated
03:07:37 <ekcs> yes great that chandan and gmann helped us figure that out quickly.
03:08:00 <ramineni_> Yes
03:08:47 <ekcs> and thanks for getting after the issue too. just a thought though, if there are queens-3 things you’d like to do, it could make sense to prioritize those. cuz these test fixes technically can wait a week or two.
03:09:00 <ekcs> on that note let’s move on to queens-3
03:09:07 <ekcs> #topic queens-3
03:09:39 <ekcs> queens-3, final client release, and feature freeze is due Thursday of the week of 1/22
03:10:14 <ekcs> let’s think about which final things we want to try and finish.
03:10:56 <ekcs> #link https://etherpad.openstack.org/p/congress-task-priority
03:11:46 <ekcs> I’ve updated the priorities a bit. planning aggressively, I think we may be able to finish everything up to #5.
03:11:54 <ekcs> 1 and 2 are basically done.
03:12:47 <ekcs> if I focus I think I can get 3 done within a week.
03:13:27 <ekcs> I’d like to try to get 5 done too but that requires a bit more exploration and less certain.
03:14:07 <ekcs> any thoughts on what we’d like to add/remove/change in final weeks of queens-3?
03:14:53 <ekcs> monasca webhook would’ve been great but as ramineni_ and I discussed last time it’s not possible now without change on monasca side.
03:17:39 <ramineni_> I can target that for rocky -1
03:20:49 <ekcs> that’d be awesome.
03:21:48 <ekcs> one more thing on queens-3.
03:22:11 <ekcs> ramineni_: did you have more thoughts about the driver config issue?
03:22:30 <ekcs> I think as of our final email we’re pretty close on how to proceed.
03:23:17 <ramineni_> Ya ..I think we agreed on the approach .. But don't remember now ..I'll look into it and start on that
03:23:24 <ekcs> if you have time to knock that out for queens-3 that’d be great. but if not i’ll try to get to it. we can also take this offline if there is more depth to discuss.
03:24:17 <ekcs> ok. great thanks!
03:24:33 <ekcs> anything else on queens-3 or anything else to discuss at all?
03:24:46 <ekcs> last 5 minutes =)
03:29:10 <ekcs> oops looks like I got disconnected.
03:30:03 <ekcs> well let’s end meeting then. looks like we’re on track for a very productive and feature-rich queens release. talk to you guys next week!
03:30:14 <ekcs> #endmeeting