16:00:16 <mestery> #startmeeting networking_ml2 16:00:17 <openstack> Meeting started Wed Feb 26 16:00:16 2014 UTC and is due to finish in 60 minutes. The chair is mestery. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:21 <openstack> The meeting name has been set to 'networking_ml2' 16:00:22 <rkukura> hi! 16:00:33 <HenryG> hi 16:00:33 <mestery> rkukura: Good morning man! 16:00:40 <mestery> HenryG: Same to you, morning! 16:00:51 <matrohon> hi 16:00:58 <mestery> matrohon: Good afternoon :) 16:01:17 <mestery> So, we have a lighter agenda today, lets see if we can plow through things! 16:01:19 <mestery> #link https://wiki.openstack.org/wiki/Meetings/ML2 Agenda 16:01:32 <mestery> I'll give folks a minute to join in here as well. 16:01:36 <amotoki_> hi 16:01:39 <pcm_> hi 16:01:46 <yamamoto> hi 16:02:00 * otherwiseguy waves from rkukura's cube 16:02:02 <mestery> yamamoto pcm_ amotoki_: Hello! 16:02:07 <mestery> otherwiseguy: Heh :) 16:02:12 <irenab> hi 16:02:12 <mestery> OK, lets get started. 16:02:18 <mestery> irenab: hi! 16:02:23 <mestery> #topic Action Item Review 16:02:31 <mestery> First, I wanted to extend a thanks to all for reviews the past week! 16:02:42 <mestery> We got all 3 of the SR-IOV patches merged! 16:02:45 * mestery claps loudly. 16:02:56 * HenryG claps louder 16:03:06 <amotoki_> nice. thanks all 16:03:11 <mestery> Great work by all reviewers and patch submitters (irenab and rkukura)! 16:03:25 <mestery> So, just a note: 16:03:29 <mestery> rkukura posted his port binding doc 16:03:37 <mestery> #link https://docs.google.com/document/d/1k8tAqfQr8Ujzx5TzpXTYEUaym-0U-pE3YL9wjrS8cDg/edit#heading=h.cpck9ip3dbdv ML2 Port Binding Document 16:03:38 <Sukhdev> Hi 16:03:45 <mestery> Please review and provide feedback in the google doc 16:03:58 <rkukura> The 1st part of that is in review now 16:04:06 <mestery> rkukura: Great, thanks! 16:04:13 <matrohon> https://review.openstack.org/#/c/76363/ 16:04:15 <mestery> One more note from last week: 16:04:17 <mestery> All ML2 MechanismDrivers which bind ports should ensure they look to make sure the binding:vnic_type attribute is something they handle. 16:04:23 <matrohon> thanks bob 16:04:27 <rkukura> The new PortContext properties are in https://review.openstack.org/#/c/76363/ 16:04:55 <mestery> Lets focus on this review this week, this is an important one to merge. 16:05:20 <mestery> #topic Migration BP 16:05:26 <rkukura> The 2nd part will be a bit more complicated, rearranging the transactions. 16:05:30 <mestery> Looks like marun posted a patch for this already! 16:05:33 <mestery> https://review.openstack.org/#/c/76533/ 16:05:59 <mestery> great work marun (who can't be here for the meeting unfourtanetly)! 16:06:02 <zzelle> Hi 16:06:09 <mestery> Getting eyes on this one will also be very important. 16:06:15 <mestery> So please spend some cycles reviewing and testing this out if you can! 16:06:28 <mestery> Any questions on the migration? 16:07:11 <amotoki_> do we need configuration migration? 16:07:13 <shivharis> the 2 patches are on the wiki 16:07:15 <shivharis> ? 16:07:42 <mestery> shivaris: For migration? The patch is here: https://review.openstack.org/#/c/76533/ 16:07:56 <mestery> amotoki_: What do you mean by configuration migration? 16:08:10 <rkukura> amotoki_: My thinking is migrating configuration should be handled by things like the puppet modules and higher level tools. 16:08:11 <shivharis> no i was slow, i meant the two above 16:08:38 <mestery> shivharis: The first one is, the second one isn't yet. 16:08:52 <amotoki_> we need some configuration change from ovs/lb to ml2. 16:09:18 <amotoki_> can we assume that configruatoin tools cover it? 16:09:30 <mestery> amotoki_: I think so, yes, like rkukura indicated above. 16:09:57 <amotoki_> it looks okay to me too. 16:10:05 <mestery> amotoki_: Cool, thanks! 16:10:13 <rkukura> looks like marun's script has instructions on manually migrating the config 16:10:35 <amotoki_> there is a question in the commit message of marun patch. 16:11:14 <amotoki_> a question on config migration. 16:11:19 <mestery> Yes, I see the question amotoki_. 16:11:31 <mestery> Lets reply in the review and continue the discussion there perhaps since marun isn't here? 16:12:20 <mestery> OK, moving on. 16:12:28 <mestery> #topic Port binding bugs affecting MDs 16:12:37 <mestery> We covered one of these: https://bugs.launchpad.net/neutron/+bug/1276395 16:12:45 <mestery> The second bug is this one: https://bugs.launchpad.net/neutron/+bug/1276391 16:12:56 <mestery> We as a team should work to merge rkukura's patch for the first one this week. 16:13:10 <rkukura> The patch in review partially resolves https://bugs.launchpad.net/neutron/+bug/1276395 16:13:27 <mestery> rkukura: OK, thanks! 16:13:40 <rkukura> It defines the new properties, but doesn't fix making them available when a port is deleted. 16:14:09 <rkukura> That will be completed as part of the upcoming patch for https://bugs.launchpad.net/neutron/+bug/1276391, which rearranges the code. 16:14:19 <mestery> OK, cool, thanks for the update rkukura! 16:14:58 <mestery> Any questions on these for rkukura? 16:15:05 <shivharis> rkukura, kmestery: this causes issues for 16:15:16 <shivharis> new plugins and the deadline of Mar 3 16:15:22 <rkukura> Please provide feedback on the google doc (or email thread) sooner rather than later! 16:15:33 <mestery> shivharis: Yes, exactly. 16:15:37 <shivharis> plugins -> mds 16:15:39 <mestery> There is a merge order here for new MDs 16:16:14 <matrohon> rkukura : should I wait for your patch to get mine merged : https://review.openstack.org/#/c/75037/ 16:16:32 <rkukura> My recommendation is, if an MD can be merged before my 2nd patch, get it in! I can update the MD with any driver API changes 16:16:42 <matrohon> I left a TODO in it 16:17:12 <matrohon> rkukura : thanks 16:17:17 <rkukura> Can we handle TODOs like this as bug fixes after March 3? 16:17:34 <mestery> matrohon: Your patch looks simple enough to merge prior to rkukura's. 16:17:45 <shivharis> rkukura: please let mark be aware of the that point 16:18:15 <matrohon> mestery : hope so, and it will be easy to backport for havana 16:18:26 <mestery> matrohon: Agreed. I just gave +2 in fact. :P 16:18:38 <mestery> OK, any more questions on the port binding bugs by anyone? 16:19:44 <mestery> #topic Open Discussion 16:19:51 * mestery wasn't kidding when he said this meeting would be short. :) 16:19:59 <mestery> Anything else anyone wants to discuss in the context of ML2? 16:20:14 <irenab> I wanted to ask for review for proposed MD: https://review.openstack.org/#/c/72854/ 16:20:20 <amotoki_> Could you take a look at https://bugs.launchpad.net/neutron/+bug/1282925 ? 16:20:44 <amotoki_> i found it while debugging nec plugin but it affects all plugins including ML2. 16:21:14 <mestery> irenab: Yes, thanks for bringing that up! 16:21:26 <asomya> Can someon eyeball this BP i filed: https://blueprints.launchpad.net/neutron/+spec/ml2-type-driver-refactor 16:21:31 <mestery> amotoki_: Good call! 16:21:43 <rkukura> amotoki_: I'll look at whether the transaction changes I'm planning will address that bug. 16:21:51 <Sukhdev> mestery: rkukura: can you please give few cycles to https://review.openstack.org/#/c/73482/ 16:22:02 <mestery> asomya: I just made myself approver for that, but it will have to be targeted at Juno, is that ok? 16:22:18 <asomya> @mestery: thanks, yeah I'm targeting it for J-1 16:22:19 <shivharis> asomya: is that for juno? 16:22:27 <asomya> @shivharis: yes 16:22:28 <mestery> asomya: Cool! 16:22:48 <shivharis> asomya: relief, could affect mds 16:22:49 <mestery> Sukhdev: Will look at that one as well. 16:23:06 <rkukura> Sukhdev, irenab, mestery: I definitely have some reviews to catch up on! 16:23:06 <mestery> It woudl be good for folks to spend lots of cycles reviewing all of this code this week. 16:23:10 <Sukhdev> Regarding Third party testing, I posted this wiki yesterday - you may want to check out https://wiki.openstack.org/wiki/Arista-third-party-testing 16:23:36 <asomya> shivharis: It could potentially affect MD's, I'm trying my best not to touch the context objects supplies to the MD's 16:23:51 <irenab> Sukhdev: Thanks. The support for recheck very helpful 16:24:00 <asomya> Most of the work is just Type manager and Type drivers 16:24:00 <shivharis> sukhdev: liked it 16:24:01 <rkukura> zzelle: Do you think https://review.openstack.org/#/c/71904/ is ready for i-3? 16:24:08 <zzelle> hi everyone 16:24:27 <Sukhdev> irenab: shivharis: Thanks 16:24:31 <zzelle> good question, i think so but i need your feed backs :) 16:24:54 <amotoki_> Re third party testing, i wrote the blog post about how to migrate to zuul. http://ritchey98.blogspot.jp/2014/02/openstack-third-party-testing-how-to.html 16:25:10 <mestery> amotoki_: Cool, thanks for sharing! 16:26:09 <zzelle> rkukura, there is also the review https://review.openstack.org/74055 which provides the same feature for tunnels (quite the same implementation) 16:26:21 <Sukhdev> amotoki_: good information 16:26:51 <rkukura> zzelle: right - I think both of these are pretty close to ready 16:27:48 <shivharis> kmestery, rkukura: please add https://review.openstack.org/#/c/60129/ to your review cycles 16:27:50 <mestery> rkukura: So, shoudl we focus on getting those in? Will they require a FFE? 16:28:00 <mestery> shivharis: Dually noted ;) 16:28:12 <rkukura> shivharis: Will do. 16:28:17 <amotoki_> I'd like to confirm the approval criteria of new MDs. In my understanding, we need to confirm thrid party testing works before approving the patch. right? 16:28:40 <mestery> amotoki_: Yes 16:28:41 <mestery> that's correct 16:28:43 <rkukura> mestery: I think zzelle's patches are on track - they had code in review before the feature proposal deadline I think 16:28:45 <mestery> We can't approve them until that happens. 16:29:00 <mestery> rkukura: OK, good to know, so then we can focus on iterating on reviews and merge when ready I think. 16:29:14 <zzelle> rkukura, mestery, the BP was proposed before the freeze 16:29:27 <amotoki_> As a review, I wonder whether I can give +2 or +1 when the code is ready. 16:29:50 <mestery> amotoki_: I think +2 with a comment about not approving until 3rd party testing is ready is appropriate. 16:30:20 <yamamoto> what should third party testing for unmerged md test? master + locally applied md? 16:30:30 <amotoki_> mestery: yeah. sounds good. we should have same policy to give +2. thanks for clarifying. 16:30:46 <mestery> amotoki_: Yes, agreed. 16:31:02 <irenab> yamamoto: till merged, we check the MD patch only 16:31:26 <amotoki_> yamamoto: your MD patch against the master branch sounds good. 16:31:56 <mestery> I think that's all it can test until it merges yamamoto. 16:32:03 <mestery> OK, anything else for this week? 16:32:12 <mestery> If not, lets focus on reviews for the rest of this week! 16:32:43 <amotoki_> it's midnight in japan. shorter meeting is really nice :-) 16:32:48 <mestery> :) 16:32:57 <mestery> OK, thanks for all your reviews, lets keep them coming! 16:33:04 <Sukhdev> amotoki_: wow!! Good Night 16:33:06 <mestery> Communicate in-channel or on the ML with questions. 16:33:09 <mestery> Thanks all! 16:33:10 <mestery> #endmeeting