15:00:19 <jimbaker> #startmeeting craton
15:00:19 <openstack> Meeting started Mon Oct  3 15:00:19 2016 UTC and is due to finish in 60 minutes.  The chair is jimbaker. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:20 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:23 <openstack> The meeting name has been set to 'craton'
15:00:44 <syed_> Hi Everyone
15:00:48 <jimbaker> #topic roll call
15:00:51 <jimbaker> syed_, hi
15:01:05 <jovon> hello
15:01:06 <sigmavirus> o/
15:01:19 <jimbaker> hope everyone had a great weekend!
15:01:23 <syed_> Hey sigmavirus
15:02:09 <jovon> indeed! you jimbaker ?
15:02:33 <jimbaker> jovon, it was very nice. went mountain biking with my son on saturday; we went orienteering together on sunday
15:02:56 <jovon> sounds excellent :)
15:03:54 <jimbaker> sulo, hi
15:05:20 <jimbaker> ok, i think it's time for us to get going here. we will look at core stuff later in today's meeting
15:05:31 <jimbaker> but all client people are here
15:05:54 <jimbaker> #topic development
15:06:29 <jimbaker> sigmavirus, anything to note on client work for this week?
15:06:57 <chrisspencer> o/
15:07:08 <jimbaker> chrisspencer, hey, good to see you
15:07:11 <sigmavirus> I have been researching how to avoid a lot of the trouble syed_'s been having getting Keystone identity to work with craton/cratonclient
15:07:31 <jimbaker> sigmavirus, +1
15:07:36 <syed_> +2
15:07:36 <sigmavirus> I have an idea, but I want to test it locally before sending syed_ on a wild goose chase
15:08:02 <sigmavirus> Also, spurred by teh client code, I'm working on simplifying our API so we arent' passing project_id more than once in a given request
15:08:59 <jimbaker> sigmavirus, great stuff
15:09:28 <jimbaker> anything you need from us in this work?
15:09:38 * sigmavirus shrugs
15:09:46 <sigmavirus> As usual though, I'm being pulled in several directions
15:09:47 <jimbaker> besides a pony?
15:09:50 <sigmavirus> Trying to focus on craton
15:10:04 <sigmavirus> 'Tis my own fault as usual as well
15:10:05 <sigmavirus> :)
15:10:48 <jimbaker> sigmavirus, i understand. hopefully we can keep that focus :)
15:10:54 * sigmavirus is working on i
15:11:34 <jimbaker> syed_, any updates on docker and functional testing?
15:12:30 <syed_> i have been working over the keystone thing to get it working for the horizon folks
15:12:59 <syed_> once that is completed i will shift more towards docker functional testing, i have been looking into it but partially
15:13:11 <jimbaker> syed_, understood, just wanted to see where that stood
15:13:37 <syed_> apart from this i have been looking into a bug , saying updating resources does not work currently, so have been fixing that up, will be able to upload soon
15:13:49 <jimbaker> let's work together this week on functional testing, especially given the research sigmavirus is doing
15:13:59 <syed_> i believe that is an important aspect
15:14:07 <syed_> sounds good to me
15:14:54 <syed_> once we are able to solve the keystone authentication craton/cratonclient, it will be really helpful for the horizon folks for integrating things and doing their part ..
15:15:28 <syed_> i have been reading into how different openstack projects are doing that. I hope soon we will be able to get it working.
15:15:40 <jimbaker> syed_, cool. glad you have had a chance to look at the api. and keystone support is the most important given it's blocking lcastell and team
15:15:59 <syed_> Agreed jim.
15:16:11 <syed_> hopefully sigma virus and i will be able to get this going soon
15:16:15 <jimbaker> syed_, also this will be important for rbac once we get there
15:16:16 <lcastell> \o/
15:16:42 <jimbaker> i believe mdorman was asking me about rbac at the talk i gave last tuesday in boulder
15:16:56 <jimbaker> (technically broomfield, colorado)
15:17:00 <syed_> Sounds good
15:17:35 <jimbaker> lcastell, good to see you. i should have put you in that reminder list
15:17:43 <mdorman> not a show stopper for us, but was just curious around the is_root and is_admin attributes, and possibility of a single user having access to multiple projects.
15:17:45 <jimbaker> of people to ping pre meetings
15:18:36 <jimbaker> mdorman, i expect us to provide that support, post barcelona. it will be a good thing to discuss in barcelona in fact
15:18:37 <lcastell> np jimbaker :)
15:19:11 <mdorman> cool
15:20:43 <jimbaker> mdorman, the is_root and is_admin attributes really need to be expanded into roles; and specific scope of those roles that
15:20:49 <jimbaker> is governed by inventory
15:21:00 <jimbaker> from project on down
15:21:36 <mdorman> makes sense.
15:22:24 <jimbaker> jovon, any updates for this week?
15:23:26 <jovon> nothing interesting more bug fixes regarding duplicate network insertion errors
15:23:55 <jimbaker> jovon, sure, which gets into the fact i need to finalize the patch for the api refactoring so you can get that dupe bug properly fixed with minimal fuss
15:23:55 <jovon> Bug #1623100
15:23:56 <openstack> bug 1623100 in craton "Duplicate insertion of network objects with the same name is allowed by API server" [High,Triaged] https://launchpad.net/bugs/1623100 - Assigned to Jovon McCloud (jovon-mccloud)
15:24:22 <jimbaker> jovon, have you tried using https://review.openstack.org/#/c/378977/ as a base there?
15:24:45 <jimbaker> still under review of course, but simplifies the required error handling
15:24:53 <jovon> i have not currently
15:25:50 <jimbaker> jovon, i would recommend doing so, since it should mean just a few lines to implement for the patch, plus tests
15:27:44 <jovon> i'll investigate this more closely directly following this meeting
15:27:44 <jimbaker> anyway, i will push up a new change set later this morning that will hopefully sidestep the other stuff we were discussing about import best practices
15:28:10 <jimbaker> jovon, sounds good. make sure to reach out to me so we can look at the relevant model code
15:28:14 <jimbaker> together
15:29:03 <jovon> ok. thanks jimbaker
15:29:16 <jimbaker> ambiguous. this is not model.py :) but the existing error handling code in dbapi that is closest to what you need to do
15:30:05 <jimbaker> sulo, are you around?
15:30:23 <sulo> o/, hello
15:30:35 <jimbaker> so as i mention, my immediate priority is to revise that patch ( https://review.openstack.org/#/c/378977/)
15:30:55 <jimbaker> by removing all controversial aspects ;)
15:31:03 <sulo> ok, sounds good
15:31:17 <jimbaker> another time for that
15:31:50 <jimbaker> sulo, what will be working on this week?
15:32:10 <jimbaker> will *you*
15:32:31 <sulo> i have mostly pending changes for labels support, and fxing some stuff for /data enpoint
15:32:39 <sulo> then ansible-endpoint
15:32:48 <jimbaker> sulo, cool, so all important stuff
15:33:10 <jimbaker> syed_, you are still planning on working on the corresponding variables support in the client, with sigmavirus ?
15:33:35 <sulo> yeah that should givive us a complet set for barcelona
15:34:11 <jimbaker> right, in terms of base inventory support
15:34:17 <sulo> thre are quite a lot of bugs that needs to be sorted as well
15:34:27 <syed_> yes that is very important
15:35:25 <jimbaker> so with sulo's suggestion re taskflow notifications vs modifying the tf persistence backend, i started looking into that
15:35:26 <syed_> Once these things i have are done my next target will be that, in the meanwhile i will keep working in bits and pieces on that so once i am done with the things on my plate, we can then shift our focus towards variables support
15:36:48 <jimbaker> syed_, np. and don't worry about this backlog. it would be a good thing for you to focus on, but others can pick up the var support as it makes sense
15:37:38 <jimbaker> sulo, so notifications look good, and certainly a lot easier
15:37:55 <sulo> awesome
15:38:20 <syed_> yes sounds good
15:38:55 <jimbaker> so my plan is simple: to use a standard naming convention for task names; and then update our job repr. so i'm going to take ownership of what sulo did in https://review.openstack.org/#/c/355938/
15:39:24 <sulo> ok
15:39:35 <jimbaker> after the immediate priority of getting that api stuff fixed up, that's my priority for this week
15:39:42 <jimbaker> (about time :)
15:40:40 <jimbaker> sulo, basically i set up some toy examples of doing such notifications with the two level graphs we have discussed, and it's just a reasonable way to do it
15:40:48 <jimbaker> so high confidence here
15:41:24 <jimbaker> team, any other dev tasks we should be aware of?
15:42:43 <jimbaker> ok, well we can discuss further on #craton and tomorrow's client/ecosystem dev call
15:43:00 <jimbaker> #topic community
15:43:34 <jimbaker> sigmavirus, so we are just waiting on someone to workflow https://review.openstack.org/#/c/375680/ ?
15:44:04 <sigmavirus> jimbaker: +2 and +W, yes
15:44:27 <jimbaker> i guess that will just take its usual time
15:44:48 <jimbaker> will be useful for the community going forward
15:45:35 <jimbaker> so as i alluded to earlier, i gave a talk at the colorado openstack meetup last tuesday on craton
15:46:18 <jimbaker> it seemed to be well received; and it gave me an opportunity to think about what sulo, sarob, and i will discuss in barcelona
15:46:58 <jimbaker> one thing i will definitely have to do: make the craton specific part a lot more concise
15:48:27 <jimbaker> the other thing we will want to do in advance of barcelona is fully triage our bugs; finalize outstanding blueprints; and especially consider how to involve more people in what we are doing
15:48:44 <jimbaker> so lots of work to do between now and oct 25
15:49:10 <jimbaker> #topic questions/wrapup
15:49:51 <jimbaker> anything else? we have a lot to do, and i need to unblock sulo
15:51:31 <jimbaker> ok, i think silence here is consent to end today's meeting
15:51:38 <jimbaker> further discussion over at #craton
15:51:43 <jimbaker> #endmeeting