10:03:16 #startmeeting requirements 10:03:16 Meeting started Wed Dec 7 10:03:16 2016 UTC and is due to finish in 60 minutes. The chair is tonyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 10:03:18 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 10:03:20 The meeting name has been set to 'requirements' 10:03:28 #topic roll call 10:03:38 o/ 10:03:52 sigmavirus, number80, coolsvap, toabctl 10:04:03 o/ 10:04:12 we have an appology from prometheanfire 10:04:26 hi 10:05:16 Hi, coolsvap toabctl 10:05:20 #topic Any controversies in the Queue? 10:05:43 I think it's pretty clean really 10:06:23 there's a high number of reviews with negative votes from jenkins but I don't see anything problematic in there 10:07:27 https://review.openstack.org/365837 is a little questionable so it'd be good to see feedback on that 10:07:46 anyone else? 10:09:11 * coolsvap is bit behind on the current status 10:09:34 coolsvap: that's fair ... I hope you're getting better 10:09:40 tonyb: yeah, I couldn't make up my mind on this yet 10:09:52 i will put things on reviews if find anything controversial 10:09:57 it certainly overlaps with other uses, I didn't want to take a side there as I'm not unbiased 10:09:58 dirk: fair enough 10:10:20 the only other thing is the posix-ipx removal I guess 10:10:31 I think its blocked on one review, if that doesn't move soon I'd vote for stop caring 10:10:50 dirk: the thing that clinched it for me was if we -1 it then oslo_messaging will end up requireing monasca-statsd which IMO doesn't seem right 10:11:04 dirk: Yeah that's blocked on the blazar stuff 10:11:26 #link https://review.openstack.org/#/q/status:open+project:openstack/blazar+branch:master+topic:feature/goal-remove-incubated-oslo-code 10:11:34 ^^^ that's the "blazar stuff" 10:11:39 tonyb: yeah.. monitoring services it the next hype it seems. way too many things 10:12:33 its not yet visible in the queue but it looks like we'r eprogressing slowly on the docker-py uncap. 10:12:45 haven't looked at where we're stuck today yet though 10:12:57 dirk: cool. IIRC that's blocked by a change in the releases repo 10:13:25 dirk: Hmm actually that seems to have gone through 10:14:20 tonyb: no, we still have the conflict over tripleo-common 10:14:37 https://review.openstack.org/#/c/406443/ 10:14:43 ah, I see tonyb commented on that. I missed that 10:14:48 * dirk will fix 10:15:37 dirk: cool. 10:16:34 dirk: I shoudl have just fixed it, sorry 10:17:01 no worries, I have not yet completely fixed my gerrit email filtering setup.. some mails go to nowhere right now 10:17:42 dirk: Ahh okay. 10:18:04 dirk: sometimes I wish it'd be nice if mine went to /dev/null ;P 10:18:17 #topic Requirements priorities for Ocata 10:18:24 #link https://etherpad.openstack.org/p/requirements-track-constraints-usage 10:18:43 dirk: I see you've done a few more awesome! I'll review them. 10:19:03 I wonder if it's worth automating them 10:19:25 I started looking at that but there are so many differences in the projects.. 10:19:37 I wonder, do we always want to use tools/tox_install.sh ? 10:19:57 or only where its needed? it seems some projects just use pip -c ${magic git url} -U {opts} 10:20:17 directly. I guess tahts okay as long as they don't run into a conflict, right? 10:21:11 dirk: if $project is listed in upper-constratints.txt then they have to use tools/tox_install.sh 10:21:39 so nova can just use the -c {URL} but python-novaclient can't 10:22:08 I see 10:22:16 that makes sense. we should have a check for that 10:22:41 dirk: so *if* grep -i $(python setup.py --name) /path/to/upperconstratints.txt is emapty then it's a quick one liner 10:22:45 probably something for validate-constraints ? 10:23:13 otherwise we need the other oneliner + tools/tox_install.sh 10:23:59 we'll hit problems with some of the neutron-stadium projects and horizon plugins butI have a cunning plan for them also 10:24:17 dirk: I don't follow? validate-constratints? 10:24:49 tonyb: well, whenever we merge reviews that extend upper-constraints we need to check that they use the proper way of restricting upper-constraints on that particular project imho 10:25:01 extend upper-constraints.txt in global requirements repo.. 10:25:38 since in theory additional upper-constraints might be pulled in by version updates somewhere that depend on new stuff.. 10:26:12 dirk: I understand now. Yeah we shoudl probably do that with a tool rather than manual checking 10:28:03 tonyb: lets put it on the task list? I can't commit to the action item right now 10:28:12 dirk: Sure. 10:28:27 the focus needs to be getting u-c supprt up 10:28:47 then we can do more of the divergent requirements work that's actually interesting 10:29:59 Anything else for priorities? 10:30:00 do we have a blueprint or bug for this? I haven't hit it but I guess at some point someone will ask for it 10:30:37 (the uc-handling-unification part I mean, to be referenced in the reviews against all the projects) 10:30:38 dirk: No we don't but I shoudl create one so we have a common place to answer questions 10:30:53 dirk: I shoudl also email os-dev about what we're doing 10:32:06 #action tonyb to create a blueprint and email os-dev to explain the divergent requirements work and how that applies to u-c 10:32:09 well, I guess a launchpad id against all the projects that are affected (or that get reviews) is a good first start to unify communication 10:33:20 dirk: I'm unlikey to create that thing, but a blueprint and a std. gerrit topic + an email will go a long way to doing the same thing 10:35:12 tonyb: wfm 10:35:19 dirk: cool 10:35:26 #topic Open Discussion 10:35:51 anything? 10:37:48 Looks like we're close to wrapping up 10:38:28 I'd just like to say thank you to all the reviewers who are working on this. I'd not fun but it is important 10:38:56 I'd also like to say I'm always open to feedback on what I can do better as PTL 10:40:20 and encourage people to think about PTL spot for Pike. I'm happy to run but if y'all want a change that's also cool 10:40:40 ugh, there are 2 more months for that :) 10:41:47 dirk: sure but if we start thinking now it wont sneak up on us 10:42:19 we should think about the PTG as well 10:42:26 * dirk will be there 10:43:02 dirk: Yeah I'll be there too. We have some "space" during the 2 horizontal days but I don't know what that means 10:45:09 okay if there isn't anything else ... 10:45:40 #endmeeting