16:02:04 <Sukhdev> #startmeeting networking_ml2
16:02:11 <openstack> Meeting started Wed Jul 20 16:02:04 2016 UTC and is due to finish in 60 minutes.  The chair is Sukhdev. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:02:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:02:15 <openstack> The meeting name has been set to 'networking_ml2'
16:02:35 <Sukhdev> can I see a show of hands - who is here for ML2 meeting?
16:02:53 <yamahata> hello
16:03:25 <Sukhdev> I will try to go slow to give others time to sign in
16:03:27 * rkukura raises hand
16:04:11 <Sukhdev> #topic: Agenda
16:04:18 <Sukhdev> #link: https://wiki.openstack.org/wiki/Meetings/ML2#Meeting_July_20.2C_2016
16:05:18 <Sukhdev> #topic: Announcements
16:05:31 <Sukhdev> N-2 was released last week
16:05:37 <Sukhdev> N-3 is on Sept 1
16:06:04 <Sukhdev> Did you know about the OpenStack Project Mascot (logo) ?
16:06:20 <Sukhdev> well, they are coming to a theater near you -
16:06:32 <Sukhdev> see - http://lists.openstack.org/pipermail/openstack-dev/2016-July/099046.html
16:07:19 <Sukhdev> If you have strong feeling about it or want to provide feedback for Neutron logo - feel free to raise your voice
16:07:47 <rkukura> Sukhdev: Is there an etherpad with proposals for neutron’s mascot?
16:08:07 <Sukhdev> rkukura : I have not looked for it
16:08:29 <Sukhdev> does anybody know about it?
16:09:35 <Sukhdev> If you missed last week - Neutron Mid Cycle is mid of next month
16:09:54 <Sukhdev> that is all from me -
16:10:05 <Sukhdev> anybody has any other announcements?
16:10:47 <Sukhdev> moving on
16:10:58 <Sukhdev> #topic: FREs
16:10:58 <rkukura> release name polls for P ends today and Q in a couple days
16:11:04 <Sukhdev> #undo
16:11:05 <openstack> Removing item from minutes: <ircmeeting.items.Topic object at 0x7f5bbf2ca450>
16:11:26 <Sukhdev> rkukura : thanks for the reminder -
16:11:55 <Sukhdev> please do vote your favorite names - P has selection of 23 names and Q has selection of 17
16:12:00 <Sukhdev> huge selection
16:12:22 <Sukhdev> any other announcements?
16:12:54 <Sukhdev> #topic: Bugs/RFEs
16:13:30 <Sukhdev> Unifying ML2 port bindings -
16:13:49 <Sukhdev> andreas is missing today
16:14:03 <Sukhdev> rkukura : you were going to review the patches - any update?
16:14:57 <rkukura> I started looking through https://review.openstack.org/309416 but didn’t get to a doing a detailed review yet
16:16:03 <Sukhdev> rkukura: schedules get hectic - totally understandable
16:16:24 <Sukhdev> anybody has any input/update on this?
16:17:45 <Sukhdev> OK on to the next topic
16:18:04 <Sukhdev> Driver API requests
16:18:18 <Sukhdev> I kept this on the agenda from previous week
16:18:27 <Sukhdev> because yamamoto was not here -
16:18:44 <Sukhdev> he is not here today either - probably on summer vacation
16:19:04 <Sukhdev> Do we want to cover any of these ?
16:19:37 <Sukhdev> Anybody wants to go through these or shall we defer these to next week when yamamoto is here
16:19:57 <rkukura> I’m curious what others think about https://bugs.launchpad.net/neutron/+bug/1592238
16:19:58 <openstack> Launchpad bug 1592238 in neutron "[RFE] ml2: needs a way for MD to pass data from precommit to postcommit" [Wishlist,Won't fix] - Assigned to Li Ma (nick-ma-z)
16:20:54 <rkukura> armax did not think it wss a good idea, and the author abandoned the patch and the bug is now in won’t fix state
16:21:24 <rkukura> But it does seem useful to me
16:22:05 <rkukura> I’m wondering if any MDs have been simply adding their own properties to the driver Context instances to pass data from precommit to postcommit?
16:22:18 <Sukhdev> rkukura : I wonder about the use case -
16:23:05 <Sukhdev> rkukura : one of the way probably is to put information in the DB during pre-commit and consume it in post-commit
16:23:17 <rkukura> That is possible too
16:24:15 <rkukura> If we were to ever make postcommit async, then we might not be passing the same Context instance, and any MD tacking on properties would break
16:24:38 <rkukura> But if we had an explicit dict for this, we could preserve it
16:25:22 <rkukura> But this could definitely be abused, for example to communicate between different MDs
16:25:46 <Sukhdev> With DB approach that I mentioned, one would be covered - even for async case as well, no?
16:25:58 <rkukura> Sukhdev: Sure
16:26:06 <rkukura> So I’m OK with leaving this closed for now, but wouldn’t mind revisiting it if a strong use case came up.
16:26:38 <rkukura> I think we can leave this one off next week’s agenda unless someone wants to re-open it
16:26:46 <Sukhdev> right, if there is a use case, then we should definately revisit -
16:27:06 <Sukhdev> I think armax is probably thinking on the same lines as well
16:28:18 <Sukhdev> anything else?
16:28:32 <rkukura> not from me
16:28:34 <Sukhdev> #topic: Open Discussion
16:28:47 <Sukhdev> any general discussion item....
16:29:15 <rkukura> only thing that comes to mind would be how to make these meetings more useful - meant to discuss that with you last week
16:29:53 <Sukhdev> right - that's why I took the poll a while back as well
16:30:41 <rkukura> nothing specific to propose on this right now
16:30:42 <Sukhdev> any suggestions? I am all ears
16:31:37 <rkukura> But these would be a lot more useful if contributers who are making significant changes to ML2 would attend.
16:33:09 <Sukhdev> I used make noise about this in Neutron meeting - perhaps we need to do it
16:33:36 <rkukura> I don’t have anything more to add on this right now - maybe we should discuss with the PTL at some point
16:34:28 <Sukhdev> that is a good idea
16:34:52 <rkukura> might as well wrap it up for today
16:34:59 <Sukhdev> I will bring it up with armax
16:35:09 <Sukhdev> yup - I have nothing more to add
16:35:18 <Sukhdev> Thanks for attending
16:35:23 <rkukura> please let me know how that goes
16:35:28 <Sukhdev> will do
16:35:39 <rkukura> thanks for chairing today!
16:35:39 <Sukhdev> bye
16:35:46 <Sukhdev> np
16:35:55 <Sukhdev> #endmeeting