03:00:38 <tfukushima> #startmeeting kuryr
03:00:39 <openstack> Meeting started Tue Feb  9 03:00:38 2016 UTC and is due to finish in 60 minutes.  The chair is tfukushima. Information about MeetBot at http://wiki.debian.org/MeetBot.
03:00:40 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
03:00:42 <openstack> The meeting name has been set to 'kuryr'
03:01:13 <tfukushima> Hi, I'm starting the regular Kuryr meeting.
03:01:22 <fawadkhaliq> hi tfukushima
03:01:37 <tfukushima> Hi fawadkhaliq.
03:02:08 <tfukushima> #link Today's agenda https://wiki.openstack.org/wiki/Meetings/Kuryr#Meeting_February_9.2C_2016
03:02:43 <tfukushima> banix gets sick today, I'm organising the meeting.
03:03:03 <fawadkhaliq> sorry to hear. get well soon banix
03:03:53 <tfukushima> So only me and fawadkhaliq are present today?
03:04:10 <fawadkhaliq> tfukushima: :-)
03:04:18 <tfukushima> We could go quick. :-p
03:04:32 <fawadkhaliq> lol yeah
03:04:56 <tfukushima> #info fawadkhaliq and tfukushima are present
03:05:05 <fawadkhaliq> Not sure if you want to discuss the k8s integration right now
03:05:15 <fawadkhaliq> I have some things to discuss on the nested containers
03:05:26 <fawadkhaliq> we can discuss that and anything else you have in mind
03:05:29 <tfukushima> Ok, let's go that path first then
03:05:42 <tfukushima> #topic Nested containers
03:05:59 <tfukushima> fawadkhaliq: Please go on.
03:06:04 <fawadkhaliq> I addressed most of the comments in the latest patchset
03:06:24 <fawadkhaliq> there is one area of discussion we are looking at Magnum folks raised
03:07:04 <tfukushima> That is?
03:07:26 <fawadkhaliq> Hongbin Lu mentioned that we should support networking for Magnum containers even when native tools like Docker CLI are used
03:08:04 <fawadkhaliq> If we go down that path, then we are looking at Kuryr agent to communicate Kuryr endpoint/Neutron over the management/API endpoints
03:08:36 <fawadkhaliq> I was thinking we should keep the scope limited in the first phase and only go with Magnum container API support
03:09:04 <fawadkhaliq> So question is what makes sense right now, solution is there to support both
03:09:34 <fawadkhaliq> however, the second requirement would introduce something we want to avoid.
03:09:40 <fawadkhaliq> tfukushima: what do you think?
03:12:09 <tfukushima> aEventually we'd look into the networking support for Magnum but at this point it's a bit ambitious.
03:13:24 <tfukushima> And the networking is happening with the K8s part.
03:13:59 <tfukushima> I mean, the networking for containers.
03:14:44 <fawadkhaliq> tfukushima: agree, that's why I would like to keep this as a phased approach.
03:15:18 <fawadkhaliq> we start with Magnum Container API and then work towards supporting the use cases, where native tools comes into the picture
03:15:48 <fawadkhaliq> tfukushima: so, cool. That was one thing. Please feel free to comment on the spec as well.
03:15:57 <fawadkhaliq> Two more small items
03:16:18 <tfukushima> It'd be good to see the coming future though.
03:16:55 <tfukushima> But we can set the narrower scope at this point and you may be able to make it clear in your comments.
03:17:13 <fawadkhaliq> tfukushima: yes exactly, we will capture that it will be handled at the later stage and define the scope clearly
03:17:54 <fawadkhaliq> That's all I wanted to discuss on the spec
03:18:03 <fawadkhaliq> So two more things
03:18:09 <tfukushima> Yep.
03:18:11 <fawadkhaliq> I have created the launchpad blueprints for Kuryr agent and Kuryr deployment via Heat
03:18:24 <fawadkhaliq> #link https://blueprints.launchpad.net/kuryr/+spec/kuryr-agent
03:18:38 <fawadkhaliq> We will capture details about the agent in this spec
03:18:39 <fawadkhaliq> and
03:19:18 <fawadkhaliq> for the action item from last week that gsagie and I had to kick off the design spec for Kuryr deployment via Heat
03:19:25 <fawadkhaliq> #link https://blueprints.launchpad.net/kuryr/+spec/kuryr-magnum-heat-deployment
03:19:43 <fawadkhaliq> we will be pushing the details very soon once
03:19:52 <fawadkhaliq> kicked off with blueprints right now.
03:20:03 <tfukushima> Heat template for Kuryr would be nice.
03:20:18 <tfukushima> Great.
03:20:57 <fawadkhaliq> next up is address the comments and hopefully start working on the details for deployment. gsagie and I will discuss
03:21:19 <fawadkhaliq> that's all on the nested containers
03:22:09 <tfukushima> Ok, thanks. I have to take a careful read on the spec.
03:22:16 <fawadkhaliq> tfukushima: please do
03:22:18 <tfukushima> #link nested container spec https://review.openstack.org/#/c/269039/
03:22:18 <fawadkhaliq> thanks
03:23:11 <tfukushima> Oh, I guess it was supposed to be "Magnum integration" topic.
03:23:31 <tfukushima> #topic Magnum integration
03:23:56 <tfukushima> #action Everyone reviews the spec again: https://review.openstack.org/#/c/269039/
03:24:19 <fawadkhaliq> thanks
03:24:28 <fawadkhaliq> I guess that's all for today? :-)
03:25:18 <tfukushima> Or do we want to discuss about K8s integration? :-p
03:25:51 <fawadkhaliq> I was thinking would be better with majority of the stakeholders here
03:26:00 <fawadkhaliq> irenab sent an update via ML
03:26:14 <tfukushima> #topic Kubernetes Integration
03:26:36 <tfukushima> #action Everyone takes a look at irenab's initiative on Kuryr ML
03:27:38 <tfukushima> fawadkhaliq: Do you have the link to her mail?
03:27:49 <fawadkhaliq> tfukushima: let me find
03:28:13 <fawadkhaliq> tfukushima: it somehow missed the openstack-dev alias :(
03:28:17 <fawadkhaliq> so no link
03:29:08 <tfukushima> Ok, anyways I missed the meeting but it seems we'd make changes to Kuryr for supporting K8s.
03:30:21 <tfukushima> And we'd have the centralized component which watches K8s API and the agent-ish distributed components on each host to do the binding notified by the centralized compnent.
03:31:41 <tfukushima> "Kubernetes Watcher API - what information can Kuryr find?"
03:32:18 <tfukushima> #action Everyone sees the Kubernets Watch API
03:32:41 <tfukushima> #link K8s Swagger UI http://kubernetes.io/third_party/swagger-ui/
03:33:25 <tfukushima> Ok, that's it.
03:33:37 <tfukushima> #topic Fullstack / Rally tests
03:34:31 <tfukushima> Anyone related to this topic is here? I'll give 30 seconds for each topic from now on and please interrupt me if you have anything.
03:35:38 <tfukushima> It seems Baohua submitted some patches.
03:35:54 <tfukushima> #link Kuryr plugin for Rally https://review.openstack.org/#/c/274014/
03:36:19 <tfukushima> #topic Packaging
03:36:50 <tfukushima> I haven't heard anything about it yet personally.
03:37:42 <tfukushima> #action apuimedo or gsagie check the status of packaging
03:37:52 <tfukushima> #topic Reviews
03:38:22 <tfukushima> We have some patches pending.
03:38:33 <tfukushima> #link Current Kuryr patches https://review.openstack.org/#/q/project:openstack/kuryr+status:open
03:38:53 <tfukushima> #action Everyone review the patches
03:39:28 <tfukushima> #action Everyone adds reviewers on submitting patches
03:39:47 <tfukushima> #topic Bugs
03:40:38 <tfukushima> #link Current Kuryr bugs https://bugs.launchpad.net/kuryr
03:41:08 <tfukushima> I also heard about pyroute2's IPDB issue on scaling cluster from banix.
03:41:20 <tfukushima> He's sick so we'd talk later.
03:41:49 <tfukushima> #action banix brings the detail of pyroute2's IPDB issue and discuss with tfukushima
03:42:16 <tfukushima> #topic Open Discussion
03:42:56 <tfukushima> fawadkhaliq: Do you have anything else you want to share? If you don't I'll finish the meeting.
03:43:04 <fawadkhaliq> tfukushima: negative
03:43:11 <fawadkhaliq> let's finish it :-)
03:43:25 <tfukushima> fawadkhaliq : Ok, thanks for attending.
03:43:45 <fawadkhaliq> thank you, bye!
03:43:48 <tfukushima> #endmeeting