16:05:38 <odyssey4me> #startmeeting OpenStack-Ansible
16:05:38 <openstack> Meeting started Thu Sep  1 16:05:38 2016 UTC and is due to finish in 60 minutes.  The chair is odyssey4me. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:05:38 <asettle> o/
16:05:39 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:05:41 <openstack> The meeting name has been set to 'openstack_ansible'
16:05:42 <michaelgugino> hi
16:05:42 <evrardjp> o/
16:05:44 <asettle> o/
16:05:50 <palendae> Hello
16:05:53 <odyssey4me> apologoies for the late start, I was a little distracted
16:06:04 <evrardjp> 5' is fine :)
16:06:22 <evrardjp> as long as you don't do that everytime!
16:06:26 <evrardjp> :p
16:06:48 <BjoernT> odyssey4me: multi threaded meetings :-)
16:07:08 <asettle> evrardjp likes to be punctual
16:07:55 <spotz> o/
16:08:01 <cjloader> hi all
16:08:40 <odyssey4me> alright, let's kick into gear
16:08:52 <odyssey4me> mhayden is out today, so you get me... sorry for that :p
16:09:04 <jmccrory> o/
16:09:13 <odyssey4me> #topic Newton-3 has been released. Onward to RC1.
16:09:29 <odyssey4me> #link https://etherpad.openstack.org/p/osa-newton-rc1
16:09:40 <asettle> Winner winner chicken dinner
16:10:02 <odyssey4me> somehow n3 released with a bug which breaks things, then stevelle found it and fixed it
16:10:12 <odyssey4me> so n3 is not entirely functional, but that's ok
16:10:22 <evrardjp> p
16:10:33 <evrardjp> that's why we have rcs
16:10:42 <odyssey4me> our primary goals are now to find bugs and get upgrades working
16:10:59 <odyssey4me> I'd like input on any known issues so that we can track and resolve them
16:11:32 <odyssey4me> can y'all add notes in the etherpad
16:14:03 <odyssey4me> ok, I added some new bits there
16:14:36 <odyssey4me> do we have any volunteers to take the lead for each unfinished initiative ?
16:14:59 <andymccr> i'll work on centralised testing
16:16:21 <asettle> I'll obvs take the install guide stuff
16:18:32 <nishpatwa> I shasha_tavil and qwang can help you guyz if you need any help
16:19:49 <evrardjp> thanks for the offer nishpatwa :)
16:23:53 <odyssey4me> alright, we can keep that going afterwards obviously
16:24:00 <odyssey4me> add to it as you think of things please
16:24:11 <odyssey4me> let's try and use that to co-ordinate things that need doing
16:24:52 <odyssey4me> we have around 2.5 weeks before all the upstream projects issue RC's, and we must too
16:26:25 <odyssey4me> ok, moving on
16:26:32 <odyssey4me> #topic Liberty -> Mitaka upgrades
16:26:50 <odyssey4me> evrardjp I believe that https://review.openstack.org/360539 is pretty much the last bit for the roles we ship with OSA?
16:26:59 <odyssey4me> and it's a cosmetic fix
16:27:08 <odyssey4me> is that a fair statement?
16:27:39 <evrardjp> fair enough statement for what we ship with osa
16:27:57 <evrardjp> ppl can have issues when they have ceph-mons in containers
16:28:13 <evrardjp> it's something I will have a look in the future
16:28:20 <odyssey4me> yeah, I saw some chat about that in the channel last night
16:28:44 <evrardjp> our prio list for that went a little down, so I had to work on something else
16:28:54 <odyssey4me> ok - are we in a place where we can take off the warning labels we have in the docs and script so that people don't get the warning about it being untested?
16:29:04 <evrardjp> there is probably some kind of docs that should be included too
16:29:08 <evrardjp> for neutron
16:29:27 <asettle> Uhhhh odyssey4me I don't know if I want the warning label off the docs. It genuinely IS untested.
16:29:43 <asettle> If you want to remove the labels, could we potentially add a 'in testing' for the install guide?
16:29:48 <odyssey4me> asettle for liberty -> mitaka we've had some deployers say they did it and all was well
16:30:04 <odyssey4me> asettle this refers to the mitaka docs, not the newton docs
16:30:13 <asettle> odyssey4me: yesss that's specifically what I am referring to.
16:30:34 <evrardjp> I could have a commit to remove that message indeed
16:30:45 <evrardjp> I'll do that today
16:31:29 <odyssey4me> I think the script should still require confirmation that the deployer intends to do it - but we can take the 'OMG THIS WILL NEVER WORK' warnings away now.
16:31:43 <asettle> Hmm okay
16:31:49 <odyssey4me> if there are still improvements, they can come
16:31:51 <asettle> I mean, I want to say 'yes it'll work it's amazing' but hey, I really don't know.
16:32:06 <odyssey4me> no, we can't make that sort of promise
16:32:46 <qwang> i want to bring up one thing. the script does not confirm the current version installed and the version of run-upgrade itself.
16:33:04 <qwang> it seems a little unsafe to me
16:33:29 <odyssey4me> qwang yeah, we should probably implement something nice in master for newton
16:33:48 <palendae> Previous iterations of the upgrade script did try to check versions
16:33:54 <odyssey4me> but I don't want to try and implement changes in mitaka now... I'll leave that to those who are actively testing and working on it
16:34:11 <odyssey4me> so mitaka -> newton upgrade reviews are open season
16:34:54 <evrardjp> you have your commit
16:35:13 <evrardjp> :)
16:35:16 <evrardjp> #link https://review.openstack.org/#/c/364448/
16:35:42 <odyssey4me> thanks evrardjp
16:37:06 <odyssey4me> ok, there's nothing else on the agenda
16:37:12 <odyssey4me> so #topic Open Discussion
16:37:18 <odyssey4me> #topic Open Discussion
16:37:29 <odyssey4me> anyone got anything specific to raise?
16:37:48 <palendae> Not I
16:37:52 <oneswig> Hi all, a quick question
16:38:00 <oneswig> Anyone using OS-A with SR-IOV networking?
16:38:20 <odyssey4me> hmm, oneswig IIRC a while back someone asked about it
16:38:36 <odyssey4me> but I don't think anyone's using it at this stage.
16:38:38 <oneswig> ... might have been me...
16:39:04 <oneswig> OK thanks odyssey4me, I'm guessing its not on the radar either?
16:39:28 <evrardjp> no I think someone else long ago asked this too :D
16:39:53 <odyssey4me> oneswig once it becomes important to someone, typically we see patches for it coming in
16:39:56 <evrardjp> oneswig: if you have issues we can maybe sort that out in the chan
16:40:01 <palendae> I don't think I've heard of anyone pushing on that; i think that's what it needs to be on the radar
16:40:05 <palendae> Similar to OVS support
16:40:18 <odyssey4me> but yeah, if you need help hacking it to work then we're always available to discuss
16:40:49 <oneswig> We use tripleo for deploying SR-IOV, I was wondering...
16:41:17 <odyssey4me> if you can advise what sort of things would be needed, we can probably put the patches together... but we have no way of testing it unfortunately
16:41:51 <oneswig> odyssey4me: if I get the chance i will certainly do that as I do have a working implementation for reference
16:41:53 <odyssey4me> oneswig soo... if you could lend us a supercomputer for a few weeks we could put it together :p
16:42:11 <oneswig> ha.
16:42:16 <evrardjp> I can find time in my week-ends to work on this :D
16:42:37 <oneswig> Thanks all, I will hopefully follow up in due course
16:43:12 <odyssey4me> alright, anything else?
16:46:08 <odyssey4me> okie dokey - I guess that's it then
16:46:09 <odyssey4me> thanks all
16:46:12 <odyssey4me> #endmeeting