22:00:40 #startmeeting containers 22:00:41 Meeting started Tue Dec 30 22:00:40 2014 UTC and is due to finish in 60 minutes. The chair is adrian_otto. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:00:42 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 22:00:44 The meeting name has been set to 'containers' 22:00:45 @link https://wiki.openstack.org/wiki/Meetings/Containers#Agenda_for_2014-12-30_2200_UTC Our Agenda 22:00:54 #link https://wiki.openstack.org/wiki/Meetings/Containers#Agenda_for_2014-12-30_2200_UTC our Agenda 22:00:59 #topic Roll Call 22:01:01 Adrian Otto 22:01:18 Hongbin Lu 22:01:39 welcome hongbin 22:01:49 :) 22:02:20 this is again a holiday week in the US, so I'm expecting thin attendance 22:02:21 Yuanying OTSUKA 22:02:37 welcome yuanying-alt 22:05:29 #topic Announements 22:06:00 This week I proposed ootsuka as a core reviewers 22:06:17 a majority of cores voted +1, and none aganst. 22:06:21 Thank you very much! 22:06:38 I will be closing that thread and formally adding you as a core today. 22:07:10 other contributors also interested in this role may contact me, and I would be happy to offer guidance. 22:07:31 so thank you yuanying-alt for your work 22:07:42 Will do my best 22:07:54 any other announcements from team members? 22:08:14 #topic Blueprint/Bug Review 22:08:43 I have prepared a set of topics for discussion. I'd like to get some input here, and follow up on ML threads for any that seem like they may be controversial at all 22:08:55 FIrst: Should we open a bug ticket for a refactor of context handling? 22:09:02 #link https://review.openstack.org/142682 Fix context is not set correctly at pod controller 22:09:31 yes i agree 22:09:59 so should we open a bug against this for refactoring, or should we simply move on 22:10:35 I have been opening a number of bugs against reviews for the purpose of keeping commit velocity up, while keeping track of areas where we can improve later (technical debt) 22:10:58 my question is if we want to include this in the debt backlog or not. yuanying-alt, your thoughts? 22:11:50 I think we should report a bug for these purpose. 22:12:21 ok, would you be willing to open a bug at http://bugs.launchpad.net/magnum for this purpose? 22:12:54 yes 22:12:58 it should simply include a pointer to the code we want to draw attention to, state our reason for wanting an improvement, and optionally ideas for how to implement. 22:13:20 implementation details are not required for a bug to be accepted. 22:13:35 ok, thanks. I'll look forward to seeing that when you have time to work on it. 22:14:04 Next example in this topic is: 22:14:05 Should we open a bug ticket for a refactor of rpcapi attribute in request objects for access by controllers? 22:14:12 #link https://review.openstack.org/143935 Add hooks to obtain conductor api 22:15:00 sdake indicated that there is a better way to do this that was less like a hack 22:15:35 if you agree, perhaps you can open a bug against this as well. I don't feel strongly about it, so I'm simply raising this for discussion purposes. 22:16:22 I think what sdake meant was that piling more things into the request object is inefficient, but he did not say exactly what his objection was. 22:16:52 and without considering any alternative options, I refrain from taking a position on the subject. 22:17:14 yuanying-alt: ideas on this? 22:18:22 In this case, If he suggest simply way to handle request in review page, 22:18:49 I fixed it. 22:19:18 ok 22:20:01 Generally, refactor needs bug report I think. 22:20:54 we also have two use cases on the agenda today 22:21:06 #link https://review.openstack.org/144209 Added container network use case to specs 22:21:44 this one got comments from me and sdake supporting the addition, with some discussion about further refinement. 22:22:21 does anyone feel that this is too ambitious, or that it should be excluded as a use case for any reason? 22:23:28 hongbin, I think the rewording I asked for is a minor thing. Did you see it that way? 22:23:43 yes I do 22:23:47 I will fix that 22:24:09 ok, thanks, I gave this one lots of consideration, and I think it's a good addition. 22:24:14 the next one: 22:24:24 #link https://review.openstack.org/144210 Added multi-region/multi-cloud use case to specs 22:24:42 this use case is a real big one, and both sdake and I coughed a bit 22:24:56 because it is certainly ambitious 22:25:19 we each suggested that we could accept is as a directional statement, but not a form project commitment 22:25:25 s/form/firm/ 22:25:34 yes, I am fine with that 22:25:43 this means that to the extent possible we would design Magnum so that it could be later adapted to suit this use case 22:26:36 but if we were faced with a hard decision that might become a problem for this use case, that we might decide to depart from this 22:26:46 but having it in the spec at all gives us a discussion point 22:27:12 so we would be encouraged to discuss the matter if we ever come to a point of conflict 22:27:43 fair enough 22:27:46 ok, so hongbin, I think we can resolve this by just creating a new section in the spec as part of this proposal for future direction. 22:27:55 and it could merge that way 22:28:08 ok 22:28:21 I will modify it accordingly 22:28:28 and when we reach a point of maturity where we feel more confident about addressing it as a real use case, we can revisit it again. 22:28:50 thanks! 22:29:09 any other work items (Blueprints or bugs/tasks) that require team discussion? 22:29:30 #topic Open Discussion 22:29:52 we are also free to adjourn early if there is nothing to discuss 22:30:36 I have nothing 22:30:46 me either 22:31:21 ok, thanks! Our next meeting it 2015-01-06 at 1600 UTC 22:31:24 I will add a bug report for implementation of heat-show. 22:31:32 but later work. 22:31:38 have a Happy New year, and I'll see you then. 22:31:44 #endmeeting