09:30:53 <prometheanfire> #startmeeting requirements
09:30:54 <openstack> Meeting started Wed Apr  5 09:30:53 2017 UTC and is due to finish in 60 minutes.  The chair is prometheanfire. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:30:55 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:30:57 <openstack> The meeting name has been set to 'requirements'
09:31:08 <prometheanfire> !topic rollcall
09:31:09 <openstack> prometheanfire: Error: "topic" is not a valid command.
09:31:14 <prometheanfire> #topic rollcall
09:31:38 <prometheanfire> toabctl: tonyb dirk dims ?
09:32:19 <dirk> o/
09:32:24 <toabctl> hi
09:34:31 <prometheanfire> k
09:34:41 <prometheanfire> #topic controversies in the queue?
09:34:58 <dirk> eventlet I guess
09:35:17 <prometheanfire> I know there's eventlet thing
09:35:18 <prometheanfire> ya
09:35:35 <dirk> the discussion on the mailing list was inconclusive to me
09:35:46 <prometheanfire> have a link? it's not showing up in gertty (unless it was abandoned)
09:35:48 <dirk> so I'm surprised to see dims proposing to not do the eventlet revert and instead drop designate
09:36:00 <dirk> #link https://review.openstack.org/#/c/448104/
09:36:01 <dirk> and
09:36:02 <prometheanfire> ya, saw that on the list too
09:36:13 <dirk> #link https://review.openstack.org/#/c/453279/
09:36:28 <dirk> I think what we should do is revert eventlet, extend test coverage to cover designate
09:36:37 <dirk> and then try to work together with them on getting that fixed
09:36:58 <dirk> the other thing that happened with e.g. networking-odl is that they merged a local overwrite to the global upper-constraints
09:37:09 <dirk> so they are running their tests against a different upper-constraints file than "ours"
09:37:11 <prometheanfire> I'd rather just fix designate, but not sure how hard that'd be
09:37:12 <dirk> which is bad as well
09:37:38 <dirk> it probably isn't that hard
09:37:40 <prometheanfire> has anyone talked to them about it?
09:37:57 <dirk> I didn't
09:38:15 <prometheanfire> nor I
09:38:21 <dirk> so we have a number of options
09:38:26 <dirk> a) just fix the thing
09:38:32 <dirk> b) give them a local overwrite
09:38:43 <dirk> c) globally revert and work with them on re-bumping it
09:38:49 <dirk> d) dropping them from g-r sync
09:38:55 <prometheanfire> with eventlet?
09:39:22 <dirk> I think a and c are good options, b and d less so
09:41:27 <prometheanfire> agreed
09:42:24 <prometheanfire> a if we can, c if we work with them and it's a bigger / longer issue, d if they are dead (upstream wise), don't think I like b at all
09:42:31 <prometheanfire> in that order that's what I'd like
09:44:45 <prometheanfire> anyway, I'll keep an eye on it as I can, but am on vacation for a week.
09:44:48 <prometheanfire> so...
09:44:59 <dirk> the action item is to discuss this with dims
09:45:08 <dirk> he seems to have a different opinion and we need to understand his reasoning
09:45:14 <prometheanfire> ya, when we can
09:51:50 <prometheanfire> anything else in the mean time?
09:52:18 <prometheanfire> want to wait til this would normally start to give people a chance to show
10:00:30 <prometheanfire> looks like dims got back to us in the other channel :D
10:01:03 <dims> prometheanfire : haven't woken up fully yet. back in a bit
10:01:12 <prometheanfire> dims: how long is a bit? :P
10:02:44 <dirk> prometheanfire: meeting time is the other favorite topic
10:02:57 <dirk> I also have the topic of upgrading the nova-func test job to voting
10:03:44 <prometheanfire> well, let's do the nova-func first
10:04:38 <prometheanfire> I've notice you mention it a couple of times :P
10:06:56 <prometheanfire> ?
10:07:38 <dirk> basically, short summary: we discussed it in atlanta to extend the list of check jobs for upper constraints changes also by running the projects functional tests
10:07:47 <dirk> the guinea pig for that was nova and it seems to work, but it is currently non-voting
10:07:54 <dirk> which in the end led to things being merged that broke it
10:08:02 <dirk> so I would like to propagate it to voting
10:08:05 <dirk> any objections?
10:08:18 <dirk> I'd also think about adding further func test jobs then, as a 2nd step
10:08:34 <prometheanfire> I'm good with making it voting
10:09:29 <prometheanfire> if it's regularly passing and adds value to our testing and is not an external job, make it vote :D
10:10:23 <prometheanfire> and the meeting time?
10:10:31 <prometheanfire> I'm fine with an hour earlier
10:16:59 <dirk> +1
10:17:12 <dirk> it seems that would be incompatible with dims
10:17:18 <dirk> and he's a frequent contributor recently
10:17:25 <dirk> dims: your input appreciated now :)
10:18:10 <dims> +1 to make it voting
10:19:59 <prometheanfire> dims: and the meeting time? is an hour ago too early?
10:22:53 <prometheanfire> also, is upstream desigante really dead?
10:23:09 <prometheanfire> or dead enough to be removed?
10:23:44 <prometheanfire> if so, a post to the list about it should probably be made, before removal
10:26:11 <prometheanfire> going to close this in a min if no response, we can discuss this stuff later, I'm dead tired
10:26:37 <dims> prometheanfire : it's 6:30 AM for me right now :)
10:26:47 <dims> prometheanfire : yes it is
10:27:17 <prometheanfire> it's 5:30 for me
10:27:24 <dims> prometheanfire : did you see monty and me posting to ML about designate?
10:27:26 <prometheanfire> and I've been up since yesterday :P
10:27:32 <dims> LOL
10:27:36 <dims> go get some sleep :)
10:27:47 <dims> i need to wake up kids for school
10:28:15 <prometheanfire> oh, guess they are tagged in that thread
10:29:22 <dirk> prometheanfire: ok, lets end the meeting
10:29:30 <dirk> and have some sleep
10:29:57 <prometheanfire> #endmeeting