16:00:13 #startmeeting 16:00:13 Meeting started Thu Jul 5 16:00:13 2012 UTC. The chair is nijaba. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:13 #meetingtopic Ceilometer 16:00:13 #chair nijaba 16:00:13 #link http://wiki.openstack.org/Meetings/MeteringAgenda 16:00:13 #info jd__ will unfortunately not be able to join us this week 16:00:14 Hello everyone! Show of hands, who is around for the ceilometer meeting? 16:00:14 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:16 o/ 16:00:17 Current chairs: nijaba 16:00:24 o/ 16:00:26 o/ 16:00:39 jd___: I thought you could not join? 16:00:46 nice to see you! 16:00:50 for now I can, I may leave early 16:00:56 flacoste: around? 16:01:03 i am 16:01:08 o/ 16:01:17 ok, great!, let's start! 16:01:27 #topic actions from previous meetings 16:01:27 I don't see any action from the last meeting. Did I miss anything? 16:01:45 IIRC the only thing that carried over was the discussion of the incubation proposal 16:01:49 yep 16:02:00 ok, we have that as a topic for today 16:02:03 well, that and "keep working on the code" :-) 16:02:17 hehe 16:02:19 #topic Discuss and vote the incubation application proposal to be submitted to PPB 16:02:19 #link http://wiki.openstack.org/Projects/IncubatorApplicationCeilometer 16:02:19 Any remarks before we vote on this? 16:02:57 I think it looks good, but do we have examples of the proposals other projects have put up (especially successful proposals)? 16:03:08 looks good to me 16:03:14 dhellmann: I don't think there has been any... 16:03:26 ceilometer: breaking new ground, yet again 16:03:42 shall we vote then? 16:03:47 yes 16:03:52 #startvote Submit the incubation application to the PPB? Yes, No 16:03:52 Please note that to register a vote you have to use #vote yes or #vote no 16:03:53 Begin voting on: Submit the incubation application to the PPB? Valid vote options are Yes, No. 16:03:55 Vote using '#vote OPTION'. Only your last vote counts. 16:04:02 #vote yes 16:04:03 #vote yes 16:04:04 nijaba: yes is not a valid option. Valid options are Yes, No. 16:04:05 dhellmann: yes is not a valid option. Valid options are Yes, No. 16:04:11 hahaha 16:04:14 #vote Yes 16:04:15 #vote Yes 16:04:21 #vote yes 16:04:22 jd___: yes is not a valid option. Valid options are Yes, No. 16:04:31 amazing these little tools... so picky ;) 16:04:32 #vote Yes 16:04:38 yeah yeah c'mon openstack! 16:04:46 #vote yes 16:04:47 flacoste: yes is not a valid option. Valid options are Yes, No. 16:04:52 ffs 16:04:54 #vote Yes 16:05:03 * dhellmann lol 16:05:11 ok, that's a wrap for the vote 16:05:14 #endvote 16:05:15 Voted on "Submit the incubation application to the PPB?" Results are 16:05:16 Yes (4): nijaba, dhellmann, jd___, flacoste 16:05:36 So, I'll send a very formal email for our application ASAP 16:05:45 thx! 16:05:48 #action nijaba to send an email to the PPB 16:06:07 #topic PTL election process discussion 16:06:08 When we had our first project meeting on April 26th, it was agreed that Loic Dachary and I would co-lead this project for the first 3 months. Time is soon up for those 3 months, so we should discuss what to do to. 16:06:11 nijaba, is that a publicly readable mailing list? 16:06:22 it is, yes 16:06:29 the archives are 16:06:35 #link http://wiki.openstack.org/Governance/Model#Governance%20for%20the%20Individual%20OpenStack%20Projects 16:06:35 #link http://www.cs.cornell.edu/andru/civs.html 16:06:35 #link http://www.opavote.org/ 16:06:35 #link http://wiki.openstack.org/Governance/Foundation/TechnicalCommittee 16:06:35 I propose that we: 16:06:36 1/ Make a call for candidate next week, asking for applicants to do so on the general mailing list 16:06:39 2/ Open a one week voting period from July 26th til Aug 2nd 16:06:41 3/ Use the opavote voting as it is less likely to produce ex-aequo 16:06:43 4/ Limit voters and candidates to people that have contributed either to the code or to the wiki pages for ceilometer prior to the call for candidates 16:06:45 Comments? Remarks? Suggestions? 16:07:04 * jd___ checks opavote 16:07:41 the voting limits seem appropriate. do you already have a list of who those people are? 16:07:46 * nijaba waits a couple minutes for people to digest 16:07:54 dhellmann: nope 16:08:07 nijaba: sounds good to me 16:08:10 but it would not be too hard to compile 16:08:10 not that many I guess 16:08:31 what's the voting system for other PTL elections? 16:08:50 they were using civs, but are about to switch to opa 16:09:04 ok, that seems fine anyway :) 16:10:01 looks like we have a general agreement then? Should we formally vote on this? 16:10:25 yes and if you want yes 16:10:27 :) 16:10:39 yes 16:11:19 #startvote go ahead with the proposed PTL election process? Yes, No 16:11:20 Begin voting on: go ahead with the proposed PTL election process? Valid vote options are Yes, No. 16:11:21 Vote using '#vote OPTION'. Only your last vote counts. 16:11:28 #vote Yes 16:11:28 #vote Yes 16:11:39 #vote Yes 16:11:57 #vote Yes 16:12:04 #endvote 16:12:05 Voted on "go ahead with the proposed PTL election process?" Results are 16:12:06 Yes (4): nijaba, dhellmann, jd___, flacoste 16:12:22 cool. Do you want me to take the call for candidate action? 16:12:52 however, I'll be off until the 28th, so someone else will need to do the opa setup 16:12:56 volunteers? 16:13:08 I probably can do this 16:13:14 thanks jd___ 16:13:43 #action nijaba to send call for candidate to the general mailing list 16:14:15 #action jd___ to setup opa voting system to start on 26th and end on Aug 3rd 16:14:28 ok, moving on 16:14:36 #topic Discuss project's next priorities 16:14:36 This topic was suggested by dhellmann, so I let him drive it 16:14:54 Thanks, nijaba. 16:15:25 It seems to me like we've reached another milestone now that we can actually store data. There are a couple of different directions we can go from here, either one at a time or in parallel. 16:15:51 There are still quite a few plugins to write, and code to integrate with other projects (especially for notifications, but also for polling). 16:16:07 We need to build the API server. 16:16:12 I think we may need additional agents too? 16:16:30 yes, good point, we may need other agents depending on where the plugins need to run 16:16:50 we need at least one more for polling services through their REST APIs 16:17:08 we may also want to implement another db backend to ensure that the plugin layer API is not mongodb-specific 16:17:29 I think we need to list these items on a roadmap wiki page, linking to bugs, that people can assign themselves to... 16:17:40 nijaba: +1 16:17:48 good idea 16:17:53 indeed 16:17:53 and then publicise it to try to recruit new coders 16:18:17 that was the idea behing documenting how to write pludgins 16:18:23 I would like to finish writing some more dev docs before we try to recruit too many more contributors -- I am doing that today and hope to have a patch ready for review tomorrow 16:18:33 sounds great 16:19:00 Should we start a general action to have a ceilometer/roadmap page populated? 16:19:09 +1 16:19:10 I can prime it 16:19:12 yes 16:19:30 does anyone have any strong opinion about what we should make our next priority? 16:19:46 #action nijaba to prime a roadmap page and invite others to populate it 16:20:09 I would like to have the data going end-to-end (agent -> collector -> db -> api -> consumer) soon 16:20:18 yeah 16:20:18 dhellmann: I'd think swift, but that will not match what YOU will want ;) 16:20:18 probably the API server 16:20:25 i think the api server is probably the missing piece there 16:20:53 and implement the meter type as we discussed several time I think it's still missing 16:20:58 nijaba, we'll get swift done, don't worry :-) 16:21:17 jd___ there's a field in the meter messages for "type" but nothing actually uses it right now 16:21:24 dhellmann: leave something for flacoste's team when it get there :) 16:21:35 dhellmann: yeah, I should really tackle this down instead of talking about it 16:21:43 nijaba: it's getting there :-) 16:21:50 jd___: feel free to assign an action to yourself ;) 16:21:51 nijaba, if flacoste wants to work on swift we can wait 16:22:13 #action jd___ handle counter/meter type for real 16:22:17 dhellmann: I suspect you will want to do someting with some other tech ;) 16:23:12 nijaba, yeah, I think ceph already has some monitoring in place, but that's up to another team to handle for now. :-) 16:23:21 k 16:23:35 ok, should we move on? 16:23:58 yes, I'm done and it sounds like we have a good plan for working out what to do next 16:24:05 #topic Open Discusssion 16:24:16 I have one: External API. Based on the mailing list discussion 16:24:16 #link http://openstack.markmail.org/thread/fwkagzdzpleeclj5 16:24:16 I believe we reached an agreement on the purpose and limitations of the API, but wanted to make sure we were... 16:24:47 nijaba: i really appreciated your clarification email 16:24:50 I think so, for now at least :) 16:24:53 and agree with it 16:25:02 flacoste: thanks :) 16:25:33 yes, nijaba, thank you for that! 16:25:44 I am assigning myself a documentation task for the API with a clear warning about the limitation of the sum and duration function 16:25:55 #action nijaba to document external API use with a clear warning about the limitation of the sum and duration function 16:26:29 well, actually, I will start a doc, which will have to wait for completion hat we have something coded :) 16:27:00 does anyone want want to bring something else? 16:27:13 * dhellmann shakes head no 16:27:22 sounds like a wrap... 16:27:25 5 16:27:28 4 16:27:31 3 16:27:34 2 16:27:36 1 16:27:40 #endmeeting