19:04:38 <dtroyer> #startmeeting openstackclient
19:04:38 <openstack> Meeting started Thu Aug  4 19:04:38 2016 UTC and is due to finish in 60 minutes.  The chair is dtroyer. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:04:40 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:04:42 <openstack> The meeting name has been set to 'openstackclient'
19:05:04 <dtroyer> well, could be short?
19:05:12 <dstanek> o/ dtroyer
19:05:36 <dstanek> pretty small following today
19:05:39 <dtroyer> #topic 3.0 release
19:05:47 <dtroyer> Let's start with where things are...
19:06:09 <dtroyer> OSC 3.0 is getting close, I'm wrapping up the detains now on the dependencies
19:06:27 <dtroyer> os-client-config 1.19.0 released last night and is working thoguht the g-r updates so we can test with it
19:06:53 <dtroyer> osc-lib is theoretically ready for 1.0 once the release notes review merges, unless I've missed something
19:07:11 * stevemar sneaks in
19:07:20 <stevemar> i will review those
19:07:22 <dtroyer> and osc itself has two things outstanding from me (the integ tests and the shell changes) and IIRC still one thing from Henry
19:07:41 <stevemar> i think henry posted the change, it looked good IIRC
19:07:47 * dtroyer has mostly not paid attention to other reviews there this week
19:07:51 <dtroyer> ok, good
19:08:24 <dtroyer> so that's the high level, I don't expect we'll propose 3.0 until monday given weekend timing and final testing.
19:08:27 <dtroyer> am I missing anything?
19:09:23 <rtheis> I think that is correct
19:09:37 <dtroyer> so let's talk specific then
19:09:41 <dtroyer> #topic reviews
19:09:53 <dtroyer> #link https://review.openstack.org/351271
19:10:02 <dtroyer> that's the osc-lib release notes.
19:10:14 <dtroyer> at this point, that's what I'll pin for the 1.0 release
19:10:33 <dtroyer> I see rtheis proposed the releasenotes job, thanks for that
19:10:39 <rtheis> dtroyer: yw
19:11:08 <dtroyer> rtheis: also, you had a comment in https://review.openstack.org/#/c/345305/ about waiting for 1.0.  Do you mean waiting until after?
19:11:49 * rtheis looking ...
19:12:05 <stevemar> i think this is henry's last patch that is neede https://review.openstack.org/#/c/283554/
19:12:26 <rtheis> dtroyer: I think we want this in 1.0 release
19:12:52 <rtheis> it changes behavior so best to do in osc-lib 1.0 and osc 3.0 instead of after ?
19:13:30 <dtroyer> re-reading it I think I'll agree.  it has a release note, so that's done, except should it be in the fixes section?
19:14:18 <rtheis> yeah may in the fixes section
19:14:22 <dtroyer> or conversly, should my summary batch be under upgrades?
19:14:52 <dtroyer> I looked at it as a new lib, so it's a change from osc, not an upgrade.  I could be persuaded though
19:15:47 <stevemar> nah, not upgrades
19:16:38 <dtroyer> ok, I'll tweak 345305 and get that ball rolling, those will be it for osc-lib 1.0 then
19:16:56 <rtheis> sounds good
19:17:25 <dtroyer> stevemar: pointed at https://review.openstack.org/#/c/283554/ as Hentry's 3.0 blocker, will that be finished soon?
19:18:09 * dtroyer will look closely tomorrow, but trusts stevemar's judgement for some reason
19:18:54 <stevemar> dtroyer: bad idea to trust it
19:19:13 <stevemar> "just a few small issues, looks good overall though! "
19:20:24 <dtroyer> https://review.openstack.org/346113 is passing again, but I'd really rather merge the integ tests (https://review.openstack.org/349079) first for peace of mind about not breaking CLI behaviour in unexpected ways
19:20:43 <dtroyer> the integ tests need o-c-c 1.19.0 to pass though, so blocked on g-r
19:20:48 <stevemar> i think we probably want to send a mailing list post about 3.0.0 being released
19:20:52 <dtroyer> time to start poking folk
19:21:14 * dtroyer looks at stevemar for jumping ahead ;)
19:21:21 <stevemar> my bad
19:21:37 * dtroyer looks at self for not having an agenda
19:22:06 <stevemar> :)
19:22:32 <dtroyer> Those are the blockers for 3.0, are there any other open reviews that would be nice to have yet?
19:23:14 <rtheis> I don't have any
19:23:52 <stevemar> negative
19:24:14 <stevemar> dtroyer: oh we should get a cliff release in there
19:24:25 <dtroyer> roger.  We'll have enough stacked up to do a 3.next in a week or two anyway, that's fine.
19:24:43 <stevemar> yeah
19:24:49 <dtroyer> oh!  That would be good to, thanks.  We're not blocked on anything I am aware of there, but that would be cleaner
19:25:09 <stevemar> dtroyer: the only blocker is https://review.openstack.org/#/c/342914/
19:25:42 <stevemar> which would help in osc's translationy stuff
19:25:53 <dtroyer> thanks stevemar, I have not kept up in that one
19:26:04 <dtroyer> yup, and should it be a change we'll want it now
19:26:54 <rtheis> dtroyer: when should openstackclient be bumped in global requirements?  It is at >=2.1.0 now
19:27:13 <dtroyer> That really depends on others using it more than us
19:27:29 <dtroyer> I prefer to be conservative there when we can
19:27:38 <rtheis> okay
19:27:51 <dtroyer> something like DevStack would be the ususal cause for a bump
19:28:29 <dtroyer> particularly when we get the startup time improved (which I have not yet measured for 3.0)
19:28:52 <stevemar> rtheis: once 3.0.0 is released, we need to make changes to devstack for the backwards incompatiable changes in role list
19:29:00 <stevemar> cause devstack will start printing deprecation warnings
19:29:34 <dtroyer> and there's when it'll get bumped in g-r ;)
19:29:39 <stevemar> then we can propose the g-r patch to bump it to 3.0.0 and ask the puppet folks to depend on it
19:29:52 <stevemar> and a few of the other usual consumers
19:30:43 <dtroyer> since we are breaking some things I think we need to move deliberately here
19:31:14 <dtroyer> doing examples and help with plugins using osc-lib 1.0 is another area we can start with then
19:31:58 <dtroyer> and I think jamie is looking forward to seeing where else I am abusing ksa :)
19:33:24 <dtroyer> so for communication, I'll post to the ML before we cut the release that will have the complete stack overview
19:33:32 <dtroyer> (now is the time stevemar)
19:33:36 <rtheis> thank you
19:34:20 <dtroyer> we'll also need to make some passes through published docs to see what needs updating, I really have not even looked in to that yet
19:34:40 <stevemar> dtroyer: i was just gonna say that we should start an etherpad to prep the mailing list content
19:34:53 <stevemar> we really want to emphasize the change to role list and role assignment list
19:35:02 <stevemar> and apologize profusely for it
19:35:25 <dtroyer> you mean like https://etherpad.openstack.org/p/osc-3.0-release?
19:36:28 <stevemar> oh hey look at that
19:37:04 <dtroyer> so lets move on while stevemar types in etherpad
19:37:08 <dtroyer> #topci bugs
19:37:12 <stevemar> :)
19:37:24 <stevemar> if i don't write things down they escape my tiny brain
19:37:26 <dtroyer> Speaking of stevemar typing, he cleaned up the bug list big time, thanks!!!!
19:37:41 <rtheis> thanks stevemar
19:37:45 <stevemar> i'm proud of my golden beetle award
19:38:21 <rtheis> you should be, I was tired just looking at all of the emails generated
19:38:27 <dtroyer> it'll be the only one to survive the award-shelf apocalypse
19:38:43 <stevemar> rtheis: lol
19:39:00 <stevemar> rtheis: and i did OSC after cleaning up keystone, keystoneauth, keystoneclient and keystonemiddleware
19:39:17 <rtheis> that's a lot of golden beetles
19:39:19 <dtroyer> tha'ts 'in the zone'
19:40:06 <dtroyer> and of course the first thing I see there is https://bugs.launchpad.net/python-openstackclient/+bug/1593784
19:40:06 <openstack> Launchpad bug 1593784 in python-openstackclient "trim requirements before 3.0.0 release" [High,Triaged]
19:40:16 <dtroyer> another good pre-release thing for the list
19:41:04 <stevemar> yeah, with osc-lib in play we'll have a lot less reqs
19:41:20 <dtroyer> I did have one report of problems with token auth yesterday, it may be fixed with the last shell review, so I'm waiting on that yet too
19:41:50 <dtroyer> I don't think it is a 3.0 blocker as it is broken today in 2.6.0 though
19:42:02 <dtroyer> not sure if a bug exists for it yet
19:42:26 <dtroyer> anything else bug-wise?
19:42:59 <rtheis> nothing from me
19:43:16 <stevemar> nope
19:43:18 <dtroyer> #topic open discussion
19:43:23 <dtroyer> what else is going on?
19:43:43 <rtheis> neutronclient merged first OSC plugin commands for network trunk
19:43:53 <dtroyer> \o/
19:44:17 <stevemar> Huanxuan Ao is doing great work, i'll just leave that comment there :)
19:44:24 <rtheis> +1
19:44:27 <dtroyer> ++++
19:44:45 <dtroyer> does he IRC?
19:45:04 <stevemar> even picking up patches that are abandoned / lagging
19:45:14 <stevemar> maybe, not our time at least
19:46:02 <stevemar> i think he works with tang, maybe reach out to tang
19:46:36 <dtroyer> looks like he does
19:48:38 <dtroyer> I had some questions from Ironic this week about handling properies in a show command differently
19:48:57 * devananda starts listening
19:48:59 <dtroyer> leaning toward resource show —properites rather than attempting to nest them
19:49:39 <dtroyer> its actually a list command under the hood, what I am unsure about is if it should be a list command at the cli
19:50:06 <dtroyer> this pattern would be very helpful in other places where we have a similar problem, so getting it right will be a Good Thing
19:50:11 <devananda> dtroyer: which command exactly?
19:50:24 <stevemar> havent looked into it
19:50:27 * dtroyer tries to remember without notes
19:50:41 <dtroyer> driver backend show, or something like that
19:50:59 <dtroyer> it is listing the driver-specific backend properties
19:51:48 <dtroyer> osc has this problem in general with resource properites, we currently just bunch them up as a single field in a show command, which is odd at best
19:51:49 <devananda> "openstack baremetal node show --help" doesn't list anything about properties
19:52:10 <dtroyer> its a new command, just a spec now
19:52:32 <devananda> ahh. ok
19:52:39 <dtroyer> I'll dig through my notes after the meeting
19:53:10 <dtroyer> I am hoping this will help us sort out how to handle this case in general
19:54:24 <dtroyer> also, (related) we'll be doing another UX study in Barcelona
19:54:33 <devananda> I lack the wider context to feel like I can propose anything general
19:54:42 * devananda goes back to ironic things :)
19:54:51 <dtroyer> The properties thing is the kind of thing I would love to use those for, but I don't think we can wait that long here
19:55:10 <dtroyer> thanks for checking in devananda
19:55:19 <devananda> np!
19:55:43 <dtroyer> anything else anyone?
19:55:45 <stevemar> oh nice re: ux
19:55:53 <rtheis> nothing else from me
19:56:11 <stevemar> nada
19:56:17 <stevemar> excited to finally get this done
19:56:27 <dtroyer> wellallrightythen, let's go 3.0
19:56:34 <dtroyer> thanks guys!
19:56:46 <dtroyer> #endmeeting