15:00:49 <witek> #startmeeting monasca
15:00:50 <openstack> Meeting started Wed Nov 28 15:00:49 2018 UTC and is due to finish in 60 minutes.  The chair is witek. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:51 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:53 <openstack> The meeting name has been set to 'monasca'
15:01:07 <witek> Hello everyone
15:01:11 <sc> \o/
15:01:18 <witek> Stefano! :)
15:01:20 <dougsz> hey all
15:01:25 <dougsz> welcome sc!
15:01:29 <sc> I got released from AWS
15:01:35 <witek> haha
15:01:56 <witek> nice to see you again
15:02:30 <witek> let's start
15:02:43 <witek> agenda is as usual here:
15:02:48 <witek> https://etherpad.openstack.org/p/monasca-team-meeting-agenda
15:02:59 <Fdaisuke_> Hi
15:03:08 <witek> hi Fdaisuke_
15:03:17 <witek> #topic CI updates
15:03:30 <witek> two updates
15:03:57 <witek> Adrian has finished the work migrating the CI configuration to native Zuul3
15:04:10 <witek> for tempest tests jobs
15:04:19 <witek> both in monasca-api and monasca-log-api
15:05:01 <witek> it's now more flexible to use with the new Zuul version
15:05:20 <witek> and also `compatible` with the rest of OpenStack CI
15:05:37 <dougsz> nice work
15:06:06 <witek> we're using now `devstack-tempest` parent job which is defined openstack infra
15:06:48 <witek> the second update is about upgrading devstack to Ubuntu Bionic
15:07:28 <witek> infra plans on migrating all the tempest tests to run on Bionic in December
15:08:18 <witek> I have tested devstack locally and after fixing some minor problems everything seems to work fine
15:08:25 <joadavis> the devstack readme page still indicates 16.04 as the most supported version. Hopefully that will update as well
15:08:54 <joadavis> I need to rebuild my devstack anyway, I'll give it a try also
15:08:56 <witek> you mean devstack or our devstack plugin?
15:09:08 <joadavis> the main devstack repo
15:09:15 <witek> I see
15:09:23 <sc> witek: why openjdk-8-jre-headless? what's missing in openjdk-8-jdk?
15:10:08 <witek> for some reason it installed openjdk-11-jre-headless next to 8 in my local env
15:10:24 <witek> 11 being default
15:10:47 <kaiokmo> 11 is default for bionic release, I think
15:11:21 <witek> yes, but I haven't installed it myself, must have been pulled in by some dependency, I guess
15:11:52 <witek> so when I added `openjdk-8-jre-headless` the dependency problem was gone
15:12:32 <sc> witek: I see but still looks strange, did you try openjdk-11-jdk?
15:13:17 <witek> no, I think upgrading to Java 11 is a bigger thing
15:14:09 <sc> Yes, and you don't what to patch java code, aren't you?
15:14:49 <witek> well, no if there's no direct reason
15:15:21 <sc> OK, we'll survive with one more package :-)
15:16:38 <witek> well, I think `openjdk-8-jre-headless` was installed anyway in Xenial already, just not explicitly
15:17:11 <witek> I'll check it once again
15:17:41 <dougsz> FWIW, kolla provides bionic and we're using openjdk-8-jre-headless there: https://github.com/openstack/kolla/blob/master/docker/storm/Dockerfile.j2#L16
15:18:52 <witek> thanks, feel free to comment in review
15:19:17 <witek> #topic notification plugins
15:19:41 <witek> dougsz: that's yours, right?
15:20:10 <dougsz> yeah - so there are a lot of patches sitting around for notification
15:20:25 <dougsz> these ones I think are the highest in the chain
15:21:01 <witek> yes, they're up for quite some while now already
15:21:36 <dougsz> There are some bug fixes which depend on these getting merged.
15:21:56 <dougsz> Around configuration of plugins.
15:22:23 <witek> the only missing puzzle is using Alembic commandline tool in devstack plugin
15:22:45 <witek> https://storyboard.openstack.org/#!/story/2001654
15:22:51 <dougsz> yes, you're right - saw your comment.
15:23:02 <witek> we still use the SQL script there
15:23:12 <dougsz> Unfortunately I'm not a devstack user - but I suppose I could try and take a look sometime.
15:24:41 <dougsz> So I guess we just need reviews on these patches.
15:25:19 <witek> thanks for adding unit tests on fingerprinting
15:25:43 <witek> looks good
15:25:46 <dougsz> np - it was worth it - I found a python 3  bug
15:26:00 <dougsz> Just looking, I think I need to update the commit ID - I can do that later.
15:26:23 <dougsz> But if anyone can see anything else which needs doing please let me know,.
15:27:04 <witek> if you need help with alembic in devstack, please ping in #openstack-monasca
15:27:32 <dougsz> ok, thanks. Do you see that as something blocking the existing patches?
15:29:00 <witek> it's not really tested in tempest tests, so I guess we can add alembic later
15:29:13 <witek> but we should definitely do this
15:29:29 <dougsz> yeah - I've switched over in Kolla and it's a really nice feature
15:30:16 <witek> any other comments?
15:30:53 <witek> #topic specs
15:31:07 <witek> https://review.openstack.org/583803
15:31:44 <witek> dougsz has posted some questions in review
15:31:53 <witek> and I added my comments
15:32:00 <joadavis> yes, thanks for the comments. I haven't tried to incorporate them yet
15:32:17 <joadavis> The point about validation is a good one
15:32:48 <joadavis> but I think the efficiency of going straight to kafka is why we wanted to shortcut the events api
15:32:51 <witek> agree, it would be easier if all the requests go through the single API
15:33:05 <joadavis> But we can use the same validation code in the Events Listener when it is implemented
15:34:00 <witek> dougsz: what do you think?
15:34:47 <dougsz> Just wondering how this fits in with merging APIs
15:35:04 <dougsz> The validation code would live in monasca-common?
15:35:11 <joadavis> already does
15:35:17 <witek> no, monasca-api
15:35:28 <witek> I mean, after merge in monasca-api
15:35:45 <witek> and event-listener also
15:36:13 <dougsz> Ah I see.
15:36:35 <joadavis> yes, sorry, I was starting to merge things in my head already
15:37:00 <dougsz> So the event listener would run as a separate service the the Monasca API?
15:37:08 <dougsz> *to the
15:37:24 <witek> yes, but reusing the code
15:39:42 <dougsz> Just wondering if it should have it's own repo but import the validation code from the Monasca API
15:41:23 <witek> won't we end up with the same issues we have with monasca-common?
15:42:24 <dougsz> Yeah - that is a disadvantage.
15:43:03 <witek> OK, let's make another round on that spec this week
15:43:28 <witek> is that OK
15:43:29 <witek> ?
15:43:33 <joadavis> I'll take some time this afternoon and incorporate revisions
15:43:33 <dougsz> Sounds good to me
15:43:49 <witek> thanks joadavis and dougsz !
15:44:11 <dougsz> thanks to you both, it'll be a very nice feature
15:44:35 <witek> I think so too
15:44:45 <witek> https://review.openstack.org/609055
15:45:34 <witek> I have moved the spec to a different dictionary and made small updates
15:45:52 <witek> I think it's in a good shape already
15:46:49 <dougsz> thanks for tweaking it
15:47:26 <witek> that's all from my side
15:47:36 <witek> do you have anything else?
15:47:51 <dougsz> Not from me
15:47:57 <joadavis> nor from me
15:48:43 <witek> Fdaisuke_: how is the progress on monasca-analytics?
15:49:44 <dougsz> dougsz: Feeling sorry for Fdaisuke_ that it's gone midnight
15:50:10 <witek> yes, not a good meeting time for Japan
15:50:45 <witek> anyway, please keep us updated if you have any news or problems
15:50:59 <witek> thanks everyone for today
15:51:04 <witek> and see you next week
15:51:05 <dougsz> thanks all
15:51:48 <witek> #endmeeting