15:00:17 #startmeeting monasca 15:00:18 hey all 15:00:18 Meeting started Wed Feb 27 15:00:17 2019 UTC and is due to finish in 60 minutes. The chair is witek. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:19 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:21 The meeting name has been set to 'monasca' 15:00:22 hi dougsz 15:00:38 the agenda for today: 15:00:43 https://etherpad.openstack.org/p/monasca-team-meeting-agenda 15:00:49 feel free to add topics 15:01:27 #topic OpenStack release dates 15:01:37 just a reminder 15:01:59 we're approaching a couple of dates in the release schedule 15:02:23 Non-client library freeze being the first ont 15:02:25 one 15:02:27 tomorrow 15:02:42 it affects mainly monasca-common 15:03:30 Would be nice to get this change in if that's possible: https://review.openstack.org/#/c/638423/ 15:03:55 Oops, sorry, note relevant 15:03:57 *not 15:04:08 would be good anyway :) 15:04:22 *monasca-agent change for statsd 15:04:50 we have 2 last minute changes in monasca-common 15:04:55 https://review.openstack.org/637190 15:05:07 and Kafka https://review.openstack.org/638725 15:05:41 I'll have a look at the validation change tonight 15:06:29 I think we can probably also merge https://review.openstack.org/626761 15:07:16 the next date is next week 15:07:21 for python-monascaclient 15:07:49 and then RC1 on March 21 15:08:55 please have a look through the repos and check which changes should land until then 15:09:12 #topic Kafka client 15:09:39 I've pushed implementation for the Kafka consumer in monasca-common 15:09:45 https://review.openstack.org/638725 15:10:08 Great to see this moving forward, need to take a closer look at that review when I have time 15:10:13 it has become much simpler than the kafka-python one 15:10:54 which makes me wondering if we indeed shouldn't move it out from monasca-common 15:12:18 If we move it out of monasca-common, how many places would we need to duplicate it? 15:12:28 I suppose the pattern of being able to swap in another kafka client and not update all the other monasca projects is a good one. 15:12:48 If the wrapper can be generic enough 15:13:19 dougsz: that's a point for leaving it in monasca-common, at least in the first phase 15:13:37 joadavis: producer in api and notification 15:13:52 consumer in persister and notification 15:14:10 then we also have monasca-transform 15:15:32 if we decide to leave it in monasca-common, I would like to merge it before the freeze this week, so that we can complete the work in Stein 15:15:50 Do you have a Monasca deployment with that wrapper running in production? 15:15:55 I know it's very short, but the code will not break anything for now 15:16:01 no 15:17:54 +1 for merging it now 15:17:56 I see, so you can get it in now, and then patch it up after the deadline 15:18:18 (if it needs patching up when you come to use it) 15:18:24 yes, I'm adding unit tests right now, should be ready tonight 15:20:16 should we move on? 15:20:42 #topic Dropping parse errors in monasca-persister 15:20:53 https://review.openstack.org/638589 15:21:41 the change allows for dropping the messages being blocked due to parsing errors in InfluxDB 15:22:55 but I think the actual reason might be the newer version of InfluxDB which we currently test 15:23:17 I haven't checked that though 15:23:59 Yeah, need to investigate more - I have hit similar problems to this with badly formatted metrics breaking the persistor 15:24:27 what was the reason? 15:25:20 I forget exactly now, but invalid characters in metric or something. Caused by metrics generated from logs not going via the Monasca API for validation. 15:27:20 I think that commit would be fine if we can get to the root cause. I might like more detail in the log message if possible, and wonder if it should be error instead of warning 15:29:21 monasca-persister is our performance bottleneck that's why I'm a little reluctant 15:29:51 and also I would like to add full support for newer InfluxDB if that's the root cause of the problems 15:30:17 yeah - I agree it needs more investigation 15:30:45 but I'm fine with merging this one if we don't fine a better way 15:31:28 #topic legacy jobs migration to Bionic 15:32:01 there is a recent thread on mailing list about moving to Bionic for all jobs 15:32:39 I've checked that we still build jars with legacy Xenial jobs 15:32:57 I think in the thread mainly testing jobs were meant 15:33:13 but we should migrate these anyway 15:33:30 egacy-monasca-{common,api,persister,thresh}-maven-build 15:33:47 legacy-monasca-{common,persister,thresh}-localrepo-upload 15:34:14 we're actually not publishing API jar anymore 15:34:35 I guess we can remove building this as well, right? 15:34:59 meaning legacy-monasca-api-maven-build 15:36:02 I'll create a story in backlog for this 15:36:14 if someone has time, can take care of it 15:36:35 I hope it works without greater problems 15:37:10 #topic metrics grouping 15:37:54 mohankumar: is that one yours? 15:38:03 Hi witek, that's mine :) 15:38:08 hi pandy 15:38:14 witek : It's pandy 15:38:37 first would like to thanks all for accepting paper to present at denver summit. 15:38:54 congratulations! 15:39:01 congratulations!! 15:39:05 +1 15:39:11 pandy : congrats mate ! 15:39:38 coming to point, currenly VM dimesions are grouping with tenant_id & resource_id 15:40:22 we had feedback from end users to have "tenant_name" & "resource_name" 15:41:20 so would like to add that feature, looking forward your advise 15:42:24 documentation says "VM name and tenant name (in addition to default IDs) can be provided as dimensions if vm_name and tenant_name are provided in the list of metadata keys." 15:43:18 so either the names can be included in the metric when generated, or they can be matched up against data in post-processing 15:43:38 https://github.com/openstack/monasca-agent/blob/9ee96263e9b7ae03c8a061cfecaa4e8fcda1bd87/docs/Libvirt.md#vm-dimensions 15:44:59 I could see that in documents, but when we query metrics with dimensions "tenant_name" getting invalid parameter ? 15:45:23 http://git.openstack.org/cgit/openstack/monasca-agent/tree/docs/Libvirt.md#n70 15:45:49 please check `metadata` in you libvirt plugin configuration 15:48:36 ok, I will check 15:48:55 https://github.com/openstack/monasca-agent/blob/master/monasca_agent/collector/checks_d/libvirt.py#L206 15:49:02 that should work, I guess 15:50:28 #topic open stage 15:50:42 do we have anything else? 15:51:05 oh, I haven't noticed the new entry in agenda 15:51:28 okay, thanks. I got it :) 15:51:31 #topic custom alerting templates 15:52:06 witek : please review and approve :) 15:52:19 https://blueprints.launchpad.net/monasca/+spec/add-custom-alert-templates 15:52:42 that's something that we have in our backlog for some time already 15:53:03 witek : okay 15:53:03 there is a story for templating Slack notifications 15:53:26 witek: I see let me check 15:53:46 https://storyboard.openstack.org/#!/story/2001308 15:54:18 we definitely need something similar for emails as well 15:54:48 btw. we don't use launchpad anymore and moved to Storyboard 15:54:49 agreed, I need to finish the Slack one 15:55:30 here is the board with current tasks: 15:55:32 https://storyboard.openstack.org/#!/board/111 15:55:46 yes, i remember that i was discussed same with dougsz some time back to customise notification at present it's verbose mode and hard to understand quickly 15:56:23 mohankumar: is that something that you could/would like to work on? 15:57:17 witek : yes , I 'll sync with dougsz and will find way to customize for email 15:57:28 dougsz, would like to know status of slack notification ? if fine i can work on customising mail onle 15:57:49 could you please create a story for this? 15:58:06 witek : sure !C 15:58:14 great, thanks 15:58:33 we have to wrap up 15:58:42 another meeting in 2 min. 15:59:02 witek: another question , for bug reporting you use storyboard or launchpad ? 15:59:03 thanks for joining 15:59:17 thanks :) 15:59:19 and please have a look on Kafka change later today or tomorrow 15:59:26 bye 15:59:37 bye all, will check Kafka change tomorrow 15:59:37 mohankumar: storyboard 15:59:39 thanks all ! 15:59:44 and tag it with `bug` 15:59:55 witek : got it ! 15:59:57 #endmeeting