14:01:18 #startmeeting monasca 14:01:19 Meeting started Wed Jun 28 14:01:18 2017 UTC and is due to finish in 60 minutes. The chair is rhochmuth. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:22 The meeting name has been set to 'monasca' 14:01:41 o/ 14:01:42 hello 14:01:50 hi witek 14:02:00 hi 14:02:02 i just realized that i still have a number of reviews to complete 14:02:06 hi koji 14:02:20 hello 14:02:27 hi kornica 14:02:42 looks like a pretty small group again 14:02:49 pleasure to meet you sire ;) 14:03:11 and the agenda is a little light 14:03:17 group is small but strong 14:03:54 so besides all the stuff that I said I would do last week that I didn't get to, are there other items to cover today 14:04:24 I'd mention that we reached a state where we have a gates in place for documentation 14:04:31 kornica works on publishing docu 14:04:38 for log-api 14:04:54 results can be examined here -> https://review.openstack.org/#/c/437889/ 14:05:04 witek: yes that's only for log-api, clearing the path 14:05:15 api-ref, api-guide, dev docs and releasenotes 14:05:32 very impressive 14:05:37 what's missing and could be a subject of discussion are install docs, but that's point for agend for another meeting I guess 14:06:07 rhochmuth is there any agenda at all, I don't want to jump in front of it 14:06:07 let's fill this docs with body first 14:07:05 one thing to mention perhaps 14:07:22 witek: btw, I addressed comments here => https://review.openstack.org/#/c/478480/ 14:07:23 CloudKitty started work on integrating with Monasca 14:07:36 yes, saw that 14:07:46 thanks for answering their questions 14:08:22 Ashwin and Joseph has given valueable input on using monasca-ceilometer 14:08:36 for that purpose 14:08:56 thanks ashwin and joseph 14:10:55 I don't have anything else 14:10:59 have they uploaded anything to review ? 14:11:21 I don't think so 14:11:23 I mean what's the scope of their integration ? I think I got little behind in that topic 14:11:27 :( 14:11:41 i don't see anything 14:12:01 i'm going to check with the OP5 folks to see if they are going to start joining the meetings 14:12:02 they want to use Monasca as backend for storing metrics they use in their project 14:12:43 i promise i'll get some reviews done this week 14:13:10 btw, next week i'll probably be out for July 4th week 14:13:57 sure things (from my side I can tell you that keystone_auth and new_client topics) are in fairly good shape, however in new_client[monasca-agent] adrian found a flaw around libvirt and ovs that I need to fix tommorow 14:14:22 but apart from those two checks everything is ok (rest of agent, ui and client while used in cli) 14:14:23 thx kornica 14:14:52 we added rate calculation to monasca-aggregator recently 14:15:14 https://github.com/monasca/monasca-aggregator 14:16:06 i guess that is it for today 14:16:09 oh, it is written in go :) 14:16:19 yes i 14:16:20 did you check that maybe with newer kafka ? 14:16:21 does it cover all the functionallity of monasca-transform? 14:16:37 I mean monasca-aggregator? 14:16:39 yes, it covers everything in monasca-transform and more 14:16:47 witek: you want to say that it is meant to replace monasca-transform ? 14:17:08 I'm wondering what should be the future of these two 14:17:28 and it is a very easy to deploy/configure, very performant, with guarantees on collection 14:17:41 from what I saw they still contribute to monasca-transform, so that's a bid odd from my POV 14:18:05 what Kafka client do you use? 14:18:06 well, monasca-transform was developed and deployed in Helion OpenStack 14:18:29 Dependencies: Dependent on only the following Go libraries: 14:18:30 • https://github.com/confluentinc/confluent-kafka-go 14:18:30 • https://github.com/prometheus/client_golang 14:18:30 • https://github.com/sirupsen/logrus 14:18:31 • https://github.com/spf13/viper 14:18:54 confluent- 14:18:57 there is a nice readme that covers what monasca-aggregator does 14:19:04 yes 14:19:22 so the same thing that we considered a while ago for the rest of monasca, but that got swallowed by other more important priorities :/ 14:19:34 right 14:19:40 cofluent is async, right ? 14:20:01 right 14:20:14 it uses librdkafka 14:20:26 the go client is written ontop of that 14:20:33 having that established, couldn't we simply switch to oslo.messaging[kafka] and use newer kafka from their dependencies that is also async 14:20:42 for the rest monasca 14:20:45 high-level idea 14:21:48 i don't recall all the specific issues, but librdkafka requires a C build of the library 14:22:29 kafka in oslo.messaging is I guess the same on as in forked monasca code (what monasca used to use) but newer 14:22:37 well, not kafka but kafkaclient 14:25:12 anything else? 14:25:49 witek: you have something ? 14:25:59 no, thanks 14:26:09 ok, bye everyone 14:26:14 small croud 14:26:19 thank you rhochmuth 14:26:20 but twice as big a last week 14:26:26 thx 14:26:26 heh, cya 14:26:28 so we are gaining momentum 14:26:29 thank you everyone 14:26:39 2^2 14:26:44 haha 14:26:45 thank you 14:26:54 #endmeeting