15:00:02 #startmeeting climate 15:00:03 Meeting started Fri Mar 28 15:00:02 2014 UTC and is due to finish in 60 minutes. The chair is DinaBelova. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:04 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:06 The meeting name has been set to 'climate' 15:00:16 o/ 15:00:16 Climate folks - you here? :) 15:00:19 \o 15:00:21 cool 15:00:48 casanch1, cmart, cmart_, dpamio, f_rossigneux, mkaminsk, pafuent, scroiset_, tomek_adamczewsk? 15:00:56 o/ 15:01:00 Swann won't be here, sorry 15:01:05 pafuent, o/ 15:01:12 Hi there 15:01:15 o/ 15:01:22 \o folks 15:01:23 i'm here 15:01:27 coolio 15:01:40 mkaminsk, I guess tomek_adamczewsk is your collegue 15:01:44 will he be here? 15:01:54 f_rossigneux, you here? 15:02:15 don't know for f_rossigneux 15:02:26 yes, me too :) 15:02:28 well 15:02:33 I think I'll start 15:02:40 today's agenda is 15:02:40 https://wiki.openstack.org/wiki/Meetings/Climate 15:02:41 yes, tomek_adamczewsk is my collegue and will be in one moment 15:02:57 tomek_adamczewsk, o/ 15:03:08 Hi all 15:03:25 first of all - let me introduce our new contributors) 15:03:44 mkaminsk and tomek_adamczewsk will help us working on PR use case) 15:03:59 and I'm really glad :) 15:04:16 these folks are from Poland :) 15:04:24 welcome, mkaminsk and tomek_adamczewsk :) 15:04:27 so our european tram is growing :) 15:04:37 hi for everyone :) 15:04:42 mkaminsk: tomek_adamczewsk: welcome 15:04:44 well - it was team, not tram) 15:04:56 welcome 15:05:03 anyway, as usually, I'll start with action items 15:05:08 #topic Action items from the last meeting + reviews queries 15:05:17 #link http://eavesdrop.openstack.org/meetings/climate/2014/climate.2014-03-21-15.00.html 15:05:33 so let's go through all of them by person 15:05:39 bauzas, let's start with you 15:05:48 sure 15:06:00 reviews are in progress 15:06:07 I had no time to do the etherpad 15:06:19 ok, so I'll add this action item to you today 15:06:24 sure thing 15:06:35 #action bauzas create etherpad about new designs and amend according BPs and share links via ML 15:06:43 ok, next one is me) 15:07:02 we have good progress with tempest testing with cmart) 15:07:10 is _cmart here? 15:07:26 cmart_ :S 15:07:31 wow, noone) 15:07:36 rare thing 15:07:48 he's not at his desk :( 15:08:00 but anyway - Christian published first change for that tempest BP 15:08:15 and I guess we'll merge it soon - just after minor fixes 15:08:22 and all comments fixed 15:08:47 yes, some nits to be done yet 15:08:52 as for Barbican folks I had no opportunity to check if they need maintenance for gerrit while their moving 15:09:01 so I'll pick that action item for me one more time 15:09:10 #action DinaBelova check one more time if Barbican folks need some maintenance for gerrit while their moving 15:09:20 I've filled form to participate in #vbrownbag Tech Talks 15:09:27 cool 15:09:32 and I had small conversation with Michael Still 15:09:33 who will be at summit ? 15:09:37 about reasons for rejecting of Climate talk 15:09:44 oh cool 15:09:47 bauzas, one moment 15:09:55 sadly, he's is in trip now 15:10:01 with rare unternet connections 15:10:16 so he'll answer me in mail when he'll had opportunity to do that 15:10:29 have you all seen the agenda for OSS related projects ? 15:10:35 currently I still dunno if reason was in lack of voices or any other reasons 15:10:58 bauzas, I did take a look on summit schedule 15:11:07 #link http://openstacksummitmay2014atlanta.sched.org/overview/type/related+oss+projects 15:12:01 well, quite narrow, but interesting topics 15:12:08 well, I don't want to point out someone, but I'm a bit frustrating to see again some talks we have at Icehouse and Havana summits... 15:12:20 well... 15:12:22 like if it was something automatic 15:12:27 sometimes that happens 15:12:31 anywat 15:12:33 anyway 15:12:37 and we can do nothing with it 15:12:51 let's take the opportunity to use the vbrownbag session 15:13:14 as for Swann - he's in process of doing / just done his action items 15:13:26 and it was action item - pafuent publish fix for PR + trusts 15:13:29 yep 15:13:31 Yes 15:13:31 pafuent? :) 15:13:38 I'm still working on it 15:13:49 well, may you provide draft at least? 15:13:56 for us start looking on ut? 15:13:57 it* 15:13:58 I think is almost done. At this moment I'm doing some manual testing 15:14:08 I think 15:14:28 #action pafuent publish fix (at least draft for core-team) for PR + trusts 15:14:38 I'll really appreciate you'll do that today 15:14:54 pafuent, are you ok with it? 15:15:00 Ok, I'll try to do it 15:15:11 ok, great 15:15:16 Yes, I found that PR plugin doesn't work with PostgreSQL 15:15:29 I'm double checking that in order to fill a bug 15:15:30 pafuent, hm 15:15:34 ok 15:15:35 ? 15:15:46 I'm not really sure 15:15:51 ok, would be interesting to know the reason 15:15:58 as we leverage SQLA 15:15:58 Returning exception (ProgrammingError) operator does not exist: character varying = integer 15:15:58 LINE 4: WHERE computehosts.id = 1 15:15:59 bauzas, +1 15:16:05 oh 15:16:11 ni-i-ice 15:16:20 ok, we'll look on bug 15:16:20 The id is an uuid and in nova is an integer 15:16:34 Seems that PostgreSQL doesn't like that 15:16:42 well, anyway 15:16:49 MySQL seems to be more permissive 15:16:49 let's go further 15:16:50 ok, let's see your proposal then :) 15:17:01 #toppic Volume reservation concept 15:17:04 agrh 15:17:09 #topic Volume reservation concept 15:17:18 #link https://blueprints.launchpad.net/climate/+spec/basic-volume-plugin 15:17:36 casanch1, may you tell us current status for that? 15:17:37 that's me :) 15:17:46 I know ;) 15:18:02 I have just posted an etherpad in the BP whiteboard 15:18:09 cinder does support extensions 15:18:13 #link https://etherpad.openstack.org/p/climate-bp-basic-volume-plugin 15:18:32 so we will need to create a repo climate-cinder 15:18:40 to capture the API calls 15:18:59 so, for lease creations it is more or less ok, we can follow the same approach done for Nova 15:19:15 but... what should happen when the lease for a volume starts? 15:19:16 well, it looks so... as we have no opportunity to keep imports from nova/cinder in one repo due to oslo duplicate option hell 15:19:39 oh, one more thing: cinder API does not have hints 15:19:47 hm 15:19:50 not nice 15:19:54 but it supports arbitrary metadata 15:20:01 sounds like there is need to commit on cinder :) 15:20:18 well, I guess metadata is more than good 15:20:25 maybe we could use this metadata to send lease info 15:20:35 but... this will be stored in cinder db 15:20:46 and of course in Climate db when the lease is created 15:21:03 I think metadata is ok for now 15:21:09 in cinder db it'll be only lease-id, i guess 15:21:16 not all info really 15:21:41 casanch1: will you be at summit ? 15:21:47 yes I will 15:21:53 bauzas - casanch1 and pafuent 15:21:53 because I think we need to discuss on this at summit time 15:22:05 that's a really wide discussion 15:22:11 well, I propose to prepare ideas/drafts before) 15:22:12 ok 15:22:18 on etherpad yes 15:22:19 so casanch1 is working on it now) 15:22:24 the etherpad is created IIRC 15:22:38 bauzas ^^ - https://etherpad.openstack.org/p/climate-bp-basic-volume-plugin 15:22:53 ok, so we'll continue discussion there 15:22:55 my biggest question is what will happen when the lease starts... 15:22:56 yep 15:23:13 casanch1, we'll try to define that in etherpad 15:23:18 ok 15:23:29 #action everyone propose ideas in https://etherpad.openstack.org/p/climate-bp-basic-volume-plugin 15:23:38 * DinaBelova going further 15:23:42 casanch1: I personally think that the call should be placed from climate at first 15:23:45 like for PR 15:23:54 #topic Energy efficiency BP status (if Francois will come to meeting) 15:23:59 ok 15:24:12 bauzas, that will be done only after new lease-creation architecture 15:24:23 so we'll need your documents first ;) 15:24:29 as for energy efficiency 15:24:41 I had funny conversation with Francois :) 15:24:53 f_rossigneux, you here? 15:25:11 well, anyway 15:25:44 DinaBelova: what's the result of this discussion ? :D 15:25:46 he really started working with it - the only moment was that he did not discuss all that stuff with others Climate team. 15:25:47 trailers are good :) 15:25:57 I asked him to write letter to ML 15:26:01 sure thing 15:26:07 but well - he said he'll do that 15:26:09 he needs to share his thoughts 15:26:12 but no letter I see 15:26:26 for example. he created repo in his own github account 15:26:28 https://github.com/frossigneux/kwstandby 15:26:43 to start working on ipmi stuff and so one 15:27:19 as i really did not think it'll be other repo (and definitely not in his github accout) I asked him to share ideas and current status 15:27:22 well, I can't say more than... 15:27:28 will continue pinging him 15:27:41 as currently I'm a little bit out of context 15:27:45 he was really busy before 15:27:56 that's why i thought he simply did not start yes 15:27:58 let's consider that if he doesn't follow the process, we give -2 to his changes 15:27:59 yet* 15:28:13 well, he's quite new to open source 15:28:16 well, he should have time now 15:28:24 so i prefer more gentle -1 15:28:25 anyway 15:28:26 he has 1 yr background :) 15:28:34 it looks like no 15:28:38 anyway 15:28:49 let's not discussing him without him here 15:29:07 sure thing 15:29:10 I'll do Francois will share his current status and ideas 15:29:17 and we'll discuss that public 15:29:39 well, Im moving on 15:29:48 #topic Open discussion 15:30:12 folks, do you have quesiotns to discuss? 15:30:16 questions* 15:30:17 I have just one 15:30:22 you're welcome 15:30:46 bauzas, could you please give +2 back to this: https://review.openstack.org/#/c/82192/ 15:30:57 casanch1: ok 15:31:00 casanch1, I love the way you say it ;) 15:31:05 :) 15:31:16 thank you bauzas 15:31:18 #info casanch1 will have vacation 2 next weeks 15:31:25 am I right? 15:31:35 yes 15:31:39 for info as well, I'm moving from one position to another 15:31:40 the following 2 weeks 15:31:57 with a focus area on Nova mainly now 15:32:12 hence my currently dotting support 15:32:13 bauzas, may I info that about your new company? 15:32:20 for us to congrat you? 15:32:26 well, not yet :) 15:32:30 ok :) 15:32:45 I'll still continue to support Climate of course 15:32:53 but with less focus than previously 15:33:00 I'll still focus on reviewing code 15:33:05 but with less commits 15:33:10 ok, got it 15:33:24 that leads to my question 15:33:28 ok? 15:33:43 we had an election 4 months ago ^_^ 15:33:49 and yes 15:34:20 I was wondering if we were planning to do another set of elections for the PTL position ? 15:34:20 we need to have #1 new PTL election and #2 take a look on current core team 15:34:34 well, I'll propose my candidate) 15:34:41 if there will be no others - well 15:34:48 possibly it's not useful here 15:34:54 do you have ideas? 15:35:17 I won't propose myself for obvious reasons :) 15:35:32 ok, so are we going to have one more election? 15:35:38 think so 15:35:59 let's ask our lovely sherpa to cornac this election :) 15:36:05 SergeyLukjanov ^^ 15:36:13 >> sherpa to cornac ???? 15:36:40 http://en.wikipedia.org/wiki/Sherpa_people 15:36:43 #info we need to organize new PTL elections 15:36:50 bauzas, DinaBelova, I'm ready to do it 15:36:57 SergeyLukjanov, cool 15:37:09 so we'll have 1 week for candidates self proposal 15:37:18 and 1 week for voting itself 15:37:22 smth like that? 15:37:27 as for the other projects 15:37:34 see Anita e-mail 15:37:50 so, you'd like to adjust elections with official programs schedule? 15:38:08 * March 28 - April 4, 05:59 UTC: Open candidacy to PTL positions * April 4 - April 11: PTL elections 15:38:15 I would love to do it, yes 15:38:24 unless there is a constraint, of course 15:38:30 okay, I'll send an email today 15:38:36 any objections? 15:38:41 we can vote :) 15:38:48 ok 15:39:15 #info start PTL elections with official programs schedule - * March 28 - April 4, 05:59 UTC: Open candidacy to PTL positions * April 4 - April 11: PTL elections 15:39:27 ok, cool 15:39:34 smth else here to discuss? 15:40:00 https://review.openstack.org/#/c/76830/ 15:40:08 and yes 15:40:27 It's the scroiset_ patch 15:40:43 yup ? 15:40:53 I really wanted to ask Swann how he tested that) but anyway - I'll do some manual testing and find out if i can change my +1 to +2 15:40:58 will try to do that today 15:41:12 bauzas, pafuent need that change 15:41:22 for his one with PR and trusts 15:41:28 needs* 15:41:31 He changed the NovaBlaBla to not use singletons, which for me is ok (I did that change in my pathc) 15:41:39 s/pathc/patch 15:41:52 and it'll be cool we'll merge it 15:41:54 really 15:42:00 yup, understood the need 15:42:17 #action DinaBelova, bauzas - review https://review.openstack.org/#/c/76830/ 15:42:21 ok 15:42:25 Thanks 15:42:30 I did test this review 15:42:36 I can already give a +2 15:42:42 it was a little changed 15:43:06 like no singleton -> it night cause issues with PR i guess... 15:43:15 well, ok 15:43:19 bauzas: Could you share how you test it? Because I need to do the same for my patch 15:43:21 although I guess it won't 15:43:34 you have to source the code by yourself 15:44:13 bauzas: I want to know how to test the PR plugin 15:44:28 that requires to setup a PR evn 15:44:32 pafuent, well, simply we need to test if it'll work ok with two original use cases (PR + VR) with different user/project combinations 15:44:45 as novaclient will use different tokens there 15:44:52 bauzas: I don't know If the steps that I'm doing are sufficient 15:44:53 DinaBelova: +1 15:45:14 pafuent, please list these steps :) 15:45:33 I'm creating the host, then creating a lease 15:46:42 I could not find a complete use case 15:46:46 you should check also if 1/ VMs are going to reserved hosts when lease starts 2/ updating lease (like end date changing) 3/ find out what happens when PR ends 15:47:03 pafuent: both steps are good 15:47:51 How I assign VMs to the host? 15:48:20 by passing --hint reservation= while VM creation 15:48:21 by passing a scheduler hint 15:48:26 +1 15:48:40 Ok 15:48:41 you may find it in nova-client repo 15:48:46 I mean filter itself 15:49:01 one moment 15:49:39 https://github.com/stackforge/climate-nova/blob/master/climatenova/scheduler/filters/climate_filter.py#L74 15:49:57 and there is nice doc in that method 15:50:10 well - smth else? 15:50:33 well, bye :) 15:50:39 Regarding https://review.openstack.org/#/c/79399/ 15:50:39 have a nice weekend - all of you) 15:50:42 oh 15:50:56 bye :) 15:51:00 dpamio - yes? 15:51:14 just to clarify, bauzas, shall I move the #noqa fix to another bug? 15:51:44 it looks like bauzas is proposing to move all changes in that file to separated one 15:51:53 ok, will do 15:52:00 ok, cool 15:52:03 just wanted to check. 15:52:12 ok, bye! 15:52:13 thnks 15:52:15 :) 15:52:18 #endmeeting