14:02:55 <dhellmann> #startmeeting releaseteam
14:02:55 <openstack> Meeting started Fri Oct 14 14:02:55 2016 UTC and is due to finish in 60 minutes.  The chair is dhellmann. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:02:57 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:02:59 <openstack> The meeting name has been set to 'releaseteam'
14:03:08 <dhellmann> courtesy ping : ttx, dims, stevemar, fungi
14:03:18 <ttx> o/
14:03:33 <ttx> As always I padded the agenda
14:03:41 <dhellmann> yes, thanks for that
14:04:32 <dhellmann> ok, call done
14:04:39 <dhellmann> #topic status update on trailing projects
14:05:06 <dhellmann> I sent out a proposed set of deadlines late last week or early this week
14:05:31 <dhellmann> #link http://lists.openstack.org/pipermail/openstack-dev/2016-October/105266.html
14:05:42 <ttx> dahsboard shows multiple RC2s
14:05:52 <dhellmann> IIRC, the consensus was that those dates worked for everyone
14:06:07 <ttx> branches up except for kolla ?
14:06:32 <ttx> and unclear status for fuel
14:06:38 <dhellmann> yes, and I think the kolla team was going to be ready
14:06:46 <dims> ttx : i have been neglecting to update the dashboard
14:06:56 <dhellmann> fuel had a patch up to get tags but without any actual release jobs it wouldn't pass validation so I added a bypass mode for them
14:07:15 <ttx> I was suspecting something like that, that is why I asked
14:07:22 <dhellmann> I don't see that patch up for review any more, so I guess it was approved
14:07:26 * dhellmann got a late start this morning
14:07:40 <dhellmann> #link https://review.openstack.org/386211
14:07:50 <ttx> ok, looks like you have a good handle on this
14:08:10 <dhellmann> I see a WIP TripleO RC3
14:08:23 <dims> dhellmann : yes, me and tony pushed that along
14:08:27 <dims> (fuel)
14:08:38 <dhellmann> dims : thanks
14:08:50 <dims> dhellmann : am waiting for the fuel team to generate the artifacts before sending a announcement
14:08:56 <dhellmann> I see we're up to RC2 at least for all projects according to https://releases.openstack.org/newton/index.html
14:08:58 <dims> (per ashtokolov)
14:09:18 <dhellmann> dims : ack, good idea. I don't know where we'll send folks to download those things.
14:09:36 <dhellmann> do they distribute via launchpad?
14:09:40 <dims> dhellmann : right there's scroll back on our irc channel with links
14:09:42 <dhellmann> or a mirantis.com site?
14:09:43 <dhellmann> k
14:10:06 <dims> [08:34:17]  <ashtokolov>	dims: we will publish packages here http://packages.fuel-infra.org/repositories/
14:10:18 <dhellmann> ah, good
14:10:20 <dhellmann> thanks, dims
14:10:38 <ttx> next topic ? Trying to keep meeting under 30 min :)
14:10:39 <dhellmann> so I think we're on track with these projects. does anyone else have a different impression?
14:10:45 <dhellmann> k
14:10:45 <ttx> nope, sounds all good
14:10:51 <dhellmann> #topic mistral pypi upload
14:11:02 <ttx> I summarized the question on the etherpad
14:11:13 <dhellmann> is 3 their current version?
14:11:19 <ttx> 3.0.1 even
14:11:35 <dhellmann> yeah
14:11:55 <dhellmann> I wish we would stop including source for projects from git
14:11:59 <ttx> We could add the publish-to-pypi bit to their release file, and do a manual upload of 3.0.1 ?
14:12:02 <dhellmann> s/including/importing
14:12:51 <ttx> dhellmann: do we, here ?
14:12:59 <dhellmann> ?
14:13:12 <ttx> dhellmann: do we import mistral from git
14:13:25 * ttx sees mistral!=2015.1.0,>=2.0.0
14:13:36 <dhellmann> looking
14:14:08 <dhellmann> global requirements has mistral 2.0.0
14:14:17 <dhellmann> this is pretty weird
14:14:28 <ttx> in this case the issue is that some mistral code is imported as a library by some projects
14:14:30 <dhellmann> maybe EmilienM has some more info, I'll ask him, since it's tripleo
14:14:44 <ttx> they are working on a mistral-lib so that you import less
14:14:52 <dhellmann> #link http://paste.openstack.org/show/585713/
14:14:55 <ttx> but in the mean time need mistral on Pypi to solve that requirements.txt
14:15:13 <dhellmann> so 2.0.0 was never uploaded?
14:15:13 <ttx> currently it's importing 3.0.0.0b1 instead of 3.0.1
14:15:37 <EmilienM> o/
14:15:39 <dhellmann> I guess that means tripleo isn't using constraints? because the constraints show ==2.0.0
14:15:58 <ttx> hmm, maybe it skips b1
14:16:40 <dhellmann> EmilienM : we're trying to figure out what's going on with users of mistral as a lib
14:16:40 <dhellmann> ttx: yeah, old versions of mistral are on pypi, just "hidden" so the beta version is skipped
14:16:48 <dhellmann> I just ran "pip install mistral" and got a 2015 version
14:16:56 <ttx> ew
14:17:13 <dhellmann> well, i suppose if I filtered I'd get 2.0.x
14:17:15 <ttx> ok, so how about we add the publish-to-pypi bit to their release file, and do a manual upload of 3.0.1 ? oir a 3.0.2
14:17:46 <dhellmann> I guess? if only tripleo is using it and they don't care about pip maybe we don't need it?
14:18:07 <ttx> do they not care ? Someone complained at least
14:18:08 <dhellmann> and I'd rather tag a new version on the same commit than manually upload an existing version #lazy
14:18:22 <dhellmann> ah, I didn't see a complaint
14:18:32 <ttx> d0ugal on the release channel
14:18:39 * d0ugal waves
14:18:45 <ttx> he was expecting to see 3.0.0 on Pypi
14:18:46 <dhellmann> sorry, I'm not prepared this morning, I didn't get home until about 2 last night and had an early meeting
14:19:10 <dhellmann> d0ugal : did you need mistral 3.x for newton or ocata?
14:19:15 <d0ugal> dhellmann: we care as far as unit tests go
14:19:34 <dhellmann> k
14:19:44 <ttx> feels like release 3.0.2 with publish-to-pypi is the rigth way around this
14:19:47 <dhellmann> ttx: do you want to propose a 3.0.2 tag
14:19:52 <ttx> I can do that
14:19:54 <dhellmann> yeah, if we need it we need it
14:20:05 <dhellmann> we should get the mistral ptl to sign off, too
14:20:11 * ttx parallelizes
14:20:12 <dhellmann> rakhmerov
14:20:33 <dhellmann> moving on while ttx runs in a separate thread...
14:20:40 <dhellmann> #topic releases for vmware-nsxlib
14:21:11 <dhellmann> the vmware team asked us to manage a release, but the repo is not part of an official team. I don't remember if we set a policy on that or not.
14:21:20 <dhellmann> #link https://review.openstack.org/#/c/384487/
14:21:33 <dhellmann> it's a lot less work now, so we could just do it
14:21:52 <dhellmann> otoh, I think that would make the results show up on the releases site
14:22:18 <dhellmann> yes: http://docs-draft.openstack.org/87/384487/1/check/gate-releases-docs-ubuntu-xenial/8b2df0a//doc/build/html/teams/vmware_nsx.html
14:22:32 <ttx> dhellmann: hmm wait... is include-pypi-link sufficient to trigger the upload, or do we need to add jobs too
14:23:00 <dhellmann> ttx: oh, right, we'd need jobs. that flag only triggers behavior in the announce email
14:23:18 <ttx> ok then I'm probably not able to do that today
14:23:23 <dhellmann> ok, I'll take that
14:23:52 <ttx> I can propose the release, but better Workflow-1 it
14:24:11 <dhellmann> do we have any other deliverables for non-official teams?
14:24:25 <dhellmann> ttx: ack, I'll wip it and then make it depends-on the jobs patch when I write that
14:25:31 <dhellmann> so I'm inclined for now to ask the vmware team to handle their own release. thoughts?
14:25:56 <ttx> dhellmann: please Workflow-1 https://review.openstack.org/386640
14:25:58 <dhellmann> oh, and that's what garyk said in his most recent comment
14:26:28 <ttx> dhellmann: I don't think we did any
14:26:48 <dhellmann> ok, so that's settled
14:26:48 <ttx> puts us on a slippery slope
14:27:00 <dhellmann> #topic barcelona agenda
14:27:09 <ttx> we have no place in openstack/releases for it anyway
14:27:37 <dhellmann> yes, true
14:27:40 <dhellmann> #link https://etherpad.openstack.org/p/ocata-relmgt-plan
14:28:07 <dhellmann> ttx, did you have something more than ^^ to bring up related to the agenda?
14:29:00 <ttx> not really
14:29:10 <ttx> trying to see if there is anythign specific we should prepare for
14:29:19 <ttx> and that I missed scanning stuff
14:29:56 <dhellmann> I'd like to put together a priority list for the things we need to discuss with infra in our work session. Maybe we can each take a pass through that plan doc and add INFRA to the lines that need to be covered, then I can pull that list out to make it easier to process in the meeting
14:30:28 * dims back in a bit
14:30:43 <dhellmann> we're hitting ttx's 30 minute limit
14:30:50 <dhellmann> #topic skipping next 2 meetings
14:31:15 <dhellmann> yes, I'll be traveling next Friday and we'll be in BCN together the next Friday so I'll announce that we won't have meetings until after the summit
14:31:18 <ttx> ack ack
14:31:21 <dhellmann> #topic open discussion
14:31:23 <dhellmann> anything else today?
14:31:43 <ttx> nope
14:31:45 <dims> +1 on skip
14:31:49 <EmilienM> I have a question
14:31:50 <dims> nothing from me
14:31:53 <EmilienM> very quick
14:31:57 <dhellmann> EmilienM : go for it
14:32:22 <EmilienM> today i'm proposing tripleo rc3, but some projects might need rc4 by 18th. is it fine?
14:32:55 <dhellmann> EmilienM : if you *know* you're going to need another candidate, it's better to wait a little bit to reduce the number of overall candidates you produce
14:33:18 <EmilienM> ok so we defer rc3 again and do it before 18th. Ok?
14:33:23 <dhellmann> but if we need to tag a candidate to make the downstream packaging machinery work for testing, we can
14:33:40 <EmilienM> dhellmann: i'll checkt that info in a minute
14:33:44 <dhellmann> yes, as long as it's done by the 18th it's ok
14:34:10 <dhellmann> ok, let us know and remove the WIP from the patch when you're ready
14:34:23 <EmilienM> ack thx. I'm done.
14:34:25 <dhellmann> k
14:34:28 <dhellmann> anything else?
14:34:51 <dhellmann> ok, I think we can close the meeting and move back to #openstack-release
14:34:55 <dhellmann> thanks everyone!
14:34:58 <dhellmann> #endmeeting