03:01:03 <hongbin> #startmeeting zun
03:01:04 <openstack> Meeting started Tue Nov 14 03:01:03 2017 UTC and is due to finish in 60 minutes.  The chair is hongbin. Information about MeetBot at http://wiki.debian.org/MeetBot.
03:01:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
03:01:08 <openstack> The meeting name has been set to 'zun'
03:01:10 <hongbin> #link https://wiki.openstack.org/wiki/Zun#Agenda_for_2017-11-14_0300_UTC Today's agenda
03:01:42 <kevinz> o/
03:02:12 <hongbin> hi kevinz
03:02:21 <hongbin> it looks it is just you and me today
03:02:34 <kevinz> Aha
03:02:37 <kevinz> Hi hongbin
03:02:46 <hongbin> we could have a quick update then end the meeting shortly
03:02:53 <kevinz> great
03:02:59 <Shunli> o/
03:02:59 <kevinz> Shunli has join
03:03:04 <hongbin> hi Shunli
03:03:10 <Shunli> hi all
03:03:23 <hongbin> first, i would like to give an update from my side
03:03:34 <hongbin> last week, it is the openstack summit
03:04:05 <hongbin> i gave two talks at the summit, madhuri gave another one
03:04:33 <hongbin> the two talks i gave is good, i collected some feedback, and there are several attendees
03:04:35 <kevinz> I've seen the video:-)
03:04:40 <hongbin> :)
03:05:08 <kevinz> Looking forward to the feedback
03:05:13 <hongbin> in general, people are interesting in the NFV use cases
03:05:28 <hongbin> there are a few other NFV talks mentioned Zun project as well
03:05:44 <kevinz> good news
03:05:48 <hongbin> in addition, there are some requests for specific features
03:05:56 <hongbin> such as cinder volume multi-attach
03:06:27 <hongbin> madhuri gave a talk about clear container integration in zun
03:06:39 <hongbin> her session is also good, there are a lot of attendees
03:06:55 <hongbin> i mentioned both of you in the talks, hope you like it
03:07:04 <hongbin> any question about the summit ?
03:07:26 <kiennt26> o/ hi
03:07:44 <hongbin> hi kiennt26
03:08:01 <hongbin> kiennt26: we were doing the openstack summit recap
03:08:15 <hongbin> kiennt26: you were in the summit as well, have anything to share with the team?
03:08:30 <kevinz> hongbin: what do we need for supporting NFV now?
03:08:46 <hongbin> kevinz: there are a few high performance features
03:09:01 <hongbin> kevinz: for example, numa topology, cpu pinning, big page, etc.
03:09:26 <kiennt26> hongbin: No, i wasn't there. The guy you had met is my co-worker in Fujitsu.
03:09:39 <hongbin> kiennt26: oh, i see
03:09:44 <kevinz> cool, these are all supported on NOVA now
03:10:13 <hongbin> yes, basically, what they need is everything that is supported in vm, and move them to containers
03:10:40 <kevinz> cinder multiple-attach is also a good features.
03:10:48 <hongbin> yes
03:11:36 <hongbin> ok, that is all from my side
03:11:49 <hongbin> #topic Introduce container composition (kevinz)
03:11:56 <hongbin> kevinz: ^^
03:13:19 <kevinz> sorry
03:13:25 <kevinz> just on a call
03:13:46 <hongbin> kevinz: ok, should we move to the next topic ?
03:14:01 <kevinz> yeah:-)
03:14:11 <hongbin> kevinz: ok, will get back to you
03:14:17 <hongbin> #topic NFV use cases (lakerzhou, Shunli)
03:14:38 <hongbin> Shunli: have anything about this topic?
03:15:00 <Shunli> i didn't spend much time on this bp
03:15:15 <Shunli> as the basic function has ready.
03:15:22 <hongbin> i see
03:15:37 <Shunli> I worked on k8s recently as my work need.
03:15:46 <hongbin> oh, i see
03:16:07 <hongbin> as the bp is closed to finish, i think we are good
03:16:08 <Shunli> maybe can introduce some k8s to zun later.
03:16:19 <hongbin> sure
03:16:35 <Shunli> k8s/k8s feature
03:16:46 <hongbin> kevinz owns the k8s integration bp, perhaps you and kevin can work together for that
03:17:00 <hongbin> (if you interest)
03:17:16 <Shunli> sure, we look into the k8s connector bp
03:17:26 <hongbin> great
03:17:28 <Shunli> sorry, we/will
03:17:37 <hongbin> ok
03:17:57 <kevinz> I'm back
03:18:07 <hongbin> #topic Introduce container composition (kevinz)
03:18:20 <kevinz> Warmly welcome Shunli to work on k8s connector:-)
03:18:29 <hongbin> :)
03:18:30 <kevinz> we can do it together
03:18:39 <Shunli> kevinz: :-)
03:19:01 <kevinz> I'm trying the cinder attach for capsule last week.
03:19:35 <kevinz> Will give a patch to support it this week about volume attach.
03:19:46 <hongbin> awesome
03:20:25 <kevinz> For K8s connector. I'm still meditating the python multiple process frame work
03:20:56 <hongbin> yes, this might be the tough part
03:21:08 <kevinz> As this connector will simultaneously handle the request both from zun and k8s
03:21:35 <kevinz> I will check kuryr-kubernetes for some reference
03:21:51 <hongbin> sounds good
03:22:04 <kevinz> That's all from my side
03:22:10 <hongbin> thanks kevinz
03:22:19 <kevinz> hongbin: yw
03:22:47 <hongbin> kiennt26: hi kien, do you have anything to discuss with the team ?
03:23:01 <hongbin> kiennt26: btw, great work for the shared network implementation!
03:23:34 <kiennt26> hongbin: i have one problem with Introduce Compute filter patch
03:23:42 <hongbin> kiennt26: go ahead
03:23:51 <hongbin> what is the problem?
03:24:04 <kiennt26> #link https://review.openstack.org/#/c/518685/
03:24:14 <kiennt26> the broker was blocked due to memory limit
03:24:17 <kiennt26> http://logs.openstack.org/85/518685/2/check/zun-tempest-docker-sql/e6d649b/logs/rabbitmq/rabbit@ubuntu-xenial-rax-dfw-0000833630.txt.gz
03:24:36 <kiennt26> so that all tempest tests failed
03:25:14 <kiennt26> #link http://logs.openstack.org/85/518685/2/check/zun-tempest-docker-sql/e6d649b/logs/screen-zun-api.txt.gz#_Nov_13_04_32_58_163426
03:25:52 <hongbin> i see, that is from the zun-api side
03:26:38 <hongbin> kiennt26: i would need some time to look into the details, we can continue it after the meeting
03:26:52 <kiennt26> hongbin: ok, thanks
03:27:02 <hongbin> all, anything else to discuss ?
03:27:28 <kevinz> Nothing from me
03:27:36 <hongbin> ok, all, thanks for joining the meeting, see you next time
03:27:39 <hongbin> #endmeeting