09:00:17 <yuval> #startmeeting karbor
09:00:17 <openstack> Meeting started Tue Aug 15 09:00:17 2017 UTC and is due to finish in 60 minutes.  The chair is yuval. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:00:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:00:21 <openstack> The meeting name has been set to 'karbor'
09:00:31 <yuval> Hello and welcome to Karbor's biweekly meeting
09:01:12 <yuval> Today, we will be discussing: PTL elections for queens, Pike status
09:01:50 <yuval> waiting 2 minutes for people to join
09:03:39 <zhonghua> hi
09:03:56 <chenying> hi
09:04:34 <yuval> hey zhonghua chenying
09:04:39 <yuval> zengchen joining?
09:05:07 <yuval> zhonghua: ?
09:05:56 <zhonghua> yuval, hey
09:06:14 <edisonxiang> hi
09:06:20 <yuval> hey edisonxiang
09:06:31 <edisonxiang> what' up yuval
09:06:40 <yuval> edisonxiang: very well, good to see you
09:06:52 <edisonxiang> yuval: me too
09:07:05 <yuval> zhonghua: will zengchen join?
09:07:25 <zhonghua> one sec
09:07:30 <zengchen> hi
09:07:39 <yuval> hey zengchen
09:07:43 <yuval> let's start
09:07:51 <yuval> #topic Queens PTL
09:08:40 <yuval> As you may have noticed, I have not placed my candidancy for Queens PTL, and Chenying sent his candidancy. I believe Chenying is the perfect fit to lead Karbor in Queens
09:09:15 <zengchen> oh, congratulation to chenying
09:09:33 <chenying> thanks everyone thanks yuval.
09:09:37 <yuval> Yes, congratulations :)
09:09:40 <zhonghua> yuval, chenying, thank you all
09:09:43 <edisonxiang> :)
09:09:45 <chenying> You all did great jobs on karbor.
09:10:18 <yuval> :)
09:11:08 <yuval> Alright, so let's continue on
09:11:13 <yuval> #topic Pike Status
09:11:26 <yuval> as you may have seen, all karbor repos branched RC1 for pike
09:12:16 <yuval> all patches entering master will be part of Queens
09:12:29 <yuval> any patch we want in Pike will have to be backported to stable/pike
09:12:40 <yuval> and then we'll have to make a release for stable/pike
09:12:51 <yuval> so, take that into account
09:13:37 <yuval> chenying: would you like to add something?
09:13:53 <chenying> yuval I hope the patch about 'Operation Engine: Support Trigger Multi Node' could be included in pike. What's your oppion about it?
09:14:21 <jiaopengju> hi
09:16:06 <yuval> chenying: I hope so as well, but I keep finding crucial basic bugs
09:16:08 <yuval> hey jiaopengju
09:17:48 <jiaopengju> yuval sorry for being late
09:17:56 <yuval> jiaopengju: it's ok
09:18:10 <chenying> yuval OK I see. we could merge it to master first. If we have time before pike relase, we coud backport to pike.
09:18:27 <yuval> chenying: sounds good
09:18:34 <chenying> jiaopengju hi
09:18:50 <yuval> In August 21st we enter final RC week
09:19:17 <yuval> basically, we are ready, only final patches will enter
09:19:21 <chenying> yuval I will aslo test karbor to find bugs these days.
09:19:39 <yuval> chenying: that's very important. Including client, openstackclient, and dashboard
09:20:01 <jiaopengju> yuval chenying: I will do some tests too.
09:20:09 <chenying> jiaopengju Thanks.
09:20:20 <jiaopengju> chenying: :)
09:20:28 <yuval> also, we need tests with and without manila/trove
09:21:33 <amrith> yuval do you want to add something to your PTG agenda to discuss trove integration? that was suggested in the review of your tests but I never heard back on that suggestion
09:22:21 <yuval> amrith: Karbor will be doing PTG meeting over video, nobody from Karbor will be arriving to Denver (for various reasons)
09:22:30 <chenying> manila tests is included in fullstack jinkens. jiaopengju Could you do some tests without manila and trove services? Have you deployed manila and trove services in your environment?
09:22:53 <amrith> that's ok, if you are doing by video I'm sure we could join you from denver if you invited us
09:23:34 <jiaopengju> chenying: ok, I will do it in the following few days
09:23:40 <yuval> amrith: tbh, the trove integration is already included in Karbor, and manually tested by chenying. The only thing we have left are tests in the gate
09:26:45 <yuval> amrith: is there anything we can do to merge the discussed patch? Would speaking this over the PTG help? Your argument is about maintaining API
09:27:28 <amrith> yuval the patch as proposed isn't going to be merged anytime soon, what you are asking for is something we're not looking to offer in that utility
09:27:57 <amrith> for reasons described in the review
09:28:51 <yuval> amrith: would trove offer any other way for other OpenStack projects to deploy a gate environment?
09:29:08 <amrith> we already do, you don't seem to like it ...
09:30:29 <yuval> amrith: I just don't like duplicating code from 'cmd_dsvm_gate_tests' in trovestack
09:31:31 <amrith> ok, in that case there's a devstack plugin like every other project. you'll have to use the devstack plugin and then actually configure and create datastores for your required database and then do what you wish. that's how a 'real user' would deploy trove
09:33:05 <yuval> amrith: that would require us to literally copy the code from 'cmd_dsvm_gate_tests' into karbor tests, which we don't want to do
09:33:57 <amrith> yes, and we're not signing up to give you a new 'api' which would do what you are looking for.
09:33:59 <yuval> amrith: or use another deployment method, specifically for gates, where trove already has one
09:35:35 <yuval> amrith: alright
09:35:39 <yuval> anything else?
09:36:18 <yuval> chenying: ?
09:36:31 <amrith> not for me, just wanted to suggest that if you want to discuss this more, we're available. if your position is that this is your only workable solution, then we are at an impasse.
09:37:36 <chenying> use another deployment method? which one?
09:38:31 <yuval> chenying: we can use trovestack and "duplicate" the 'cmd_dsvm_gate_tests' function in our tests
09:39:02 <yuval> chenying: that would require us to monitor that script for every change amrith and the trove team are doing over there and port it to karbor tests
09:39:14 <yuval> amrith: correct?
09:39:36 <amrith> and when it goes away, you have the only remaining copy
09:40:45 <yuval> amrith: I guess if it will 'go away' it will be part of trove's rearchitect effort, right?
09:41:51 <amrith> no, getting rid of it has been the plan for a while now
09:41:56 <amrith> it will likely happen quite soon
09:42:20 <chenying> yuval It means that cmd_int_tests of trove will be running in karbor's jenkins.
09:44:22 <yuval> amrith: ok, found no info about planned removal in trove meetings / mailing list
09:44:53 <amrith> this has been discussed about two years ago.
09:44:55 <amrith> feel free to look up the words trove-integration
09:45:01 <yuval> amrith: we'll be happy if the replacement will make the gate deployment easier for us (as we would like to concentrate on building karbor, and not on writing trove deployment)
09:46:56 <amrith> ok, if that is your position, so be it
09:47:07 <yuval> chenying: ?
09:48:27 <chenying> yuval I don't think the tests about trove also port to karbor's jenkins. we only need the script to prepare trove service.
09:49:53 <yuval> chenying: right
09:51:03 <yuval> lets move on
09:51:08 <yuval> #topic Open Discussion
09:51:39 <chenying> hi all I have listed the plans for queens here.[1] Everyone add new plan, and we can discuss them. The detail usecase about these plans could be added here[2].
09:51:39 <chenying> [1] https://ethercalc.openstack.org/karbor-queens
09:51:39 <chenying> [2] https://etherpad.openstack.org/p/karbor-queens-planning
09:52:23 <yuval> #link https://ethercalc.openstack.org/karbor-queens
09:52:28 <yuval> #link https://etherpad.openstack.org/p/karbor-queens-planning
09:54:26 <yuval> chenying: will do
09:54:42 <yuval> jiaopengju: you missed the beginning, anything you'd like to add?
09:55:12 <jiaopengju> yuval: nothing :)
09:55:16 <yuval> jiaopengju: :)
09:55:26 <yuval> thank you all for attending
09:55:33 <yuval> I guess the next meeting will be run by chenying
09:56:19 <yuval> #endmeeting