15:01:05 <witek> #startmeeting monasca
15:01:06 <openstack> Meeting started Wed May 29 15:01:05 2019 UTC and is due to finish in 60 minutes.  The chair is witek. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:07 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:10 <openstack> The meeting name has been set to 'monasca'
15:01:18 <witek> hello everyone
15:01:54 <hosanai> hi
15:02:03 <dougsz> o/
15:02:17 <witek> hi hosanai and dougsz
15:02:33 <witek> agenda for today:
15:02:37 <witek> https://etherpad.openstack.org/p/monasca-team-meeting-agenda
15:02:47 <akhil_jain> Hello all!
15:02:56 <witek> hi akhil_jain: nice to see you
15:03:25 <witek> #topic notification reviews
15:04:15 <witek> dougsz: do you want to take over?
15:04:29 <dougsz> thanks witek
15:05:09 <dougsz> So here a bunch of patches to the notification services
15:05:26 <sc> yo
15:05:26 <dougsz> Most importantly, there are some bug fixes which affect various stable releases
15:06:26 <dougsz> As witek has pointed out we are still using deprecated yaml config in devstack for the notification service which makes it difficult for people to test
15:07:16 <dougsz> I had a look at removing the yaml config, but it turned into quite a large task as many of the tests use it
15:07:42 <witek> isn't https://review.opendev.org/633194 covering that?
15:08:39 <dougsz> Ah yes, I need to return to that
15:09:34 <dougsz> So the most important ones for me are:
15:09:35 <dougsz> https://review.opendev.org/#/c/656783/
15:09:54 <dougsz> and https://review.opendev.org/#/c/656780/
15:10:02 <dougsz> Which are affecting the stable rocky release
15:11:03 <witek> I think we need bandit fix cherry-picked to rocky
15:11:21 <witek> https://review.opendev.org/661953
15:11:43 <dougsz> thanks - yeah, I think that is the main blocker at the moment
15:11:47 <witek> this should fix the failing pep8 job
15:12:34 <dougsz> So the templating of Slack notifications, you'd like to leave that until we switch to Oslo config in devstack? https://review.opendev.org/#/c/519630/
15:12:42 <dougsz> (that's ok with me)
15:13:21 <witek> I'd prefer, yes
15:13:27 <sc> ready to push the magic button on the bandit patch, but I'd like someelse to +1/+2 before
15:14:29 <witek> the monasca-log-api tests are still failing, so I've marked this job non-voting to unlock monasca-api repo
15:14:36 <dougsz> yeah, thanks for that.
15:14:44 <witek> https://review.opendev.org/659264
15:15:12 <witek> after this being merged, we can rebase oslo config change for devstack
15:15:27 <witek> https://review.opendev.org/633194
15:15:39 <witek> and then move on with notification changes on master
15:16:07 <dougsz> Makes sense, thanks. And I will look at backporting the bandit patch
15:17:15 <witek> sorry for holding this that long
15:17:27 <dougsz> No worries
15:17:40 <witek> anything else on notification engine?
15:18:01 <dougsz> That's it from me on Monasca notification.  Just wanted to collect the patches into a list to help drive them through
15:18:17 <witek> #topic persister bug
15:18:43 <witek> thanks dougsz, yes, let's make progress on these
15:18:46 <dougsz> So has anyone else hit this? I took a look at the code and I can't see how it can work at the moment
15:19:19 <witek> I have one idea, please check the driver configuration in persistre
15:19:39 <witek> there was a bug in the default config
15:20:00 <dougsz> Ah sorry, just saw your comment to post config
15:20:16 <dougsz> Ok thanks, I will do that. That's all from me on this one.
15:20:50 <witek> we might want to cherry-pick the bugfix if it turns out to be the same one
15:21:38 <witek> #topic monasca-log-api
15:22:16 <witek> sc: is that you?
15:22:18 <sc> I think we have to set some test tonot voting
15:22:41 <witek> oh yes, I've checked with adriancz today
15:22:43 <sc> we are in a loop like for api
15:23:11 <witek> he's made some progress on https://review.opendev.org/660958
15:23:29 <sc> I check locally all changes are good all together
15:23:44 <dougsz> py3 tests pass now?
15:23:48 <witek> have you run tempest tests?
15:23:59 <sc> no, not tempest
15:24:10 <witek> these are still failing
15:24:13 <sc> this is the missing part
15:24:27 <witek> yes, still needs some love
15:24:57 <sc> I commented on requirements-check
15:25:57 <witek> in the meantime, I set monascalog-python3-tempest non-voting in monasca-api repo
15:26:05 <dougsz> Looking close now - good job adriancz
15:26:47 <witek> I think https://review.opendev.org/659264 is ready to merge
15:26:54 <witek> yes, thanks adriancz
15:28:26 <witek> #topic CFP Summit Shanghai
15:28:42 <witek> it seems again much too early
15:28:57 <witek> Deadline: July 3rd, 7am UTC
15:29:53 <witek> akhil_jain: do you plan to attend?
15:30:18 <akhil_jain> witek: it depends on session selection
15:30:26 <akhil_jain> not planned yet
15:30:43 <witek> it's closer for you this time
15:31:11 <akhil_jain> haha, hopefully this time it will be easy for me to join
15:31:17 <witek> :)
15:31:51 <witek> #topic Akhil nomination for core team
15:32:23 <witek> I would like to nominate Akhil to the core reviewer team
15:32:41 <witek> I've sent an email to the list already:
15:32:47 <witek> http://lists.openstack.org/pipermail/openstack-discuss/2019-May/006538.html
15:33:02 <dougsz> +1 from me, thanks for your work Akhil
15:33:45 <witek> I find the X-project work in particular important
15:34:37 <witek> thanks Akhil for the reviews
15:35:19 <witek> #topic InfluxDB upgrade
15:36:34 <witek> dougsz: is it yours?
15:36:41 <dougsz> Ah yes
15:37:17 <dougsz> So I've done an upgrade to InfluxDB 1.7 and unlike 1.4 it seems to work out of the box
15:37:39 <dougsz> It's been running for a week or so now with a fair bit of load
15:37:55 <dougsz> The migration procedure from 1.3 was fairly straightforward
15:38:14 <witek> nice, what has to migrated?
15:38:22 <dougsz> As part of the migration I enabled disk backed index which allows for higher cardinality data sets
15:38:43 <dougsz> Not as good as TimeScale claims, but it has helped to reduce memory usage for us
15:39:13 <dougsz> Basically they rely on the kernel to cache the most frequently uses indicies, rather than holding all indices in memory all of the time
15:39:28 <dougsz> It seems to work pretty well
15:39:54 <dougsz> Once the gate is fixed is I can write some notes and try and get the devstack change to upgrade it merged.
15:40:21 <witek> you mean migration notes?
15:40:31 <dougsz> Yeah
15:40:43 <witek> cool, looking forward to this
15:40:53 <dougsz> and something about enabling the disk based index which is off by default
15:40:55 <witek> have you tried running tempest tests?
15:41:12 <dougsz> Not yet - I'll try that in CI
15:41:30 <witek> sounds good, thanks a lot!
15:41:37 <dougsz> I've spent some time testing CLI commands manually, and using the Grafana datasource
15:41:46 <dougsz> no problem - it was suprisingly easy for once
15:41:57 <dougsz> Also, it fixes the security issue in 1.3
15:42:09 <dougsz> Where tenant isolation is broken
15:42:19 <witek> also, is supposed to be much faster
15:42:21 <dougsz> And a tenant can dump dimensions values from all tenants
15:43:24 <witek> #topic open stage
15:44:11 <witek> I've added some tasks from the auto-scaling SIG to our board
15:44:17 <witek> https://storyboard.openstack.org/#!/board/141
15:45:06 <witek> one thing I would like to add is libvirt plugin configuration in DevStack
15:46:08 <witek> it would allow developers to use instance tenant metrics more easily
15:47:09 <witek> Rico Lin wants to write tempest tests for auto-scaling with monasca and heat, it would be based on devstack plugin
15:47:41 <witek> apart from this I have attended Telemetry meeting last week
15:47:48 <witek> here the logs:
15:47:55 <witek> http://eavesdrop.openstack.org/meetings/openstack_telemetry/2019/openstack_telemetry.2019-05-23-08.00.log.html
15:48:16 <witek> good news is, they agreed to merge our Monasca publisher
15:48:40 <witek> which is great
15:49:38 <witek> also, Trinh, the new PTL, considers migrating to Monasca as a possible goal for Telemetry in 2-3 cycles
15:50:03 <witek> we could work together on documenting this
15:50:51 <witek> I've also join the billing initiative meetings
15:50:59 <witek> http://eavesdrop.openstack.org/meetings/publiccloud_wg/2019/publiccloud_wg.2019-05-23-14.00.log.html
15:51:11 <witek> http://eavesdrop.openstack.org/meetings/publiccloud_wg/2019/publiccloud_wg.2019-05-28-14.05.log.html
15:51:50 <witek> the first AI from there is to collect the list of metrics relevant for billing
15:52:05 <dougsz> Good stuff
15:53:00 <witek> that's all from me
15:53:32 <witek> any other topics?
15:53:44 <sc> no
15:54:10 <witek> thanks everyone, see you next week
15:54:58 <witek> bye bye
15:55:02 <witek> #endmeeting