13:00:12 #startmeeting monasca 13:00:13 Meeting started Tue May 26 13:00:12 2020 UTC and is due to finish in 60 minutes. The chair is chaconpiza. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:14 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:17 The meeting name has been set to 'monasca' 13:00:25 Hello everyone 13:00:39 hello 13:00:57 Hi Matthias 13:01:42 The agenda as usual https://etherpad.opendev.org/p/monasca-team-meeting-agenda 13:02:24 hi 13:02:36 I have changed a little bit the order in the agenda to start with the action items from previous meetings 13:02:40 Hi Witek 13:03:05 witek, thanks for leading last week. 13:03:09 #topic Action items from previous meeting 13:03:30 I think there was one for me, which I forgot... 13:04:01 From 2 weeks ago, I promised: chaconpiza Set minimum version for InfluxDB and influxdb-python-client. Clean the influxdb/metrics_repository 13:04:46 It will be on this change https://review.opendev.org/#/c/725779/ Still to do. 13:05:37 I think witek means: #action create a story to fix Cassandra installation in DevStack plugin 13:05:48 and #action witek mark monasca-tempest-python2-cassandra job as non-voting 13:06:00 o/ 13:06:05 right, will do right after the meeting, promise :) 13:06:18 hi Wasaac 13:06:28 Hi Isaac, welcome 13:06:38 thanks witek 13:07:05 alright, let's move to the first technical topic 13:07:17 #topic monasca-docker and zookeeper 13:07:31 OK, I'll try to explain quickly 13:07:49 We had a customer issue: disk space usage is grwoing for monasca-thresh. 13:08:26 chaconpiza's investigation has shown: Caused by snapshots and logs from zookeeper 13:08:47 Many steps of investigation later: 13:09:35 It looks like we're using in-process zookeeper in monasca-thresh for nimbus and supervisor (as mentioned in thresh log) 13:09:55 And, we're using zookeeper container for event_spout and metric_spout. 13:10:04 This sounds very strange to me. 13:10:15 Does anybody know a reason for this? 13:11:01 My guess: This happens unintentionally; ideally, only zookeeper container should be used. 13:11:33 However, I haven't figured out how to use zookeeper container for supervisor and nimbus 13:11:56 Any thoughts about this? 13:12:18 here is a note on using monasca-thresh with Storm 13:12:20 https://github.com/monasca/monasca-docker/tree/master/monasca-thresh#running-with-and-without-storm 13:12:34 The issue is easy to reproduce: Start the containers from monasca/monasca-docker and check the space used in /tmp inside the thresh container. It will slowly grow until no space left. 13:13:39 witek: OK, thanks, that might help! I will check for this 13:14:13 From Strom documentation: "It's critical that you set up a cron to compact Zookeeper's data and transaction logs. The Zookeeper daemon does not do this on its own, and if you don't set up a cron, Zookeeper will quickly run out of disk space. See here for more details." http://storm.apache.org/releases/current/Setting-up-a-Storm-cluster.html 13:14:20 Now, topic 2.6: I assume, nobody has come across the issue with huge disk usage? 13:14:21 I reckon it might be using localhost, because monasca-docker runs one-node installation 13:15:10 one more note on container configuration, you could compare the Kolla setup as well 13:15:20 Unfortunately, the functionality "purgeTxn..." (don't remember the name exactly, doesn't seem to be available in "in-process zookeeper" 13:15:58 Current approach: chaconpiza is going to write a script that deletes these files 13:16:46 Can confirm monasca-thresh volume disk usage problem in kolla-ansible too 13:17:10 Wasaac: thanks 13:17:17 Aah, ok. Did you resolve the issue? 13:18:01 Wassac, or how do you workaround it? 13:18:08 Have been looking at containers to mount and clean the volume 13:18:33 But it would be nicer to approach it from the source of the issue 13:19:26 chaconpiza: Could you pls. quickly explain the solution with zookeeper 3.4.14 - autopurge? 13:20:03 Other way to workaround it for the case of https://github.com/monasca/monasca-docker is to stop, remove containers, recreate containers and restart them. 13:21:09 bandorf: Sure, official zookeeper docker image 3.4.14 implements the autopurge with an env. variable 13:22:19 zookeeper 3.4.5 implements this code: https://github.com/krux/zookeeper-pkg/blob/3.4.5/src/java/main/org/apache/zookeeper/server/PurgeTxnLog.java 13:23:13 But as Matthias said, the reduced zookeeper provided by storm doesn't includes this part 13:24:55 Ok, we will keep you informed about this topic. 13:25:07 OK, thanks for your input anyway. Wasaac: We can stay in touch, regarding the temporary solution, we're going to provide. We can keep you updated. 13:25:17 OK, chaconpiza was faster :-) 13:25:32 :) 13:25:36 #topic kafka version in monasca-docker 13:26:08 Pls. check just the agenda. monasca-docker is stil using kafka 0.9. Is there any reason for this? 13:26:32 I think not, it just hasn't been updated 13:26:49 OK, thanks - topic done :-) 13:27:09 from what I recall we tested it in monasca-docker env 13:27:34 Yes, we executed system tests @est 13:27:45 #action Update kafka in monasca/monasca-docker 13:27:48 +1 13:27:57 +1 13:28:04 bandorf, would you like to own this action? 13:29:46 Let's continue with the last topic 13:29:49 I suggest, that we (EST) will take care of it. Assignment to an individual can be done later. But it will take time. 13:29:59 bandorf, thanks 13:30:12 #topic PTG planning 13:30:44 Kendall and Kendall contacted me to choose the option for the virtual PTG 13:30:56 haha 13:31:13 or short, Kendalls :) 13:31:14 one option was zoom and the second jitsi running on opendev 13:31:24 right, the Kendalls 13:31:53 I decided jitsi because it includes an etherpad 13:32:19 It will look like: https://meetpad.opendev.org/TestJitsiOnOpendev 13:32:45 do we have an option to have zoom as backup solution? 13:32:53 The final channel is not yet defined 13:33:21 #action Ask The Kendalls for a zoom backup channel 13:33:33 good point, I will ask them. 13:34:42 alright, concerning the content and topics for the vPTG 13:35:07 witek have already created this etherpad https://etherpad.opendev.org/p/monasca-ptg-victoria 13:36:30 I think it would be helpful to create Stories for each topic shortly describing the motivation and possible implementation direction 13:37:24 witek, do we have already the two time slots reserved? Tuesday June 2nd (7 - 8 UTC and 13 - 17 UTC) 13:37:32 yes 13:37:39 Great, thanks 13:38:20 we're in the Folsom room 13:38:26 https://object-storage-ca-ymq-1.vexxhost.net/swift/v1/6e4619c416ff4bd19e1c087f27a43eea/www-assets-prod/ptg/PTG2020-Schedule.pdf 13:38:58 oh, I think all participants should register! 13:39:08 it's for free 13:39:24 +1 13:40:04 Wasaac feel free to add the topics you consider important 13:40:48 chaconpiza: thanks 13:41:10 Wasaac do you know if Doug Szumski can join the PTG? 13:41:18 Is there a link to register? 13:41:34 https://www.openstack.org/ptg 13:41:35 Merged openstack/monasca-common master: Bump sublibraries https://review.opendev.org/730749 13:41:50 Thanks! 13:42:11 I think we could send an email to the mailing list advertising our sessions 13:42:51 #action advertising our sessions. I will take this action 13:42:52 also, there is the Automation SIG session on Thursday 13:43:29 chaconpiza: I don't know what dougz schedule is like but he'll make it if able I'm sure 13:43:52 the agenda is here 13:43:53 https://etherpad.opendev.org/p/Automation-2020-VPTG 13:45:25 I've just noticed our Etherpad is not linked correctly 13:45:30 http://ptg.openstack.org/etherpads.html 13:45:30 Thanks for the SIG agenda 13:46:16 I can update it with openstackptg bot 13:46:22 instructions at the bottom 13:46:49 thanks 13:48:23 Ok, let's continue this week extending the Monasca PTG topics. 13:48:49 #topic AOB 13:49:27 I have a small technical question 13:49:46 In monasca-notification and monasca-persister still some traces of ujson 13:49:46 http://codesearch.openstack.org/?q=ujson&i=nope&files=&repos=openstack/monasca-agent,openstack/monasca-analytics,openstack/monasca-api,openstack/monasca-ceilometer,openstack/monasca-common,openstack/monasca-events-api,openstack/monasca-grafana-datasource,openstack/monasca-kibana-plugin,openstack/monasca-log-api,openstack/monasca-notification,openstack/monasca-persister,openstack/monasca-specs,openstack/mon 13:49:46 asca-statsd,openstack/monasca-tempest-plugin,openstack/monasca-thresh,openstack/monasca-transform,openstack/monasca-ui,openstack/python-monascaclient 13:50:18 do we need them? 13:50:52 it's only lower-constraints, right? 13:50:55 yes 13:51:20 no, should be dropped 13:51:49 #action remove ujson traces from notificatin and persister 13:51:59 witek, would you like to send the change? 13:52:05 oki 13:52:09 thanks 13:52:50 Ok, those are all topics from the agenda. 13:53:45 another AI for everyone 13:53:57 think of topics for PTG 13:54:11 already posted and new 13:54:54 Will discuss with the StackHPC team 13:55:10 thanks 13:55:33 Thanks for joining today's meeting see you next week in the vPTG. 13:55:42 o/ 13:55:44 thanks chaconpiza 13:55:48 thanks everyone 13:55:53 see you next week 13:56:03 o/ 13:56:05 Tnaks, good-bye 13:56:08 #endmeeting