16:00:54 <Sukhdev> #startmeeting networking_ml2 16:00:54 <openstack> Meeting started Wed May 11 16:00:54 2016 UTC and is due to finish in 60 minutes. The chair is Sukhdev. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:55 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:58 <openstack> The meeting name has been set to 'networking_ml2' 16:01:39 <Sukhdev> #topic: Agenda 16:01:47 <Sukhdev> #link: https://wiki.openstack.org/wiki/Meetings/ML2#Meeting_May_11.2C_2016 16:02:35 <Sukhdev> who is here to attend ML2 meeting this morning - I do not see usual suspects on the channel 16:03:22 <sadasu> Hello! 16:03:32 <Sukhdev> sadasu : hi 16:03:57 <Sukhdev> #topic: Announcements 16:04:27 <Sukhdev> rkukura is running late this morning 16:04:39 <Sukhdev> Anybody has any announcement to make? 16:05:38 <yamahata> hello 16:05:57 <Sukhdev> yamahata : hi 16:06:26 <Sukhdev> #topic: ML2 Charter for Newton 16:06:45 <Sukhdev> Based upon discussion last week following two Items I added to the agenda 16:07:11 <Sukhdev> Trunked ports integration with ML2 16:07:49 <Sukhdev> Integration of the segments service plugin with ML2 16:08:18 <amotoki> hi 16:08:42 <Sukhdev> did I miss anything or would anybody like to see this team cover anything additional for this release cycle? 16:08:45 <Sukhdev> amotoki : hi 16:10:04 <amotoki> regarding trunk port support, after the discussion at austin, we need an response from mechanism drivers for trunk port 16:10:22 <amotoki> so i think it fits ML2 part. 16:10:43 <rkukura> Hi - sorry I’m late 16:10:48 <amotoki> it includes an interface definition or possible more. 16:11:14 <Sukhdev> amotoki : correct - that is why I think we need to drive this in this team 16:11:56 * tbachman waves at rkukura 16:12:10 <Sukhdev> amotoki : other than Rosella, is there anybody else who should we be discussing this with? 16:12:53 <Sukhdev> rkukura : hi 16:13:21 <rkukura> Sukhdev: Sounds like routed networks? 16:13:23 <amotoki> Sukhdev: i think rosella and armando are stakeholders and sponsor cores. 16:13:52 <amotoki> I believe all of them are on board. you can count me too. 16:14:35 <rkukura> or still on trunked ports? 16:14:40 <Sukhdev> yes, we all were there on Friday had an excellent discussion - so, we are good to go on that front 16:14:51 <Sukhdev> rkukura : trunked ports 16:15:03 <amotoki> rkukura: discussing the charter of the team 16:15:48 <Sukhdev> I will follow up with rosella off line and get the status as to where she is and update the team 16:16:02 <Sukhdev> Next Item on the charter - 16:16:27 <Sukhdev> Integration of the Segments service plugin with ML2 - 16:16:33 <Sukhdev> carl_baldwin : are you here 16:16:51 <carl_baldwin> Sukhdev: I'm about 20% here. :) 16:17:05 <Sukhdev> we'll take it :-) 16:17:46 <Sukhdev> carl_baldwin: based upon discussion in Friday's session in Austin, I updated the RFE 16:18:05 <Sukhdev> for the manila-neutron integration use case 16:18:09 <carl_baldwin> Sukhdev: Do you happen to have a link to that RFE handy? 16:18:21 <Sukhdev> just a sec - looking 16:18:54 <Sukhdev> carl_baldwin : https://bugs.launchpad.net/neutron/+bug/1573197 16:18:55 <openstack> Launchpad bug 1573197 in neutron "[RFE] Neutron API enhancement for visibility into multi-segmented networks" [Undecided,Incomplete] 16:19:00 <Sukhdev> #link: https://bugs.launchpad.net/neutron/+bug/1573197 16:19:10 <carl_baldwin> Thanks, I'll take a look. 16:20:02 <amotoki> re: routed networks, i wonder which is better, ML2 folks join the meeting specific to routed network, or discuss routed network support in ML2 meeting. 16:20:04 <amotoki> thought? 16:20:31 <carl_baldwin> amotoki: That's a good question. 16:21:15 <carl_baldwin> We came up with some things to discuss yesterday in our meeting. I imagine there will be a small group of ML2ers that could help us out with that discussion. 16:22:14 <Sukhdev> carl_baldwin : when does routed network meeting takes place? 16:22:36 <amotoki> http://eavesdrop.openstack.org/#Neutron_Routed_Networks_Meeting 16:22:42 <carl_baldwin> Sukhdev: https://wiki.openstack.org/wiki/Meetings/Neutron-Routed-Networks 16:22:45 <carl_baldwin> amotoki: Thanks. 16:23:14 <carl_baldwin> Sukhdev: Be careful, the meeting alternates each week, mostly due to the Neutron meeting alternating around the same time. 16:23:55 <Sukhdev> carl_baldwin : it is not PST friendly :-), but, I will try to attend 16:24:19 <carl_baldwin> Basically, what we're looking for is some discussion around integrating the segments service plugin with ML2. Mainly to create and delete segments on an existing networ. 16:24:19 <amotoki> IMHO it would be best if all folks interested in the effort can join the routed network meeting. 16:24:22 <carl_baldwin> network. 16:25:16 <Sukhdev> I will start to attend the routed network meeting and then update the ML2 team about the status 16:25:22 <carl_baldwin> Agreed, it is early for PST. Maybe I could start an ML thread to start out with. Then, if we need higher bandwidth, we can catch the next routed networks meeting that starts at 1500. 16:25:38 <rkukura> I will try to as well, or at least read the minutes. 16:26:05 <carl_baldwin> rkukura: Sukhdev: That would be great. 16:26:20 <Sukhdev> carl_baldwin : yes that makes sense - 16:26:40 <carl_baldwin> Watch for an email with [Neutron][ML2] in the next 24 hours. 16:26:54 <carl_baldwin> Maybe [Routed Networks] too. 16:27:08 <Sukhdev> carl_baldwin: sounds good - will do 16:27:33 <carl_baldwin> It isn't the most urgent thing on our agenda. But, you know how the cycle slips by if you put things off for too long. 16:28:03 <andreas_s> I have a question on that 16:28:21 <andreas_s> at the summit we discussed also the multi port binding stuff 16:28:34 <andreas_s> do you think this is a combined effort? 16:28:47 <andreas_s> externalizing the segemtns and externalising multiple bindings? 16:28:58 <andreas_s> #link https://review.openstack.org/309416 16:29:13 <andreas_s> there's this spec for multiple portbinding (which I still need to update...) 16:29:21 <rkukura> I see those as orthogonal 16:30:00 <andreas_s> ok, so keep them separated? 16:30:28 <Sukhdev> andreas_s : I have not see this spec earlier - does it propose a multi-segmented networks to achieve multi-port binding? 16:31:01 <rkukura> I had not seen the spec either, so my comment may be premature 16:31:19 <andreas_s> Sukhdev, That's basically what we discussed at the Fridays meetup 16:31:35 <andreas_s> the conclusion was to have a new subresource or so for a port 16:32:02 <rkukura> I think the multi-segment should provider read/write access to the static segments of a network, and also read access to the dynamic segments, which get created during port binding 16:32:20 <rkukura> But the exact details of the port binding, especially for a distributed port, are more than that 16:33:19 <amotoki> i think you first need to file RFE to clarify what we need. what does multi-port binding bring to users? 16:33:31 <Sukhdev> andreas_s : let me review this spec and then we can decide if these are different or same 16:33:45 <Sukhdev> amotoki : +1 16:34:17 <andreas_s> Sukhdev, ok thanks - please keep in mind - it does not reflect the state after our summit discussion 16:34:33 <andreas_s> Sukhdev, I sitll have the todo to update it... 16:34:56 <andreas_s> Let me ping you as soon as I did it 16:35:15 <Sukhdev> andreas_s : understood - also, as amotoki points, can you please file an RFE as well? 16:35:22 <rkukura> andreas_s: please let me know as well 16:35:26 <andreas_s> hopefully end of this week 16:35:36 <andreas_s> yep will do and create a rfe 16:35:46 <Sukhdev> sounds good 16:36:00 <andreas_s> by the way I'm scheuran - I decided to change my irc nick to something more obvoius ;) 16:36:43 <Sukhdev> andreas_s : I noticed - It was good meeting you in person in Austin 16:37:13 <andreas_s> Sukhdev, definitively! 16:37:34 <Sukhdev> So, continuing our discussion on the team charter - 16:38:06 <Sukhdev> we covered three items so far 16:38:33 <Sukhdev> can anybody think of anything else that this team should be driving for Newton? 16:38:48 <amotoki> do we need to cover baremetal and manila stuffs here? 16:39:20 <amotoki> i am not sure what we need to do but you Sukhdev knows more :) 16:39:26 <Sukhdev> amotoki : we have weekly meetings to cover baremetal (ironic-neutron) meeting 16:39:43 <amotoki> i know. 16:40:31 <amotoki> so there is no or few ml2 specific topics. right? 16:40:54 <Sukhdev> and, for item manila, the RFE that I posted should cover it 16:41:21 <Sukhdev> amotoki : yes, it seems like we are down to 2 or 3 topics :-) 16:41:58 <amotoki> sounds good to me. thanks for clarification 16:42:56 <Sukhdev> Moving right along the agenda 16:43:02 <Sukhdev> #topic: Bugs 16:43:26 <Sukhdev> https://wiki.openstack.org/wiki/Meetings/ML2#Meeting_May_11.2C_2016 16:43:47 <Sukhdev> andreas_s : you had added this to the agenda - is there anything you want to discuss about it? 16:44:06 <andreas_s> Sukhdev, no, just make your aware of this... 16:44:11 <Sukhdev> opps - sorry - wrong link 16:44:27 <Sukhdev> https://bugs.launchpad.net/openstack-manuals/+bug/1567792 16:44:29 <openstack> Launchpad bug 1567792 in openstack-manuals "Mitaka Configuration Reference Nework section is missing ML2 parameters" [Undecided,Confirmed] 16:44:54 <Sukhdev> andreas_s : thanks 16:45:03 <andreas_s> I try to work on that... to get it fixed 16:45:26 <Sukhdev> andreas_s : let us know if you need any help with this 16:45:34 <andreas_s> sure, thanks! 16:45:44 <Sukhdev> moving right along 16:45:54 <Sukhdev> #topic: Open Discussion 16:46:30 <Sukhdev> carl_baldwin : you wanted to discuss the routed network items here 16:46:41 <Sukhdev> carl_baldwin : this is good time 16:46:51 <carl_baldwin> Sukhdev: I think we'll start with the ML as discussed above. 16:47:06 <carl_baldwin> Unless someone has some thoughts about create / delete segments now. 16:47:42 <Sukhdev> rkukura : any thoughts? 16:48:12 <rkukura> Sukhdev: I think starting with the ML makes sense 16:48:32 <Sukhdev> sounds good 16:48:51 <Sukhdev> carl_baldwin : look forward to an email from you then 16:49:13 <Sukhdev> Anybody wants to discuss anything - this is Open Discussion time 16:50:20 <Sukhdev> silence is golden - it also means that we are done 16:50:49 <Sukhdev> Thanks folks for attending today's meeting 16:50:57 <rkukura> thanks Sukhdev! 16:51:01 <Sukhdev> bye 16:51:09 <carl_baldwin> bye and thanks! 16:51:13 <amotoki> thanks 16:51:25 <rkukura> bye 16:51:26 <Sukhdev> #endmeeting