19:01:14 <dtroyer> #startmeeting openstackclient
19:01:14 <openstack> Meeting started Thu Jul  7 19:01:14 2016 UTC and is due to finish in 60 minutes.  The chair is dtroyer. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:01:15 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:01:17 <openstack> The meeting name has been set to 'openstackclient'
19:01:33 <dtroyer> priorities: lunch vs meeting
19:01:38 <rtheis> lunch
19:01:44 <stevemar> dtroyer: don't eat so late
19:02:01 <dtroyer> I hate coming back from vacation
19:02:14 <dtroyer> the pile-o-things to go through is…blech
19:02:33 <rtheis> yeah, not looking forward to that.  I'm out next week on vacation
19:02:37 <dtroyer> so, uh, did I miss much?
19:02:38 <stevemar> welcome back!
19:02:52 <rtheis> floating IP patches merged
19:02:53 <stevemar> dtroyer: we did a point release of osc-lib
19:03:51 <stevemar> not sure if anything else major has happened
19:03:56 <dtroyer> I saw that… did that take care of the breakage?
19:04:04 <stevemar> dtroyer: sort of
19:04:16 <stevemar> dtroyer: it took care of one bug and then broke our gate :P
19:04:26 <stevemar> dtroyer: puppet's gate broke
19:04:38 <stevemar> we fixed it with the osc-lib point release
19:04:47 <dtroyer> ok
19:04:49 <stevemar> that resulted in our py27 gate breaking
19:04:59 <stevemar> but ... just a few tests, cause of the way we mocked things
19:05:06 <dtroyer> methinks we pushed osc-lib out into other projects waaay too soon
19:05:22 <stevemar> i have the 10 failing tests skipped for now, and i plan to fix them, just distracted
19:05:40 <dtroyer> what area is that in?
19:05:52 <stevemar> the failing tests?
19:05:56 <dtroyer> yes
19:06:22 <stevemar> various tests, anything that did extensive mocking of find_resources
19:06:30 <dtroyer> oh, ok
19:06:38 <dtroyer> I'm really beginning to hate that function
19:06:40 <stevemar> nothing crazy, functionally it still works
19:06:43 <dtroyer> it's tooooooo fragile
19:07:09 <stevemar> not really fragile, it silently ignores errors and tries another way
19:07:16 <stevemar> it's openstack, we can't have nice thigns
19:07:37 <dtroyer> but there are a bunch of assumptions in and around it, and periodically people try to change one without knowing the follow-on effects
19:08:26 <stevemar> i opened a bug for it
19:08:32 <dtroyer> We're one of the nice things, so hiding the ugly inside OSc is a Good Thing(TM)
19:08:35 <stevemar> https://bugs.launchpad.net/python-openstackclient/+bug/1599333
19:08:35 <openstack> Launchpad bug 1599333 in python-openstackclient "latest osc-lib breaks osc gate" [Undecided,New]
19:08:42 <dtroyer> ok, good
19:08:44 <dtroyer> thanks
19:09:01 <stevemar> 10 tests, nothing crazy
19:09:14 <dtroyer> Looks like tang's IP commands are merged?  did they all go in?
19:09:21 <stevemar> i think it's cause i changed the order of how they search
19:09:27 <stevemar> they should be, yes
19:09:31 <rtheis> yes
19:09:39 <dtroyer> cool
19:09:50 <dtroyer> so the bulk of the stuff holdin up 3.0 is mine then, eh?
19:10:05 <stevemar> what else is holding it up?
19:10:11 <stevemar> just the osc-lib migration?
19:10:26 <dtroyer> The rework of Shell and ClientManager, plus more auth cleanup
19:10:38 <dtroyer> o-c-c stuff too
19:10:42 <stevemar> ah
19:10:51 <dtroyer> we still are duplicating too much between the three pieces
19:10:58 <stevemar> yeah
19:11:20 <stevemar> you should see if you can recruit jamielennox to look at things, he is presently on a tear cleaning everything up
19:11:33 <dtroyer> I see jamie got one or two of the things on my list
19:11:54 <dtroyer> but I'd like to finish the cutover…
19:12:33 <dtroyer> Timing is always such fun
19:13:28 <dtroyer> stevemar: wanna pounce on https://review.openstack.org/#/c/331418/ in o-c-c?  That's the next step in de-duping us
19:14:27 <stevemar> i can
19:16:38 <dtroyer> I don't have much of a list here of other stuff…
19:16:49 <dtroyer> There are a couple of reviews to add some columns
19:16:55 <dtroyer> in list commands
19:17:16 <dtroyer> https://review.openstack.org/#/c/337900/ is the one I'm looking at now
19:17:28 <stevemar> yeah, someone wanted to do that with image names and ids
19:17:43 <stevemar> y, marc has been on an osc tear
19:17:55 <dtroyer> right… we don't include both anme and ID anywhere else where it is not the primary resource being listed
19:17:59 <dtroyer> why do we need it here?
19:18:24 <stevemar> dtroyer: whats the argument over not listing it?
19:18:26 <dtroyer> and, I wonder if adding image to the default columns is needed.  why not flavor then too?
19:18:38 <stevemar> servers have an image after all
19:18:47 <dtroyer> the balance between all and not nearly enough stuff
19:18:56 <dtroyer> servers also have network ports
19:19:01 <dtroyer> shouldn't those be there too?
19:19:10 <dtroyer> volumes?
19:19:17 <stevemar> we include networks in the list
19:19:24 <dtroyer> I'm getting silly, I know, but I think we need a good reason to add things
19:19:39 <dtroyer> and the networks are a formatting problem too
19:19:50 <stevemar> i'd be OK with flavor, volume and image :P
19:19:55 <stevemar> what does horizon show?
19:20:29 <stevemar> david-lyle: o/
19:20:34 <dtroyer> image is one of the simpler ones in this category as there is only 1 per server
19:20:51 <rtheis> horizon shows "Instance Name 	Image Name 	IP Address 	Size 	Key Pair 	Status 	Availability Zone 	Task 	Power State 	Time since created 	Actions"
19:20:54 <dtroyer> volumes, networks, etc have the one-to-many formattign problem we have not solved yet
19:21:17 <dtroyer> no image ID
19:21:47 <stevemar> i agree that if it's tough to get into one cell in the table, then it probably shouldn't be included
19:22:19 <stevemar> doing -c volumes and getting back "vol1, vol2" doesn't help with scripting
19:22:39 <stevemar> (os server list --long -c volumes) (if we had that option)
19:22:47 <dtroyer> right…in this case Im trying to push just hard enough to have a good reason for adding things
19:23:08 <stevemar> maybe just one of image id / name
19:23:38 <dtroyer> I'll leave a comment in the review, I just wanted to get a sense of what you huys thought in general
19:24:36 <dtroyer> oh, did we get Henry's stuff merged yet?
19:24:44 * dtroyer is still looking
19:25:04 <rtheis> I don't think so
19:26:41 <stevemar> he did not
19:26:46 <stevemar> he's busy with other stuff
19:27:31 <dtroyer> is that something we can complete for him?  at least part of it is a compat issue and I'd hate to miss this chance to get that taken care of
19:27:32 <stevemar> do you want to hold up 3.0.0 for that?
19:27:50 <stevemar> we could probably do it for him
19:27:56 <stevemar> he's got some stuff up
19:28:40 <dtroyer> ok, I'll be at least a week yet on the osc-lib stuff, let's see how it looks next week
19:28:40 <stevemar> it's basically taking out all the junk from `role list` right? and making that assignment list
19:28:44 <stevemar> i forget the details
19:29:04 <dtroyer> that is part of it IIRC.  I have also forgotten details
19:29:11 <stevemar> dang it
19:29:19 <stevemar> OK, let me take a look at it
19:29:29 <stevemar> maybe i can find time on the weekend
19:30:52 <dtroyer> also, gyee has https://review.openstack.org/#/c/279854/ still open for TOTP.  Is that a compat or 3.0 priority?
19:34:34 <stevemar> dtroyer: neither
19:34:45 <stevemar> dtroyer: that can go in at any time
19:35:18 <dtroyer> is it ready?  I had been holding off until the Keystoney-bits were settled ;)
19:36:09 <stevemar> haven't looked at it in a while
19:36:20 <stevemar> IIRC i didn't like the option names, too long
19:36:23 <dtroyer> ok, no worries then.  I'll put it on my list
19:36:33 <dtroyer> oh, right
19:36:47 <stevemar> also dependency on cryptography now? ugh
19:37:03 <dtroyer> we already have it, just not explicit/top-level for us
19:37:21 <stevemar> what depends on it?
19:37:26 <dtroyer> I'm not sure what brings it in, SDK maybe?
19:37:43 <briancurtin> not i
19:38:16 * dtroyer makes note that briancurtin is paying attention ;)
19:38:47 * briancurtin made the mistake long ago of having irc highlight on SDK
19:38:56 * briancurtin but is sort of paying attention
19:39:24 <dtroyer> so while you are here… did I see the compute refactor get released?
19:40:08 <dtroyer> sdk sdk sdk :)
19:40:27 <briancurtin> it did. 0.9.0 is the first release of anything with refactored resources/proxy. just this morning i got image v2 “done” minus tests, so that’s close. however, it’s lacking some of the recently documented stuff so not really “complete"
19:40:37 <briancurtin> should be easy enough to expand it once the current stuff is pushed, though
19:40:45 <stevemar> dtroyer: none of the openstack stuff depends on it.. http://codesearch.openstack.org/?q=cryptography&i=nope&files=requirements.txt&repos=
19:41:07 <dtroyer> cool news!  thanks!  Image is the one I am most anxious to move over for OSC
19:41:10 <stevemar> (at least not directly)
19:41:17 <stevemar> dtroyer: damn warlocks
19:41:25 <dtroyer> stevemar: requests?
19:41:45 <stevemar> we'll still have that around, cause of the API layer we have
19:41:48 <dtroyer> what??? warlock brings it in???   yet another reason to prioritize dumping glanceclient
19:42:04 <stevemar> oh sorry, i was jumping topics
19:42:24 <stevemar> my "damn warlocks" was related to removing glanceclient and using SDK
19:42:29 <stevemar> nto crypto
19:42:34 <dtroyer> oh
19:42:36 <stevemar> i haven't figured out how that is installed
19:43:35 <dtroyer> at one point cryprography was the "blessed" Python path forward so I wouldn't be surprised that something pulls it in by now
19:44:55 <stevemar> i'll keep looking
19:45:48 <dtroyer> what else do we need to talk about?  I am sufficiently unorganized (even by my standards) that this is just Open Discussion Meeting Day
19:46:16 <stevemar> thats cool, i think we're all a little distracted
19:46:48 <stevemar> oh, given any more thought to a virtual midcycle?
19:46:53 <stevemar> or a bug squash day?
19:47:03 <stevemar> even a bug squish day
19:47:41 <dtroyer> I think the bug squash is probably closer to what we could use best right now… I have not looked for a date yet though
19:48:52 <dtroyer> I do think we would get some benefit just from having a hangout (or whatever) open for a set time focus and to coordinate stuff
19:48:59 <rtheis> I thought the bug smash was this week, but haven't see a lot of smashing
19:49:00 <dtroyer> no matter what the tasks are
19:49:29 <dtroyer> oh, that organized one?  it is, IIRc that's where tangchen is, right?
19:50:20 <rtheis> I thought so
19:50:37 <stevemar> yes, there was the china bug smash
19:51:25 <stevemar> looks like there were 2 patches from folks in china https://review.openstack.org/#/c/338908/ and https://review.openstack.org/#/c/338795/ (minor bug smashy type ones)
19:51:38 <stevemar> maybe tangchen is helping them get their feet wet
19:52:09 <stevemar> on a somewhat related note, Huanxuan Ao is doing a lot of great work
19:52:10 <dtroyer> ooooh!  I like that first one
19:52:33 <stevemar> this is not a small amount of patches: https://review.openstack.org/#/q/owner:%22Huanxuan+Ao%22
19:52:47 <stevemar> he's been picking up on the volume stuff
19:53:02 <stevemar> doing all the good stuff for each patch (doc, bug, release note)
19:53:08 <dtroyer> he works with tangchen, right?
19:53:34 <stevemar> i believe so, yes
19:55:38 <dtroyer> I need to run off for a minute, will endmeeting now to not hold up the channel
19:55:40 <dtroyer> Thanks guys!
19:55:46 <stevemar> dtroyer: rgr that
19:55:46 <dtroyer> #endmeeting