15:00:19 <nijaba> #startmeeting Ceilometer 15:00:19 <nijaba> #meetingtopic Ceilometer 15:00:19 <nijaba> #chair nijaba 15:00:19 <nijaba> #link http://wiki.openstack.org/Meetings/MeteringAgenda 15:00:19 <openstack> Meeting started Thu Nov 1 15:00:19 2012 UTC. The chair is nijaba. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:20 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:21 <openstack> The meeting name has been set to 'ceilometer' 15:00:23 <openstack> Current chairs: nijaba 15:00:29 <nijaba> Hello everyone! Show of hands, who is around for the ceilometer meeting? 15:00:29 <nijaba> o/ 15:00:34 <dhellmann> o/ 15:00:40 * dhellmann made it on time! 15:00:43 <anniec> o/ 15:01:09 <nijaba> #topic actions from previous meeting 15:01:16 <eglynn_> o/ 15:01:26 <krtaylor> krtaylor, o/ 15:01:44 <nijaba> #topic scrubbing actions from meeting notes 15:02:07 <nijaba> dhellmann: I think you were done with that right? 15:02:16 <dhellmann> yes 15:02:22 <dhellmann> I'm hunting for the right etherpad link... 15:02:24 <nijaba> same for you eglynn? 15:02:35 <nijaba> #link https://etherpad.openstack.org/grizzly-ceilometer-actions 15:02:54 <dhellmann> thanks 15:02:55 <eglynn> nijaba yep, I put a first cut on grizzly-ceilometer-actions 15:02:57 <nijaba> what about you jd? 15:03:04 <nijaba> jd__: ^^ 15:03:20 <eglynn> (I'll polish some more later on today ...) 15:03:29 <jd__> o/ 15:03:29 <nijaba> and I think I am done done (my part was easy)... 15:03:52 <jd__> done! 15:04:04 <nijaba> thanks, so this is complete ! 15:04:22 <nijaba> #topic jd to propose an implementation for source and wait for our comments 15:04:38 <nijaba> jd__: comments? 15:04:50 <jd__> this has been implemented 15:05:01 <nijaba> and approved, right? 15:05:10 <jd__> yes 15:05:16 <jd__> I can find the commit id, a minute 15:05:31 <jd__> https://github.com/stackforge/ceilometer/commit/f6868e5e4827e96754f832a0eae4c4210ea41df7 15:05:37 <jd__> and https://github.com/stackforge/ceilometer/commit/acf8fb88d56b776cf0b83041f5e87e1b6ee88fd2 15:05:43 <jd__> for the record :) 15:05:45 <nijaba> perfect. thanks a lot 15:05:54 <nijaba> #topic nijaba to update project objective on relevant pages 15:06:10 <nijaba> I think this was done wherever I could think of. 15:06:13 * jd__ is sorry to not have prepared the link before :) 15:06:24 <nijaba> let me know if you find some missing pages... 15:06:36 <jd__> nijaba: great :) 15:07:06 <nijaba> #topic Review actions from https://etherpad.openstack.org/grizzly-ceilometer-actions and agree on priorities 15:07:57 <nijaba> so, what do you think we should be prioritizing out of this work? 15:08:14 <dhellmann> I've marked a few of the items as done :-) 15:08:41 <jd__> just see this indeed 15:09:21 <eglynn> assessing commonality with other OS measurement-related initiatives is a high-priority work item IMO 15:09:28 <nijaba> should we all have 5 minutes on the etherpad to write our names next to the actions we consider most important? 15:09:37 <eglynn> cool 15:09:39 <jd__> remove all nova imports is almost complete done I think we should do it for G anyway 15:09:47 <jd__> not sure it's listed, but the multi-publisher blueprint I scrubed from the etherpad page beyond-metering should be included 15:12:59 * nijaba is done 15:13:44 <eglynn> jd__ does https://blueprints.launchpad.net/ceilometer/+spec/multi-publisher subsume https://bugs.launchpad.net/ceilometer/+bug/1073988 ? 15:13:45 <uvirtbot> Launchpad bug 1073988 in ceilometer "Adapt ceilometer for multiple publication conduits" [Undecided,New] 15:13:46 <eglynn> (I raised the bug to capture work from http://lists.openstack.org/pipermail/openstack-dev/2012-October/001840.html) 15:13:59 <eglynn> seems like a dupe ... 15:15:02 <dhellmann> eglynn: yes, those seem like the same thing 15:15:53 <sandywalsh> (hey! side note ... I'm nearly done the bp wiki page for architectural commonality) 15:18:00 <nijaba> ok, that looks good. So I proposed to sum this back in the wiki page as follow: 3+ votes=high, 1or2 vote=medium, low for rest in terms of priorities? makes sense? 15:18:31 <jd__> nijaba: +1 15:18:39 <dhellmann> +1 15:18:54 * jd__ have been disconnected, I hope I didn't miss anything 15:18:54 <dhellmann> are there any interested parties whose input we want but who are not present? 15:19:04 <nijaba> if you know of a bug# for each item, please add it to the list. otherwise we'll have to open one 15:19:15 <dhellmann> jd__: nope 15:19:20 <nijaba> dhellmann: angus? 15:19:22 <eglynn> asalkeld would be interested 15:19:24 <jd__> dhellmann: thanks :) 15:19:51 <nijaba> not seeing input from jtran either 15:20:18 <dhellmann> eglynn, nijaba : yes to both 15:20:27 <nijaba> I am a bit worried of asking other to complete the etherpad on the ml though 15:20:34 <nijaba> we may get a lot of rogue input 15:20:42 <dhellmann> nijaba: private email? 15:20:42 <eglynn> so perhaps we shuld leave the "vote" open for 24 hours? 15:20:50 <nijaba> dhellmann: yes, that sounds better 15:20:51 <eglynn> (to hit al TZs) 15:20:55 <jd__> +1 for private mail 15:20:57 <nijaba> eglynn: +1 15:21:17 <dhellmann> nijaba: it would be interesting to run a surveymonkey survey to the mailing list, too, as long as it's clear we aren't committing to follow the results 15:21:25 <dhellmann> eglynn: +1 15:21:49 <nijaba> dhellmann: I can try to prepare that for next week 15:22:28 <dhellmann> nijaba: ok 15:22:35 <nijaba> #action nijaba to send private email to all comitter to come vote on the etherpad in the next 24h 15:23:22 <eglynn> so I guess there's an implication that a vote for some high priority item comes with some (loose?) commitment to contribute time to it? 15:23:24 <nijaba> #action nijaba to then update wiki page as follow: 3+ votes=high, 1or2 vote=medium, low for rest in terms of priorities 15:23:38 <jd__> eglynn: not sure 15:23:40 <nijaba> eglynn: agreed 15:23:48 <dhellmann> eglynn: that's a good way to phrase it :-) 15:23:53 <eglynn> cool 15:24:09 <nijaba> eglynn: but I won't send the collector squad to enforce... ;) 15:24:24 <eglynn> nijaba: LOL 15:24:28 <dhellmann> I would like to have some input from "users" too, but only to help rebalance my own priorities 15:24:39 <eglynn> that's sensible 15:24:47 * nijaba cares about your kneecaps :P 15:25:44 <nijaba> I guess I should action the survey too... 15:25:46 * jd__ wonders if that does mean mean nijaba will have to code ;) 15:26:08 <dhellmann> jd__: "contribute" != "code", right? :-) 15:26:08 <nijaba> #action nijaba to prepare survey for ml next week 15:26:24 <jd__> dhellmann: ;) 15:26:31 <nijaba> jd__: it's halloween for reviewers, I would guess 15:26:38 <jd__> nijaba: lol 15:27:16 <nijaba> next topic? 15:27:20 <dhellmann> +1 15:27:31 <nijaba> #topic Changing the versioning to adhere to OpenStack standards 15:27:38 <nijaba> So I had a discussion with ttx 15:27:57 <nijaba> he would like us to adhere to openstack standard for versioning 15:28:21 <jd__> I don't see any problem with that 15:28:29 <nijaba> something like ceilometer -2013.1~g1~20121025.16635.tar.gz 15:28:46 <eglynn> seems reasonable 15:28:49 <nijaba> instead of current 0.2, that looks like a release but is not 15:29:07 <nijaba> example job: https://jenkins.openstack.org/view/Nova/job/nova-tarball/ 15:29:43 <nijaba> anybody with strong feelings about versioning? 15:30:10 <dhellmann> no, that seems fine 15:30:15 <anniec> following the standard is good 15:30:20 <dhellmann> is there a guide for decoding that format? some of it makes sense, but some is odd 15:30:35 <nijaba> anyone caring about it enough to take the action? 15:31:12 <dhellmann> I guess I'll take it 15:31:35 <dhellmann> #action dhellmann update versioning in ceilometer repo to match openstack standards 15:31:43 <nijaba> thanks dhellmann 15:31:46 <anniec> thanks dhellmann 15:32:16 <nijaba> that's it on the topics for today 15:32:27 <dhellmann> I have a couple of items to raise 15:32:27 <eglynn> so how much leeway do we have during the incubation period? 15:32:28 <nijaba> #topic open discussion 15:32:30 <eglynn> is the idea that we start adopting all standard openstack project practices ASAP or gradually? 15:32:45 <nijaba> eglynn: ASAP 15:32:49 <dhellmann> eglynn: we tried to follow most of them from the start 15:33:02 <dhellmann> is there something else we need to do besides the version numbering and milestones? 15:33:04 <eglynn> nijaba, dhellmann: cool, makes sense 15:33:08 <sandywalsh> versioning should really be following quickly 15:33:16 <nijaba> eglynn: apparently, the renaming of git will require a gerrit reboot, so that will have to be planned 15:33:20 <sandywalsh> we ran into that with novaclient ... made a big difference once we switched 15:33:39 <sandywalsh> the CI guys were very grateful. 15:33:42 <nijaba> I had a meeting with ttxt, mordred and jeblair on the subject this week (they are at UDS) 15:34:07 <dhellmann> I'll look into it between now and our next meeting 15:34:20 <nijaba> dhellmann: versioning and project name are the last bits, it seems 15:34:20 <eglynn> so I was wondering about the milestone releases, IIRC last week someone mentioned the idea was to get in train for G-3 ... did I understand that right? 15:34:30 <nijaba> eglynn: you did 15:34:36 <dhellmann> "in train"? 15:34:39 <nijaba> eglynn: but we can join eralier 15:34:49 <nijaba> earlier, even 15:34:53 <eglynn> in train = line up with with the other projects 15:35:02 <dhellmann> aha 15:35:03 <eglynn> so maybe earlier would be better 15:35:12 <jd__> nijaba: so there's a timeline for the switch or what? 15:35:24 <jd__> s/what// 15:35:35 <dhellmann> do we need to appoint an official release manager? 15:35:41 <nijaba> jd__: wait for an email on the ml announcing a downtine for gerrit 15:35:48 <jd__> I think the release team handle all project, no? 15:35:50 <nijaba> dhellmann: ttx is the release manager 15:35:54 <jd__> nijaba: ok! 15:35:56 <eglynn> yep ttx 15:35:57 <dhellmann> nijaba: ah, ok 15:36:10 <dhellmann> I didn't know if we needed someone to coordinate with him, but I guess that's nijaba 15:36:29 <eglynn> usually he coordinates with the PTLs AFAIK 15:36:38 <dhellmann> that would make sense 15:36:48 <nijaba> yup. I may ask for some help to go to the release meeting every other week, as I have a personal conflict with it 15:37:08 <nijaba> in fact, is there a volunteer to cover for me for next tue? 15:37:14 <jd__> what's the time? 15:37:33 <eglynn> 21:00 UTC 15:37:36 <nijaba> http://wiki.openstack.org/Meetings/ProjectMeeting 15:38:21 <eglynn> nijaba: I can cover it 15:38:30 <nijaba> eglynn: thanks a lot! 15:38:30 <jd__> can do that 15:38:35 <jd__> ok :) 15:38:46 <nijaba> jd__: feel free to join too 15:38:55 <jd__> yeah, I'll add it to my agenda :) 15:38:56 <dhellmann> I'll try to eavesdrop 15:39:01 <eglynn> cool 15:39:09 <jd__> btw what about eglynn proposal to core? 15:39:21 <jd__> are we good on that? 15:39:27 <dhellmann> that was one of the 3 things I wanted to bring up 15:39:39 <dhellmann> we voted on the list, do we need anything more formal than that? 15:39:46 <jd__> sorry, didn't mean to steal your points :o) 15:39:48 <dhellmann> or just an action item for nijaba to flip the bit 15:40:00 <dhellmann> heh 15:40:01 <jd__> I think the delay is 5 days without -1 15:40:04 <nijaba> dhellmann: I think the later 15:40:13 <jd__> so we should be almost good anyway 15:40:14 <nijaba> jd__: ah... 15:40:28 <jd__> I don't think anybody will dare -1 ! :) 15:40:51 <eglynn> hope not anyway ;) 15:40:58 <jd__> process is at http://wiki.openstack.org/Governance/Approved/CoreDevProcess 15:40:59 <nijaba> #action nijaba to add eglynn to ceilometer drivers on satureday if all goes well 15:41:19 <nijaba> saturday, too... 15:41:19 <dhellmann> I would also like to propose asalkeld for core 15:41:33 <nijaba> dhellmann: same process? via ml? 15:41:39 <dhellmann> ah, right, ok 15:41:54 <jd__> dhellmann: copy/paste the mail I sent for eglynn 15:42:03 <nijaba> "When a developer feels he has met the criteria and would like to apply for Core Developer status, He/She should send an email to the main OpenStack mailing list requesting to become a Core Developer for a specific sub-project. " 15:42:03 <dhellmann> jd__: reading my mind 15:42:06 <jd__> I meant to do it, but prefered to wait for eglynn to be done first :) 15:42:13 <nijaba> "We will use lazy consensus for the approval vote from the current sub-project core developers. The lazy consensus process shall last five days after the email is sent to the mailing list. During the lazy consensus period the applicant must get approval responses from either 5 active Core Developers or a simple majority of the active Core Developers, whichever is smaller, with no rejection votes. " 15:42:54 <dhellmann> ok, I'll send a proposal email 15:43:01 <nijaba> dhellmann: thanks 15:43:02 <jeblair> o/ (sorry, got dragged into another discussion; any CI questions i can answer?) 15:43:24 <dhellmann> jeblair: we were talking about version numbering before, and your name came up related to that 15:43:25 <nijaba> jeblair: nope, I was just relating the discussion we had about having to restart gerrit 15:43:38 <dhellmann> sorry, that 15:43:41 <nijaba> jeblair: not for versioning, but for naming 15:43:51 <nijaba> dhellmann: :) 15:44:12 <jeblair> i think there is another project that wants a name change 15:44:19 <dhellmann> ok, the other thing I wanted to ask for was a review of the api auth fix in https://review.openstack.org/#/c/15133/ 15:44:46 <dhellmann> that needs one more +2 15:44:51 <jeblair> openstack-common -> oslo-incubation or something, so i think we want to batch that. let's check on the timing for that, and hopefully execute in a couple of weeks. 15:45:21 <nijaba> jeblair: thanks! 15:46:08 <nijaba> dhellmann: will do asap. Sorry, a bit busy with 3 week on the road. coming back to normal next week 15:46:58 <nijaba> anything else? 15:47:05 <sandywalsh> yes 15:47:07 <dhellmann> nijaba: I understand, and thanks 15:47:21 <eglynn> one other thing, how do we regenerate http://ceilometer.readthedocs.org/en/latest/install.html in the light of https://review.openstack.org/14831 ? 15:47:27 <eglynn> (zykes- and davidkranz both tripped over the notification_driver mis-direct in the last while ...) 15:47:49 <sandywalsh> as I mentioned, I'm nearly done the bp on common code for StackTach/Ceilometer/etc ... hope to wrap it up today. Really looking forward to feedback on it. 15:47:51 <dhellmann> eglynn: I need to push those changes to my github repo, which is tied to readthedocs 15:47:52 <nijaba> eglynn: it is not generated, is just an rst in the source 15:48:09 <eglynn> dhellmann: a-ha, cool 15:48:13 <nijaba> dhellmann: isn't this automatic? 15:48:14 <dhellmann> #action dhellmann update readthedocs copy of our docs 15:48:30 <dhellmann> nijaba: it's automatic when I push to my private repo, because I don't have permission to turn on the hook on the shared repo 15:48:32 <eglynn> sandywalsh looking forward to reading that! 15:48:53 <dhellmann> sandywalsh: me, too 15:49:10 <nijaba> dhellmann: infra team would certainly have that rigght... 15:49:25 <dhellmann> nijaba: and now that we're incubated, I can ask them to do things like this :-) 15:49:31 <anniec> sandywalsh: will make sure y! open stackers too 15:49:38 <dhellmann> nijaba: maybe we can coordinate with/after the name migration 15:49:38 <sandywalsh> cool ... hope you guys had a chance to look at the stacktack video? 15:49:48 <nijaba> dhellmann: sounds good 15:49:56 <dhellmann> sandywalsh: not yet, have a link for the logs? 15:50:00 <sandywalsh> http://www.youtube.com/watch?v=pZgwDHZ3wm0 15:50:16 <sandywalsh> latest code is up on github 15:51:17 <eglynn> step-thru' videos are great for driving adoption, we should consider one for ceilo also 15:51:28 <eglynn> (one the grizzly features satrt shaping up ...) 15:51:37 <eglynn> s/satrt/start/ 15:51:59 <nijaba> eglynn: great idea. I guess jd__ and I could work on that together 15:52:05 <nijaba> jd__: up to it? 15:52:21 <jd__> sure 15:52:23 <eglynn> cool! 15:52:26 <davidkranz> Is the API stable enough to start adding some simple tempest integration tests? 15:52:28 <dhellmann> nice 15:52:43 <dhellmann> davidkranz: probably not 15:52:53 <nijaba> eglynn: hope you don't mind if the video comes with a frnech accent ;) 15:53:03 <dhellmann> davidkranz: soon, though, I hope 15:53:15 <davidkranz> dhellmann: Ok. I think that is an important part of openstack process so just let me know when. 15:53:15 <dhellmann> nijaba: subtitles? 15:53:16 <eglynn> all the better, add some Euro-focus ;) 15:53:17 * dhellmann ducks 15:53:26 <nijaba> dhellmann: accessibility rules! 15:53:32 <dhellmann> davidkranz: agreed! 15:54:20 <nijaba> #action jd__ and nijaba to start preparing a video demo of ceilometer 15:54:55 * jd__ laughed at "subtitles?" 15:55:18 * nijaba laughed first, then thought about hearing impaired 15:55:41 <nijaba> anything else? 15:55:52 * nijaba starts the 30s count down? 15:55:58 * nijaba starts the 30s count down!! 15:55:59 * dhellmann has a room full of hungry coworkers who want to go to lunch 15:56:08 <m4dcoder> i have a question about data backend... 15:56:20 * nijaba throws virtual peanuts 15:56:33 <nijaba> m4dcoder: sure. shoot 15:56:34 <m4dcoder> for ceilometer, any direction to push data to NoSQL backend? 15:56:52 <nijaba> m4dcoder: we have one backend implemented: mongo 15:56:52 <eglynn> other than MongoDB? 15:57:15 <nijaba> m4dcoder: but the abstraction would allow you to do anything else you want... 15:57:23 <jd__> anyone can come with a new storage engine 15:57:52 <nijaba> jtran is finishing the SQLAlchemy one 15:58:30 * nijaba resets the 30s count down 15:58:31 <m4dcoder> that answers my question. thx. i just started catching up on this project. so still figuring this out. 15:59:19 <nijaba> m4dcoder: feel free to join #openstack-metering for other questions you may have 15:59:23 <nijaba> #endmeeting