15:00:27 #startmeeting monasca 15:00:29 Meeting started Wed Jul 27 15:00:27 2016 UTC and is due to finish in 60 minutes. The chair is rhochmuth. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:30 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:32 o/ 15:00:33 The meeting name has been set to 'monasca' 15:00:46 Agenda is at, https://etherpad.openstack.org/p/monasca-team-meeting-agenda 15:00:55 o/ 15:01:02 Agenda for Wednesday July 27, 2016 (15:00 UTC) 15:01:03 1. OpenStack Barcelona Summit Voting 15:01:03 2. Number of reviews or various organizations/contributors. 15:01:03 3. Java deprecation plan 15:01:03 4. Reviews 15:01:03 1. https://review.openstack.org/#/c/343905/ 15:01:03 2. https://review.openstack.org/#/c/347481/ 15:01:03 o/ 15:01:03 3. https://review.openstack.org/#/c/347532/ 15:01:04 4. https://review.openstack.org/#/c/332731/ 15:01:04 5. https://review.openstack.org/#/c/334856/ 15:01:05 3. Colorado Openstack Meetup 6th Birthday party: 15:01:05 o/ 15:01:10 o/ 15:01:18 o/ 15:01:19 o/ 15:01:20 o/ 15:01:23 hi everyone 15:01:56 #topic OpenStack Barcelona Summit Voting 15:02:00 o/ 15:02:18 The voting opened yesterday for the OpenStack Summit in Barcelona 15:02:45 As usual, you can search for topics as well as use the categories 15:02:48 For example, https://www.openstack.org/summit/barcelona-2016/vote-for-speakers/presentation/15929/?q=monasca 15:02:58 give all the Monasca sessions 15:03:13 It would be good to see folks vote on these based on interest 15:03:40 Does anyone know about the Masakari 2.0 project 15:04:07 that is listed as a session, and they mention Monasca and their interest/TODO collaboration 15:04:34 I guess not. 15:04:58 i don't 15:05:17 So, please take a look and vote for topics you are interested in. There are a number of session on monitoring, Vitrage, Watcher and Kubernetes that look interesting 15:05:47 will do 15:05:49 #topic Number of reviews or various organizations/contributors. 15:06:12 So, I've been looking at the overall numbers of reviews occurring by various organizations and individuals 15:06:47 My concern is that there is a huge dissaproportinate number of reviews from my team mostly at HPE 15:07:19 but your team is so good at it 15:07:53 Going forward I'm trying to figure out what to do, but new code submissions is going to have to take a back seat 15:07:59 to reviews 15:08:18 So, here is a good review to take a look at 15:08:20 https://review.openstack.org/#/c/334688/ 15:08:48 hpe 65% new lines of code, 73% reviews 15:08:50 It was posted on June 27th, and other than Tomasz, I don't see any other reviewers 15:09:05 apologies, will look 15:09:10 However, I see code showing up from other companies 15:09:46 and I constantly get pressure to priortize those reviews higher than others that are doing reviews 15:10:03 that probably needs to change and become a more equitable mix 15:10:15 Additionally, I'm reviewing the list of core members 15:10:32 Reviews is one of the primary reasons for being core 15:10:49 Being core is not lifelong position 15:12:27 So, I think folks just need to be aware of the amount of code that is going through the system and the need for taking part in more reviews to balance the load a little more 15:12:42 sounds like a good idea 15:13:31 agree 15:14:21 #topic Java deprecation plan 15:14:49 we briefly talked about ^^ at the midcycle 15:15:17 should we set a goal for deprecation at least? 15:15:19 thoughts? 15:15:36 yes, we should plan the steps to deprecate java 15:15:57 i would like to see a goal too 15:16:11 by deprecate you mean: 'hey java is deprecated' or 'rm -rf */java' ? 15:16:19 there is a plan to have packages to install monasca? just wondering =) 15:16:20 cassandra seems to be key to this plan 15:16:44 or a replacment for influxdb 15:17:07 if we at least get it to be 'deprecated', then we could stop maintaining java with new features -- i could see even only doing cassandra in python 15:17:09 iurygregory: There are packages for monasca 15:17:38 rhochmuth, like apt-get install monasca? 15:17:42 I think we agreed on doing cassandra only in python 15:17:42 O.o 15:17:53 maybe not apt...but they are on python pypi 15:18:03 iurygregory: i see. No we don't have a plan for that 15:18:23 oh my idea is to impro puppet-monasca and this would be good =) 15:18:27 improve* 15:18:51 Oh. So should I stop writing java cassandra? 15:18:51 maybe even that is part of the java deprecation plan iurygregory 15:19:07 you could add install of the python api/persister to puppet-monasca 15:19:26 bklei, yeah 15:19:59 shinya_kwbt it's a suggestion 15:20:37 shinya_kwbt: I agree with bklei, cassandra in Java is only additional effort 15:21:07 If support for Vertica and Cassandra was added to the persister, then we could deprecate the Java persister today. 15:21:42 The reason I mention that is that the persister is relatively simple 15:21:50 sounds good 15:22:03 good first step, would save shinya_kwbt java dev 15:22:34 Similar for the API 15:22:42 but i was thinking about staging development 15:22:58 althoguh if only the persister supports cassandra, then the api has to too 15:23:28 is anyone from DT here? 15:23:41 i can't think of a reason to add cassandra support for java for api or persister 15:24:44 if we are at the topic of open changes - there's also notification-story to think of, at the moment already having impl in java and python from what I saw 15:25:47 I agree with Roland, we have to make sure that Cassandra is the right way for Monasca first 15:25:59 and make progress with that 15:26:17 even more reason to do the 'pilot' only in python 15:27:14 I agree too. 15:27:38 If we had support for Vertica in the Python implementation, then HPE and TWC could switch to the Python implementation completely 15:28:03 We would essentially have parity at that point 15:28:05 * boom * 15:28:24 Then we could add support for Cassandra and hopefully deprecate InfluxDB 15:28:57 If Cassandra was a suitable replacemnt for Vertica, then we don't need to add support for Vertica to Python 15:29:34 hmm -- not sure on that -- existing deployments have the issue of data migration 15:29:39 so, shall we say: 'brace yourself - python is hissing' ? Personally still having both languages makes things harder, maybe not from POV of new features, but still any bugfixing and so on 15:29:42 i'd like to see vertica in python 15:30:40 So, if we added Vertica to Python, we have parity with what we have today, and we can stop development on Java 15:30:52 at least the Java API and Persister 15:31:35 That work woudl fall on HPE and TWC since we are the only one's using it today 15:31:42 using Vertica that is 15:31:52 yes -- we could split that work 15:32:20 ok, i'll have to look into that, but seems like a reasonable proposal/start 15:32:30 we should be GA for MaaS by 9/1, then i'll have some time 15:32:34 The Threshold Engine would remain in Java 15:33:05 threshold could be in python, but that would require writing that with spark I guess, like monasca-transform ? and that's also a lot of work 15:33:05 Then in parallel we could add support for Cassandra 15:33:19 wouldn't it be better to join the power and concentrate on Cassandra instead of Vertica? 15:33:33 + not realty sure if that is possible 15:34:06 witek: i would love to say yes, i just dno't know how well cassandra will compare to Vertica 15:35:16 also i don't know how long cassandra will really take 15:35:17 rhochmuth: we answer that question only if we check 15:35:48 i mean comparison with vertica 15:36:39 I don't know how speed and use space too. > i just dno't know how well cassandra will compare to Vertica 15:37:23 so, i think the vertica discussion is primarily a hpe/twc discussion 15:37:46 yup 15:37:46 i think we shoudl focus on a plan for cassandra with the greater community 15:38:26 i'll look into resources that are available to work on this over the next few months 15:38:35 and get back to you next week 15:38:40 great 15:38:45 from an hpe stand-point 15:38:47 i'll get approval for some of my time in sept rhochmuth 15:38:53 thx 15:39:24 i'll also follow-up with DT 15:39:44 witek: do you know from your end what would be available 15:39:45 rhochmuth: i could write them 15:39:52 thx 15:39:54 sounds good 15:40:38 rhochmuth: we should have some resources for cassandra, don't know exact numbers 15:41:05 so, if we could get resources and focus on this from everyone starting up in sept or sooner and lasting a few months we might be able to get over this hurdle 15:41:44 1 from HPE, TWC, DT, Fujitsu and NEC is a pretty serious effort 15:41:54 hope i didn't leave anyone out 15:42:00 i usually do 15:42:04 :) 15:42:42 so, how about we reconvene on this next week after everyone checks back with headquarters 15:42:57 good idea 15:43:06 sounds good -- and we have to train rhochmuth to say CHARTER, not TWC :) 15:43:13 hehe 15:43:20 you'll always be TWC to me 15:43:24 :) 15:43:53 #topic Colorado Openstack Meetup 6th Birthday party 15:44:00 http://www.meetup.com/OpenStack-Colorado/events/231949941/?gj=co2&rv=co2 15:44:00 Thursday 7/28 6:30 pm 15:44:00 CableLabs, 858 Coal Creek Circle, Louisville, CO, US, 80027 15:44:12 thought i woudl drop that in before the reviews 15:44:31 sounds like bklei_ is going 15:44:51 yes -- just want CO folks to know they are invited 15:45:00 beer, food, giveaways 15:45:05 thanks, i don't hink I'm going to make this 15:45:08 what, free beer 15:45:12 maybe 15:45:14 :) 15:45:33 #topic reviews 15:45:42 https://review.openstack.org/#/c/343905/ 15:45:53 That's mine. 15:45:56 yup 15:46:16 I was hoping to get this merged since it's been sitting around for a while, but there's been a flurry of comments I need to address now. 15:46:36 :) 15:46:47 roland said nobody is doing any reviews, so I thought I will post some buster there ]:-> 15:47:33 tomasztrebski: thanks for all your reviews 15:47:41 :-) 15:48:04 rbak: so it sounds like this is getting some attention 15:48:27 right. I'll address those comments today. 15:48:36 i gave a +1, but didnt review closely, as JS isn't my area 15:48:49 i'll try and take another look though 15:49:19 https://review.openstack.org/#/c/347481/ 15:49:36 I think I will also try to spend some time on JS there, maybe I will find something else to comment about 15:50:00 that review is just me, just want to advertise the change 15:50:06 It looks like this one is getting attention 15:50:06 I'd rather not fix every issue in that repo right now. We've been using this code for a while, this is just the repo move. 15:50:11 i have a bit more tempest testing to do 15:50:45 rbak: sounds like you want to get merged, and then start making improvements 15:50:56 as well as getting others involved 15:51:09 bklei_: yup 15:51:47 https://review.openstack.org/#/c/347532/ 15:52:40 shouldn't the other review go first? 15:53:23 Moving onto: https://review.openstack.org/#/c/332731/ 15:54:03 tomasztrebski: that is yours 15:54:27 looks like I have to look at this 15:55:00 it looks like you are free to merge that one 15:55:19 sorry, was writing a comment for previous one 15:55:44 yes, it sits there for some time and basically I'd like to know if anyone is fine with that or is there something that should be fixed 15:55:55 i'm fine with it merging 15:56:53 last one 15:56:54 https://review.openstack.org/#/c/334856/ 15:57:31 looks good to me 15:59:02 I don't see any room for improvement by my eye 15:59:19 but if someone would like to test a bit or review a code, it is always welcomed 15:59:54 thx tomasztrebski 16:00:02 i meant to have tsv look at this 16:00:18 and we might want to integrate into our helion distribution 16:00:35 rhochmuth, sure, will review 16:00:54 so, i need to end the meeting 16:01:01 thank you 16:01:21 #endmeeting monasca