15:00:29 #startmeeting monasca 15:00:30 Meeting started Wed Dec 20 15:00:29 2017 UTC and is due to finish in 60 minutes. The chair is witek. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:30 thx; I'll try to have a look for the topic 2 15:00:31 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:33 The meeting name has been set to 'monasca' 15:00:35 thanks for chairing tobberydberg ! 15:00:50 Hello 15:00:54 hello everyone 15:01:43 agenda quite empty today 15:02:17 I have created the etherpad for PTG planning 15:02:52 greetings 15:02:58 hello joadavis 15:03:36 people attending the PTG please add your names 15:04:00 and also if you already have some topics, feel free to add in the etherpad 15:04:07 https://etherpad.openstack.org/p/monasca-ptg-rocky 15:05:25 Monasca sessions are planned on Wed 28.2. and Thu 1.3. 15:06:18 Is that fixed already or just what we are shooting for? 15:07:36 there is no official schedule, but from what I understood we're free to fill 'teams days' 15:07:42 which are Wed-Fri 15:07:53 Ok 15:08:09 As for topics: for me that would mainly be 15:08:27 https://storyboard.openstack.org/#!/story/2001214 15:09:15 yes, we probably won't be able to finish it in Queens 15:09:30 I haven't gotten very far on in Queens (Grafana 4 and now Cassandra on SOC 8 sidetracking me :-/ ) and I don't think we'll make much more progress in Queens 15:09:37 s/on/on it/ 15:10:55 #topic Cassandra 15:11:24 Cassandra changes have been merged 15:11:38 testing framework still need review 15:11:52 ...and they even mostly work on Pike :-) 15:12:00 nice 15:12:05 (test drove James' unofficial backport today) 15:12:30 There are still a few rough edges (for instance libvirt aggregate metrics are giving me some trouble, but overall it works) 15:13:35 you mean the metrics you've added in your recent change? 15:13:44 Yes 15:13:59 So the problem may actually be with the metrics themselves 15:14:45 I have to take a closer look - so far I've only notice them being absent in Grafana and some suspicious submission failures on the collector side that started appearing when I switched to Cassandra 15:15:28 jgr: hhmmm let me know if I can help. cassandra does not touch the agent collector. 15:15:47 yes, that looks weird to me too, please keep us updated 15:16:12 jgu: no, it doesn't. The problem starts in monasca-api or further up. 15:16:37 jgu: as I said, I'll have to take a closer look first. For now I'm happy I got it running at all :-) 15:17:04 jgr: okay. I just joined. I'll check the meeting log for the the full details. wow, that's very cool! 15:18:01 #topic milestone 2 15:18:20 I have postponed creating the tags until Cassandra is merged 15:18:37 I'll do it after the meeting 15:18:52 witek: that was very much appreciated! 15:19:31 updated Java versions are also on the way 15:20:53 do you have other topics for today? 15:21:36 re the updated Java versions, I didn't find threshhold has 1.2.1. If I should have missed it, pls let me know 15:21:49 is already merged 15:22:30 now waiting for api to get green 15:22:31 Ah, these Java versions. That might be something to discuss at the PTG, too. 15:22:51 or rather deprecating Java support 15:23:19 Heh. +8^t from me for that :-) 15:23:51 we can put some more time on the cassandra python persister to see if we have the same performance as the java version. 15:23:52 I don't understand your dialect :) 15:24:33 8^t: digit 8 with a 't' for "transposed" in the exponent. Aka "lying 8", aka "inifinity". 15:24:59 wow 15:25:22 yeah, wow. I thought it was just a grumpy pointy-nose smiley 15:25:36 jgu: that would be beneficial in any case 15:25:46 (saw that in my university's math building once. Some joker had labelled the 8th floor button in the elevator with an infinity sign and a 'T' exponent :-)) 15:26:06 :) 15:26:53 What do we do about monasca-thresh, though? 15:27:21 Is there a Python equivalent for that on the horizon? 15:28:13 i've seen some python library for storm, we could also choose to go away from storm 15:30:12 What would be the alternative to Storm? 15:30:37 Spark has been good for monasca-transform. Or Go as was used in monasca-aggregator. (no, I'm not signing up to write either) 15:32:01 replacing storm with Spark is an interesting idea as one less 3rd party library 15:32:01 Spark is not very helpful from my point of view. It's another square Java peg to fit into a round RPM hole :-) 15:32:26 jgr: but less square pig :-) 15:32:54 is there other python alternative? 15:32:58 jgu: I've already built the Storm package, though... :-) 15:33:59 I think we should start with collecting the functional features the new thresh should have 15:34:01 And you did some good work on a Spark rpm ;) 15:35:06 jodavis: there is that. We'll need Spark anyway, so if we can drop Storm that's probably not too bad... 15:36:23 Agree on collecting features. I'm not familiar enough with it to know if Spark would handle everything thresh is currently getting from Storm 15:37:33 jgr: have you deployed monasca-transform already? 15:38:39 witek: yes. I think it was even in Cloud 7 from the first release. 15:39:40 witek: or wait...no. 15:40:14 witek: I got that mixed up with openstack-monasca-log-transformer (the simple logstash based log transformer) 15:40:26 ok 15:40:49 witek: you mean https://github.com/openstack/monasca-transform, correct? 15:40:54 yes 15:41:20 I've put new monasca-thresh as the topic for PTG 15:41:37 Sounds good 15:42:15 witek: is there an etherpad for ptg? 15:42:26 https://etherpad.openstack.org/p/monasca-ptg-rocky 15:42:33 fresh :) 15:43:34 nice :-) 15:44:39 anything else for today? 15:45:23 if not, I wish you all Merry Christmas 15:45:43 Thanks :-) 15:45:47 Merry Christmas to you too 15:45:47 Merry Christmas! 15:45:49 A Merry Christmas to you, too! 15:46:02 we skip the meeting next week 15:46:11 and meet next year 15:46:12 then Happy New Year to all :-) 15:46:22 :) 15:46:29 Happy New Year 15:46:59 see you, bye bye 15:47:02 Merry Christmas all 15:47:26 #endmeeting