09:59:39 #startmeeting climate 09:59:40 Meeting started Tue Dec 17 09:59:39 2013 UTC and is due to finish in 60 minutes. The chair is DinaBelova. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:59:41 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:59:43 The meeting name has been set to 'climate' 09:59:49 hola 09:59:57 o/ 10:00:01 (again :D) 10:00:07 yeap 10:00:11 Me and Sergey are also here 10:00:24 f_rossigneux_, scroiset_? 10:00:33 Hi 10:00:40 Ok, nice 10:00:51 scroiset_, are you here? 10:00:55 scroiset is still on paternity leave :) 10:01:04 Ok, forgot about it 10:01:08 let's start then 10:01:10 sure 10:01:23 https://wiki.openstack.org/wiki/Meetings/Climate#Agenda_for_December_17_2013 10:01:33 it's our agenda 10:01:35 for today 10:01:40 pretty busy agenda :) 10:01:44 we have to be quick :) 10:01:56 #topic Action items from the last meeting + reviews queries 10:01:59 esp. there are quite chatty topics 10:02:20 As for the actioan items from last session 10:02:23 #link http://eavesdrop.openstack.org/meetings/climate/2013/climate.2013-12-09-20.01.html 10:02:37 Let's move through them quickly 10:02:48 Nick, you had two of them 10:02:55 I can't get the webpage, so I will only speak by mind 10:03:06 bauzas, ok 10:03:23 Nikolay_1t, as I see you have delivered new patchsets as discussed 10:03:26 great 10:03:34 yeap 10:03:44 ok, any reviews needed ? 10:03:44 and also for vm_plugin 10:03:49 As for my action item, I also reviewed it https://review.openstack.org/#/c/57200/ 10:03:55 saw it 10:03:55 bauzas: yes, please 10:04:13 ok, which ones to be prioritized ? 10:04:21 today we'll test all this stuff about vms 10:04:24 well 10:04:28 all of them ? :D 10:04:34 bauzas, they are connected with each other 10:04:38 you were interested in openstack utils 10:04:44 sure, will do again 10:04:50 the last won't work without first two :) 10:04:50 my concern is just about time :) 10:05:04 bauzas, they all are needed for the release 0.1 10:05:08 because I will be on holidays starting end of this week, until Jan 4 10:05:29 bauzas, we need to try review all of them 10:05:32 as yours ones 10:05:36 utils-->trust-->plugin 10:05:40 ok, will do 10:05:48 because in the other case we need to postpone release 10:06:00 yup, that's something we need to discuss afterwards 10:06:16 due to the holidays in FR and RU it sounds impossible to release 0.1 before the NY 10:06:23 bauzas, you had two action items 10:06:30 SergeyLukjanov, looks so for me too 10:06:31 so, it sounds reasonable to move release to the mid-end Jan 10:06:31 ... 10:06:39 ... 10:06:42 what are your vacancies ? 10:06:53 SergeyLukjanov, maybe that's a good point 10:07:07 me and Dina have 2w vacations around the NY 10:07:09 ok, let's discuss that on the next topic item 10:07:16 ok 10:07:20 so for my action items 10:07:29 bauzas, you have two of them 10:07:31 I had to fix a blueprint 10:07:37 it's done 10:07:44 great 10:07:53 sorry, can't give you the link :( 10:07:54 the next one was Agree on delivery date 10:08:07 don't worry, we'll find it 10:08:13 yup, and as said, we're focusing on delivering the merge code by this week 10:08:19 the second one was about when you 10:08:34 ... will have all needed for hosts reservation implemented 10:08:40 so I would say the prototype would be there, but it would still require reviews 10:08:52 bauzas, ok, nice 10:09:07 we have the folowing open reviews now 10:09:10 #link https://review.openstack.org/#/q/status:open+project:stackforge/climate,n,z 10:09:43 changes by Nikolay_1t are essential for the VM reservation 10:10:01 we're still missing a Climate filter for hosts reservations, but that's only for preventing other requests to be spawn on the hosts 10:10:19 and we also need to hack the client 10:10:26 and we have 3 changes by bauzas and one by f_rossigneux_ 10:10:32 bauzas, as for client 10:10:38 I propose to merge first commit 10:10:44 after some testing 10:10:54 because we have large patchset really... 10:10:56 sure 10:11:01 too large... 10:11:10 which one is huge ? 10:11:11 * SergeyLukjanov thinking about the pros/cons of weekly reviews sync 10:11:24 client CR ? 10:11:29 in climate it's Francois's one 10:11:37 and in client it's my 10:11:52 I'll try to find way of splitting for both of them 10:12:02 I think we should move to next topic 10:12:05 sorry, can't get the point 10:12:18 bauzas, they are too huge to be quick reviewed 10:12:23 ah ok 10:12:38 I think we should move to next topic, because we do not have much time 10:12:40 My patchet could be splitted in two parts: the reservation plugin and the reservation algos. 10:12:56 f_rossigneux_, would be nice if you'll do it 10:13:01 Host Admin Manager and Physical Host Reservation are the only ones scoped for 0.1 10:13:11 because it's almost impossible to review huve amount of code quickly 10:13:15 there will be a 3rd one 10:13:23 huge* 10:13:26 bauzas, ok 10:13:29 the review will be the merge of all the code 10:13:58 bauzas, can't get your point 10:14:02 sorry 10:14:27 I will provide another CR for merging both Francois and mine' code 10:14:43 hm... 10:14:49 they are quite big 10:14:53 themselves 10:14:58 yup 10:15:02 if there is possibility to stay for them separated 10:15:06 yep 10:15:07 it would be nice 10:15:20 to review them in more comfortable way 10:15:23 ok? 10:15:29 I think we have discussed also the second topic somehow :) 10:15:30 that's why I'm doing a merging CR, only for managing the few discrepancies 10:15:42 but let's discuss that offline 10:15:47 ok 10:16:04 because the automatic merge is failing on some patches 10:16:12 I think we may move to the Openstack new Program or not? 10:16:22 because the second topic was also discussed 10:16:29 Is that ok for you? 10:16:33 which date do we plan ? 10:16:59 Jan 21/24 should work 10:17:03 bauzas, I think that should be second half of the Jan 10:17:04 agree 10:17:08 because of all these holidays 10:17:11 ok 10:17:11 21/23 10:17:18 ok 10:17:27 DinaBelova: could you please raise that point ? 10:17:31 #agreed New release date Jan 21/23 10:17:39 ok 10:17:42 %(meeting_chair) please add info item 10:17:48 see it ;) 10:18:08 let's move to the Openstack new Program or not? https://wiki.openstack.org/wiki/Governance/NewPrograms 10:18:10 ok? 10:18:14 yup 10:18:16 sure, new topic ? 10:18:17 : 10:18:18 #topic Openstack new Program or not? https://wiki.openstack.org/wiki/Governance/NewPrograms 10:18:18 :D 10:18:31 ok, as we want to be incubated somehow 10:18:40 we need to think about our program 10:18:43 so, I don't feel quite comfortable with this 10:18:52 there are some existing ones 10:18:55 #link https://wiki.openstack.org/wiki/Programs 10:18:57 that's something which is unclear, even for TC 10:19:04 we are a new team 10:19:09 it looks like there is now existing program fully compatible with project's scope/plans 10:19:18 s/now/no ? 10:19:25 yep 10:19:29 *no* 10:19:30 bauzas, yes 10:19:45 there is no such program 10:19:46 So I propose to heve new program like "Resource Reservation" 10:19:54 or something like this 10:20:01 or just reservations 10:20:04 if we agree with that, we would need to ask for a mission statement 10:20:10 SergeyLukjanov, yes 10:20:31 bauzas, so we need to create document, describing our project, our view on it 10:20:31 I think that we should start working on Incubation application without sending it 10:20:38 SergeyLukjanov, 10:20:39 +1 10:20:43 just to be sure that everything is ok 10:20:45 just one comment, our current usecases could all fit under the Compute umbrella :) 10:20:54 not really... 10:21:16 bauzas, there are plans for supporting all other services/resources for reservations 10:21:20 well, reserving Nova resources could be seen as a feature of Nova :) 10:21:25 like stacks, networks and etc. 10:21:35 floating IPs are Nova 10:21:39 no 10:21:44 we need to provision routers 10:21:51 bauzas, we have plans about new resources - Neutron's, Heat's 10:21:55 IPs is neutron 10:22:12 and we were talking about reserving of storage nodes 10:22:15 and so on 10:22:18 yup 10:22:22 so that's not about Nova 10:22:22 on last summit they said that it's a good point to manage network resources using Neutron 10:22:29 Compute, sorry*** 10:22:31 I'm just speaking as Devil lawyer :) 10:22:36 bauzas, 10:22:37 :D 10:22:49 ok, so agreed on creating new prorram 10:22:53 because if we apply as new Program, there will be some contradictors 10:22:59 #agreed ? :D 10:23:06 bauzas, sorry 10:23:13 lost your disagreements 10:23:23 what kind of contradictors? 10:23:34 there are still no programs with more than one project in it :) 10:23:38 nah nah, I'm just saying I'm OK, but we need to prepare our mission statement 10:23:46 bauzas, absolutely! 10:24:05 provisioning Neutron routers should be explicit 10:24:06 except maybe TripleO + Tuskar, but I don't know much about their relationship 10:24:33 SergeyLukjanov: I could speak about them, but we would be running out of time :) 10:24:45 #agreed Prepare new program's description and start working on incubation proposal with mission, etc. 10:24:55 okay 10:25:05 these discussions should be public thanks to the ML 10:25:05 writing new program application + incubation application will help us to analyze the gaps in scope/etc. 10:25:18 SergeyLukjanov, +1 10:25:24 bauzas, of course 10:25:39 are you ok with moving to the next topic? 10:25:41 ok, once we agreed that, we need to go to the next point 10:25:53 because there is a consequence 10:25:59 #topic PTL election questions 10:26:10 I've asked Sergey to help us 10:26:19 because he has much more experience :) 10:26:20 jd__ could also help 10:26:26 bauzas, of course 10:26:31 there are many options on how to make it, but I think that the right approach for climate is to use the common OS process 10:26:37 sure 10:26:41 1w for proposing candidates 10:26:45 a Condorcet election 10:26:48 self-propositions I mean 10:26:55 + 1w for elections 10:26:59 yup http://civs.cs.cornell.edu 10:27:03 yup, we need to formalize that 10:27:04 #info http://civs.cs.cornell.edu 10:27:13 there is a link for this 10:27:15 hold on 10:27:26 yup, process is well defined 10:27:46 that's on wiki 10:28:07 we just need to propose the process and vote for it 10:28:32 bauzas, I can do it for the next meeting 10:28:39 we've done the same in Savanna 10:28:48 and I can find old etherpads 10:28:50 SergeyLukjanov, yes, I saw that 10:29:04 https://wiki.openstack.org/wiki/Election_Officiating_Guidelines 10:29:05 that's basically only matter of giving the wikipage, and asking for +1/-1 ? 10:29:07 some info ^^ 10:29:22 so, we agree on the proposal, and then we go thru it 10:29:46 I think that the process is well defined, so, we just to agree on that by voting on the next meeting for example 10:29:59 then wait foor candidates for a week 10:30:06 exactly 10:30:11 and then setup condorset elections 10:30:14 that's what I was trying to sat :) 10:30:16 say 10:30:17 and then have 1week lonf elections 10:30:30 I think that I now several questions 10:30:40 it will be a little hard due to the vacancies 10:30:41 that we'll need to discuss 10:31:08 SergeyLukjanov, what are they? 10:31:13 bauzas, what do you mean? 10:31:26 1. for which time range we'll elect ptl 10:31:37 2. how we'll collect ATCs list 10:31:37 +1 10:31:52 okay 10:32:01 1. 6 months sound okay to me 10:32:11 so, for the #1 icehouse will be the best option 10:32:18 to be aligned with openstack elections 10:32:26 well, that's a good point 10:32:38 in savanna we've elected me for the end of current and next cycle 10:32:46 it works too :) 10:32:58 so we are speaking about time till the end of Icehouse release 10:33:01 am I right? 10:33:04 yup 10:33:07 4 months 10:33:12 and then we revote 10:33:15 If we'll complete elections in mid-end Jan than it'll the mid Icehouse 10:33:24 looks like that it's ok 10:33:30 ok, nice 10:33:33 let's setup voting right now? 10:33:52 Nikolay_1t, are you ok with that? 10:33:53 it'll be better to have voting results for all our decisions 10:34:00 SergeyLukjanov, +1 10:34:19 +1 for icehouse PTL 10:34:27 +1 for Icehouse timeframe 10:35:00 #startvote Elect PTL for the Icehouse timeframe? Yes,No 10:35:01 Begin voting on: Elect PTL for the Icehouse timeframe? Valid vote options are Yes, No. 10:35:02 Vote using '#vote OPTION'. Only your last vote counts. 10:35:09 #vote Yes 10:35:14 #vote Yes 10:35:14 #vote Yes 10:35:24 #vote Yes 10:35:26 #vote Yes 10:36:15 ok, we can close the vote :) 10:36:17 looks like that's enough :) 10:36:21 #endvote 10:36:22 Voted on "Elect PTL for the Icehouse timeframe?" Results are 10:36:23 Yes (5): bauzas, f_rossigneux_, DinaBelova, SergeyLukjanov, Nikolay_1t 10:36:28 yay! 10:36:31 :) 10:36:45 the next question is about choosing electorate 10:36:51 so the next point was about how we'll collect ATCs list 10:36:52 ok, next question was : how we can find Climate ATCs ? 10:37:01 the common practice is take all commiters 10:37:07 eactly 10:37:15 yeap, that's good 10:37:23 but I have a concern that there was some initial code that was obsolete and removed afaiu 10:37:26 but the question is for what period 10:37:26 do we also take current reviews ? 10:37:38 well 10:37:39 Icehouse? Or whenever> 10:37:41 why not? 10:37:42 ? 10:37:49 bauzas, in OpenStack only existing commits used 10:37:49 on review I think 10:37:54 :( 10:37:58 okay 10:38:11 * SergeyLukjanov looking for the doc 10:38:20 well, briefly checking at the git log 10:38:31 That's for the Icehouse http://stackalytics.com/?release=icehouse&metric=commits&project_type=stackforge&module=climate&company=&user_id= 10:38:45 I have some scripts to build ATC list 10:38:45 let's vote then 10:38:49 That's for all periods http://stackalytics.com/?release=all&metric=commits&project_type=stackforge&module=climate&company=&user_id= 10:39:10 are there any initial thoughts to exclude initial climate code? 10:39:27 I don't think we have to exclude 10:39:49 because any committed code is good for climate 10:39:58 even if that's useless now 10:40:03 but it was replaced, isn't it? 10:40:05 that would send a bad signal 10:40:22 doesn't really matter for me 10:40:30 and that doesn't change 10:40:34 the list of ATCs 10:40:48 but that would only be a terrible mistake in terms of communication 10:40:51 Except otherwise-noted in the program description, the electorate for a given program PTL election are the Foundation individual members that are also committers for one of the program projects over the Grizzly-Havana timeframe 10:41:12 generally speaking, a 2 cycle-period is a timeframe for electing ATCs 10:41:18 yup 10:41:19 As i looked, the list of ATC will be the same anyway 10:41:28 ok 10:41:33 so, let's vote to confirm it 10:41:40 ok 10:42:25 #startvote Use 2 OpenStack release cycles as a timeframe for electing ATCs? Yes,No 10:42:26 Begin voting on: Use 2 OpenStack release cycles as a timeframe for electing ATCs? Valid vote options are Yes, No. 10:42:27 Vote using '#vote OPTION'. Only your last vote counts. 10:42:41 #vote Yes 10:42:44 #vote Yes 10:42:47 #vote Yes 10:42:48 #vote Yes 10:42:55 #vote Yes 10:43:03 #endvote 10:43:04 Voted on "Use 2 OpenStack release cycles as a timeframe for electing ATCs?" Results are 10:43:05 Yes (5): bauzas, f_rossigneux_, DinaBelova, SergeyLukjanov, Nikolay_1t 10:43:10 ok 10:43:11 okay 10:43:12 ok 10:43:24 do we have any other topics to discuss? 10:43:31 a big one :( 10:43:32 except elections 10:43:40 Cafe 10:43:42 I think for this topic that's it 10:43:47 let's move to the next 10:43:50 I need some time to think about other elections-related stuff 10:44:07 SergeyLukjanov, okay 10:44:13 #action SergeyLukjanov to prepare aggregated doc about Climate PTL elections 10:44:17 #topic Cafe project https://wiki.openstack.org/wiki/Cafe 10:44:28 how did you find them ? 10:44:34 Okay, today I found one email 10:44:38 because I can't find any code on Stackforge 10:44:41 #link http://www.mail-archive.com/openstack-dev@lists.openstack.org/msg11857.html 10:44:53 because they have not shared it yet 10:45:02 just from mailing list, yes 10:45:08 But their initiative was after ours 10:45:18 And Nick and I wrote them to look on Climate 10:45:27 missed it ?! 10:45:35 bauzas, looks so... 10:46:02 ok, anyway, let's wait for feedback 10:46:12 that's something we need to know 10:46:19 because they definitely have overlaps with us.... 10:46:32 what's bad is that they did prototype without checking what's existing 10:46:33 overlaps, or climate do the same? 10:46:45 SergeyLukjanov, not really a copy 10:46:53 but the most of things - yes 10:47:12 ok, then maybe that's good news then :) 10:47:30 the only thing they have different from us - they propose the users creating mechanism in the Cafe 10:47:31 because if they have new usecases, then they could contribute to Climate 10:47:33 it'll be really cool to have one more team and usecases pack 10:47:39 +1 10:47:44 +1 10:47:49 so let's wait their reaction 10:47:53 sure 10:48:07 anyway, we're pretty well implemented now 10:48:16 so I'm not so worried 10:48:28 and the thing is they did not share their code proposal... 10:48:37 btw today will be tc meeting with the final barbician incubation discussions, so, it could be useful for you guys 10:48:41 so we know nothing about what do they relly implemented 10:49:02 Climate is already known by TC 10:49:09 SergeyLukjanov, let's move it now to the open discussion 10:49:20 bauzas, that's about incubaion process 10:49:28 yup, followed ity 10:49:28 #topic Open discussion 10:49:55 bauzas, I'm speaking not about talking about climate but listening about how other guys trying to become incubated ;) 10:50:01 ok, so I think we may take a look on this process closer :) 10:50:05 SergeyLukjanov, yep 10:50:09 SergeyLukjanov: yup yup 10:50:10 They had problems 10:50:21 we may prevent them on our case 10:50:27 SergeyLukjanov: I'm just saying that Climate is already known 10:50:32 :) 10:50:34 contrary to Cafe 10:50:38 yup and it's good 10:51:00 and that's cool that Tim B. points to climate in response to the cafe proposal 10:51:06 well, about holidays, maybe we should just share our periods 10:51:15