15:01:41 <ttx> #startmeeting releaseteam
15:01:42 <openstack> Meeting started Fri Mar 17 15:01:41 2017 UTC and is due to finish in 60 minutes.  The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:43 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:45 <openstack> The meeting name has been set to 'releaseteam'
15:02:02 <ttx> courtesy ping dims, sigmavirus, tonyb, stevemar
15:02:15 <ttx> Agenda at https://etherpad.openstack.org/p/pike-relmgt-tracking
15:02:24 <ttx> (in R-24 week)
15:02:35 <ttx> #topic Pike plan status updates
15:02:48 <ttx> I wanted to update the tracking
15:02:58 <ttx> I think this can be considered done: Consolidate the release tools that use the releases repository data in the releases repository (dhellmann)
15:03:14 <ttx> Same for 'Port all release tools to work under python 3 (dims, dhellmann, others can help too)' ?
15:03:22 <dims> o/
15:03:33 <dhellmann> a big portion of the first is done; I'll look into whether it's fully complete
15:03:49 <dhellmann> for the second, we still need to update the tools in the project-config repo
15:03:54 <ttx> oh, ok
15:03:57 <dhellmann> at least one of those relies on python2, though
15:04:02 <ttx> let's keep that open then
15:04:04 <dims> i didn't end up doing much here for this
15:04:17 <dhellmann> the launchpad updater
15:04:26 <ttx> Next I wanted us to mark "PIKE-1" tasks that are assigned to you that you intend to complete by then
15:04:45 <ttx> so feel free to go through the list and tag those
15:05:09 <ttx> dhellmann: we have a common one around asking for a TC resolution... should we try to do that bny Pike-1 ?
15:05:11 <dhellmann> ah, the announcement email stuff still needs to be either moved or ported to python 3 or both
15:05:11 <ttx> by*
15:05:37 <ttx> I have a long backlog of things to push to TC
15:05:43 <ttx> so I wouldn't mind Pike 2
15:05:55 <dhellmann> yeah, I can take a stab at writing a draft of that when I get back from PTO at the end of the month
15:06:10 <ttx> ok, let's target to pike1 then
15:06:58 <ttx> Finally we don't have an assignee for the "support storyboard in openstack/releases" task
15:07:10 <ttx> I could probably tackle it in Pike2
15:07:22 <ttx> but if more uregnt I wouldn't mind people taking it over
15:07:35 <ttx> in case anyone wants to play with StoryBoard API
15:08:01 <ttx> (dhellmann already did the first two steps)
15:08:33 <ttx> Ok, I'll mark it as mine for Pike-2, with a "feel free to take over" mention)
15:08:41 <dhellmann> I was surprised that there was no python-storyboardclient release
15:08:44 <fungi> worth noting we have a few repos migrating from lp to sb on monday
15:08:50 <fungi> monasca, refstack and defcore
15:08:51 <dhellmann> that's going to make the task a little more difficult
15:09:13 <dhellmann> their releases will work, but the stories won't get comments when they are released
15:09:17 <ttx> dhellmann: we could make that happen I think
15:09:35 <ttx> (storyboardclient releases)
15:09:38 <dhellmann> we need to decide what we actually want to have happen there, do we want to try to update the state or just leave a comment?
15:09:50 <dhellmann> if there's no release, is the code even functional?
15:09:56 <ttx> just leave comment for now
15:10:05 <fungi> yeah, i don't think those teams are too worried about the lack of release automation commenting in resolved stories for the moment
15:10:54 <ttx> OK, anything else to report on the pike-plan ?
15:11:22 <fungi> one other point on presence/absence of a client lib for sb... when jeblair wrote boartty he just had it do direct calls to the rest api
15:11:23 <dhellmann> nothing from me
15:11:46 <ttx> #topic Priority reviews
15:11:47 <dhellmann> fungi : yeah, I guess the sb team is focused on the JS client
15:11:58 <ttx> Looks like we are on top of our review queue
15:12:56 <dhellmann> that's the shortest it has been in a while :-)
15:13:01 <ttx> Could use russellb's advice on https://review.openstack.org/#/c/443748/
15:13:09 <dims> yay
15:13:30 <ttx> dhellmann: he mentions a discussion with you on release models
15:13:33 <dhellmann> it sounds like we need to explain why we want them to use the cycle model there
15:13:54 <dhellmann> yeah, russellb and I talked on irc
15:14:06 <dhellmann> or maybe even in our 1:1, now that I think of it
15:14:21 <russellb> yeah we talked briefly, i was convinced enough of the value
15:14:23 <ttx> I don't know the project in question, so i don't want to make assumptions on what is the best model for them, but I suspect Russell has enough view on both sides
15:14:28 <dhellmann> I'll try to compose something and put it in the patch today
15:14:36 <ttx> ok
15:14:37 <russellb> because i don't see the stable branches as taking *that* much extra overhead
15:14:59 <ttx> imho it depends how tied to openstack it is
15:14:59 <russellb> because i'd imagine they sit mostly idle ... it probably won't be that active of a repo
15:15:16 <russellb> from a technology perspective, it's not tied to openstack anymore
15:15:25 <russellb> but openstack is still the vast majority consumer
15:15:33 <russellb> there's one other project that isn't openstack that is using it
15:16:21 <ttx> ok, I'll let you guys elaborate on the review
15:16:26 <dhellmann> the main benefit is that the downstream packagers keep coming back and asking why we're releasing "parts of openstack" that don't have stable branches
15:16:27 <dhellmann> it interferes with the automation they have in place to consume what we produce
15:16:36 <russellb> dhellmann: +1
15:16:53 <ttx> Any other review we need to pay attention to ?
15:16:56 <russellb> consistency so that it's just 1 of many repos that fit a pattern is why i'm good with it
15:17:14 <russellb> anyway, happy to help in follow up ...
15:17:19 <dhellmann> thanks, russellb
15:17:33 <ttx> #topic Open discussion
15:17:59 <ttx> There are reveiws up for setting shade/requestsexceptions model
15:18:08 <ttx> doesn't seem to be in a hurry since it's not approved yet
15:18:30 <dhellmann> we're waiting on the governance change to make that a standalone team
15:18:36 <ttx> also unclear whether they would actually want independent instead of cycle-based
15:18:38 <dhellmann> the argument for using cycle-with-intermediary is the same here
15:18:40 <dims> y was wondering if they want to be independent
15:18:44 <dims> y
15:18:53 <dhellmann> downstream, RH wants those stable branches
15:18:58 <dhellmann> upstream we probably won't use them much
15:19:07 <ttx> ok
15:19:25 <ttx> With dhellmann in PTO next week I propose we skip the meeting
15:19:35 <fungi> i realized earlier this week i need to press forward on swapping the pike development key we created into production, so hope to get that through system-config today and then have the releases site update (with the correct dates) proposed shortly thereafter
15:20:22 <ttx> ok to skip ?
15:20:33 <dhellmann> ttx: I'm out on my next two release rotation days (next week and the 28th)
15:20:43 <dhellmann> skipping is obviously fine with me :-)
15:20:51 <ttx> you're covering... Tuesday or Thursday those days ?
15:20:55 <dhellmann> Tuesday
15:21:01 <ttx> ok, will be extra attention then
15:21:05 <ttx> pay*
15:21:10 <dhellmann> usually Monday and Tuesday, but mainly Tuesday
15:21:29 <ttx> anything else ?
15:21:49 <dhellmann> I didn't have anything to bring up this week
15:22:01 <ttx> ok, let's close early then
15:22:22 <dhellmann> ++
15:22:40 <ttx> Thanks everyone. Have a great weekend. dhellmann, enjoy your time out
15:22:50 <dhellmann> thanks!
15:22:51 <ttx> #endmeeting