16:01:19 <andymccr> #startmeeting OpenStack-Ansible
16:01:20 <openstack> Meeting started Thu Sep 22 16:01:19 2016 UTC and is due to finish in 60 minutes.  The chair is andymccr. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:21 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:24 <openstack> The meeting name has been set to 'openstack_ansible'
16:01:24 <spotz> o/
16:01:28 <mattt> \o
16:01:56 <michaelgugino> O/
16:01:59 <johnsom> o/
16:02:05 <evrardjp> o/
16:02:10 <palendae> 'lo
16:02:42 <jmccrory> o/
16:03:02 <andymccr> hey all, lets get this going!
16:03:34 <asettle> o/
16:03:37 <automagically> o/
16:03:42 <andymccr> #topic Newton-3 has been released. Onward to RC1.
16:03:57 <andymccr> #link https://etherpad.openstack.org/p/osa-newton-rc1
16:04:36 <odyssey4me> ok, while I know we're working on various things - I think we should just RC based on what we have pinned now
16:04:47 <odyssey4me> we don't have to branch from RC1
16:05:08 <stevelle> you want to delay branching still though?
16:05:16 <odyssey4me> any objections to that approach?
16:05:44 <automagically> none here
16:05:44 <odyssey4me> good question - thoughts?
16:08:03 <andymccr> id say we delay branching then, we have quite a bit in progress around testing and ansible 2.1.1 which, whilst it doesnt affect the functionality much right now, its not that uniform
16:08:33 <odyssey4me> branching now would mean we'd need to backport the changes in flight which we want included in Newton
16:09:05 <automagically> I’m good either way
16:09:14 <odyssey4me> so, personally I think it's worth issuing an RC for help testers use what we have now to find more bugs - but delaying branching until we're more confident that we have a solid release
16:09:31 <andymccr> i agree with that approach
16:09:41 <jmccrory> sounds good
16:09:44 <automagically> makes sense
16:10:09 <odyssey4me> cool, will go with that
16:10:17 <andymccr> ok cool any other issues with newton-3 etc that anybody wants to discuss?
16:10:52 <andymccr> #topic Octavia and OpenStack-Ansible
16:11:37 <automagically> We have more on that topic this week or is it a straggler from last week?
16:11:57 <andymccr> hmm perhaps - i missed last week to be fair, so its possible we didnt remove it properly!
16:12:08 <andymccr> i'll clean that up!
16:12:19 <andymccr> #topic Release Planning and Decisions
16:12:25 <johnsom> I joined to answer any additional questions, but I don't think we had any open topics
16:12:39 <odyssey4me> this is release week - any specific blockers for a liberty/mitaka release?
16:13:57 <andymccr> johnsom: thanks! we didnt' have anything listed at least, so hopefully its all ok at the moment
16:14:00 <andymccr> odyssey4me: i think we're good
16:15:02 <odyssey4me> alright, I'll push up a release request after the meeting
16:15:32 <andymccr> i think thats all so opening up for open discussion
16:15:36 <andymccr> #topic Open Discussion
16:16:01 <automagically> I want to thank odyssey4me for his tireless work as PTL to date and point out https://github.com/openstack/election/blob/master/candidates/ocata/OpenStackAnsible/andymccr.txt
16:16:19 <automagically> Thanks for offering your talents andymccr
16:16:43 <odyssey4me> :) welcome andymccr!
16:16:59 <andymccr> lets see how it goes! but looking forward to next cycle, even though its a short one.
16:17:02 <odyssey4me> and thanks automagically - it's been a pleasure, for real reals
16:17:31 <palendae> https://review.openstack.org/#/q/(topic:inventory-export-hosts+OR+topic:inventory-debug)+project:openstack/openstack-ansible - some inventory patches that will be useful if people have time to give them a review
16:17:38 <evrardjp> thanks indeed to both of you, for different reasons :)
16:17:43 <odyssey4me> I'd like to thank the core team and all our contributors for helping making Newton the awesome release that it is
16:17:50 <automagically> palendae: Thx for the reminder, I’ll give em another pass
16:17:53 <asettle> odyssey4me: for 'real reals' okay man, we've sat next to each other for *too* long
16:18:05 <asettle> I mean, YAY andymccr
16:18:08 <asettle> \o/
16:18:11 <michaelgugino> congrats to andymccr
16:18:16 <odyssey4me> fo' shizzle, my nizzle asettle
16:18:19 <asettle> odyssey4me: oh bro
16:18:38 <andymccr> ok anybody else got anything to raise? otherwise i think we're good to go back to fixing all the thigns!
16:18:57 <automagically> I raised a question about our usage of tempest-horizon in the os_horizon gate earlier
16:19:12 <odyssey4me> thanks palendae for reverse engineering the dynamic inventory and making it better :)
16:19:24 <asettle> andymccr: could we add a check in topic for OSA install guide testing next week please?
16:19:29 <automagically> Issue is that it appears to not support turning off SSL cert validation which is causing the Xenial gate failure for the Ansible 2.1.1 os_horizon work
16:19:32 <andymccr> asettle: will do
16:19:36 <odyssey4me> automagically oh yeah
16:19:39 <evrardjp> asettle: just add it to the agenda :p
16:19:44 <automagically> Wondering how folks would like to proceed
16:19:45 <odyssey4me> automagically can we turn SSL off for the os_horizon role test?
16:20:00 <asettle> Thanks evrardjp - I always lose the agenda. I'm useless :p
16:20:14 <odyssey4me> asettle it's in the IRC channel topic :p
16:20:14 <evrardjp> asettle: I'll teach you how to fish
16:20:22 <asettle> OMG WOT
16:20:25 <asettle> This explains *much*
16:20:30 <andymccr> automagically: hmm.
16:20:34 <automagically> odyssey4me: I _think_ we can toggle the external_ssl var, which will prevent the self-signed cert gen
16:20:39 <automagically> But, may cause other issues
16:21:18 <automagically> The question really, is what do you all think is the ideal way forward, and should we take a short term approach if the ideal has a longer timeframe?
16:21:39 <andymccr> automagically: so is the fix in horizon itself?
16:21:41 <andymccr> (ideall)
16:21:42 <automagically> I’d suggest that the ideal approach is to get a patch merged in tempest-horizon so we can continue as we have been
16:21:50 <odyssey4me> automagically well, I guess the question is what are we trying to test
16:22:22 <andymccr> is there an upstream bug with tempest-horizon already?
16:22:27 <automagically> odyssey4me: Fair point, the scenario basically just validates a horizon auth and some page views
16:22:29 <odyssey4me> it would certainly be nice to be able to thoroughly test horizon in the integrated build, so yes - an upstream patch to fix the plugin would be great
16:22:38 <automagically> And I can’t imagine anyone wanting to run Horizon without SSL
16:22:50 <automagically> andymccr: None yet that I could find
16:23:01 <odyssey4me> for now we'll have to work around it, as an upstream patch could take a long time to get in
16:23:07 <automagically> Righto, agreed
16:23:20 <automagically> I’ll twiddle with the external_ssl toggle to see if that gets us through gate on our side
16:23:23 <evrardjp> it's still worth it
16:23:27 <odyssey4me> automagically also, we happen to know the current and next cycle PTL :)
16:23:31 <automagically> And file a bug, raise a patch with tempest-horizon as needed
16:23:44 <automagically> ORLY, who dat?
16:24:18 <odyssey4me> current PTL is robcresswell, a Racker - I believe that r1ch4rdj0n3s is going for PTL next cycle
16:24:35 <asettle> Rob is Cisco
16:24:40 <odyssey4me> whoops - rob is cisco, richard is a Racker
16:25:01 <automagically> Gotcha, cool
16:25:10 <automagically> Interestingly, looks like Tempest has no candidate
16:25:13 <odyssey4me> anyway - we have relationships there and can facilitate a discussion
16:25:28 <odyssey4me> does that surprise you? :)
16:25:36 <andymccr> ok sounds like we're doing ok!
16:26:13 <automagically> Appreciate folks weighing in with thoughts
16:26:14 <stevelle> automagically: looking for more work? :)
16:26:27 <automagically> stevelle: Whaddya got?
16:26:39 <stevelle> just in case you wanted to head Tempest
16:26:48 <automagically> Aw hell no...
16:26:55 <evrardjp> :D
16:26:58 <andymccr> next cycle maybe!
16:27:02 <odyssey4me> there isn't enough whiskey in the world to fix that
16:27:23 <stevelle> I just assume automagically will want your job when you're done with it andymccr
16:27:35 <automagically> We’ll see about that
16:27:40 <stevelle> :)
16:27:46 <odyssey4me> head hunting already - sjoe give the guy a break
16:27:57 <andymccr> automagically is in demand
16:28:13 <andymccr> ok! well thanks all
16:28:16 <andymccr> i think thats it for today
16:28:18 <andymccr> so.
16:28:19 <andymccr> #endmeeting