15:05:07 <Swami> #startmeeting distributed_virtual_router 15:05:08 <openstack> Meeting started Wed Jun 4 15:05:07 2014 UTC and is due to finish in 60 minutes. The chair is Swami. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:05:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:05:13 <openstack> The meeting name has been set to 'distributed_virtual_router' 15:05:19 <Swami> hi folks, 15:05:53 <Swami> mrsmith: hi 15:06:01 <mrsmith> hi hi 15:06:34 <Swami> ping armax 15:06:45 <armax> Hi Swami 15:07:14 <Swami> Agenda: 15:07:20 <Swami> DVR Progress update 15:07:47 <Swami> L3 Extension 15:07:56 <Swami> L3 Agent update 15:08:08 <Swami> ML2/OVS update 15:08:29 <Swami> L3 Extension - Update 15:08:39 <armax> Swami: I can go 15:09:00 <Swami> armax I did see that you have pushed patches on the L3 Extension. 15:09:08 <Swami> How is the review going. 15:09:12 <Swami> can you give an update. 15:09:48 <armax> so the patch in question is this one: https://review.openstack.org/#/c/84223/ 15:10:15 <armax> I did some preliminary refactoring in: https://review.openstack.org/#/c/97028/ and https://review.openstack.org/#/c/97275/ 15:10:32 <armax> as for patch 84223, it has gone through a number of reviews 15:10:38 <armax> I am now in the process of adding UT’s 15:10:52 <armax> and I think it’s then ready to go back into swami’s hands 15:11:19 <armax> reason being, there are a number of issues unvelied during the review cycles that I am not sure how to proceed forward 15:11:34 <armax> I expect to be done by the end of today, as for the UT’s 15:11:54 <Swami> armax: Thanks for your help and patch update 15:12:08 <armax> I tried the patch locally and it should get all the Jenkins’ jobs to pass 15:12:24 <Swami> Sorry I was a busy last couple of days, I can take a look at the review comments in a day or so. 15:12:29 <armax> we will miss functional testing 15:12:45 <armax> because I’d hate to go the ‘wrong’ way of testing API extensions 15:12:49 <armax> so for now I am sticking to UT proper 15:13:09 <armax> when functional testing in neutron is ready (something that marun_afk is working on) 15:13:13 <armax> I’ll get that done 15:13:31 <Swami> armax: yes that sounds reasonable. 15:13:32 <armax> Swami: ping me back when you have some time 15:13:57 <armax> that’s all from me 15:14:22 <Swami> armax: I did you that you have created an additional table for the handling the distributed routers, was that based on the review comments. 15:14:37 <armax> I did it for a number of reasons 15:15:01 <armax> we could have kept the extra column in the router table 15:15:23 <armax> but that was affecting plugins that did not support the extra attribute 15:15:43 <armax> also everything in the logic was cobbled together 15:15:56 <armax> making extensibility and maintainability a lot more difficult 15:16:15 <Swami> armax: Yes I realized it after seeing the patches. 15:16:23 <armax> now l3 base and l3 dvr are a lot more decoupled 15:16:33 <armax> which I think it’s beneficial 15:16:50 <Swami> armax: Yes I went that route in my first couple of patches and then moved back again. 15:17:12 <armax> also, I am planning of reconciling nsx’s implementation of distributed routing 15:17:20 <armax> so that we’ll have a unique extension point.. 15:17:29 <armax> but that’s for later 15:17:46 <Swami> armax: Ok thanks for the information. 15:17:53 <armax> Swami: what made you go back on your steps? 15:18:32 <armax> Swami: even though we can take this offline, I don’t want to hog the meeting 15:18:36 <Swami> armax: I heard from the Juno summit that for any attribute extensions we don't need to introduce additional extensions. 15:18:53 <Swami> That's what compeled me to go back to the original table instead of extensions. 15:19:23 <armax> is there a discussion on the ML? 15:19:48 <armax> there’s a discussion on idempotent DB migrations, but that’s slightly differnet 15:20:09 <Swami> Yes I sent out an email to Mark/Kyle asking for their feedback on this. 15:20:14 <armax> ic 15:20:31 <Swami> ok, I can discuss about this offline. 15:20:35 <armax> Swami: sure 15:20:49 <Swami> armax: Thanks for the update 15:21:13 <Swami> L3 Agent Update 15:21:24 <Swami> mrsmith: Any updates on the L3 Agent. 15:22:10 <mrsmith> we got SNAT sync'd up with Icehouse-ga 15:22:19 <mrsmith> started moving to juno/top 15:22:34 <Swami> mrsmith: great 15:22:35 <mrsmith> the merge doesn't look as bad as previous (ice-2 and ga) 15:22:50 <mrsmith> working on some corner-cases, etc 15:23:04 <mrsmith> as part of moving to ga, we got all unit tests passing 15:23:32 <mrsmith> not much else to report 15:23:40 <mrsmith> one question 15:23:47 <mrsmith> as far as the wip code 'patch' 15:23:49 <Swami> mrsmith: that's good news 15:24:06 <mrsmith> should we update that patch with juno/top or start a new submittal? 15:24:23 <Swami> you might have to push the code with the top. 15:24:34 <mrsmith> abandon the previous patch? 15:24:34 <carl_baldwin> mrsmith: Use the same patch. 15:24:36 <Swami> You don't need to start a new patch. Just upate the original one with the new one. 15:24:45 <mrsmith> ya - got it 15:24:56 <mrsmith> same patch, but updated to juno/top - right? 15:25:42 <Swami> Yes, use the same patch, just rebase to the new top of the branch code and then merge your changes and do a git review. 15:25:49 <mrsmith> k - sounds good 15:26:06 <Swami> mrsmith: Thanks mike 15:26:20 <mrsmith> swami - one outstanding issue is our rpc/plugin port binding issue 15:26:22 <mrsmith> that is ongoing 15:26:36 <mrsmith> you and murali are working that out right? 15:27:25 <Swami> mrsmith: I think we can move to the plugin instead of rpc, if there is not timing issue. 15:27:39 <mrsmith> k 15:28:08 <mrsmith> that is all from me 15:28:18 <Swami> mrsmith: For the FIP, we don't have any issue for the plugin as far as the plugin does the port binding. 15:28:35 <mrsmith> I think you're right - we have SNAT remaining 15:28:50 <Swami> For the SNAT I will check with Murali and confirm to you. 15:28:59 <mrsmith> great 15:29:14 <Swami> mrsmith: Thanks for your information. 15:29:18 <mrsmith> np 15:29:33 <Swami> I think vivek is on FTO, so we don't have any update on the L2 Agent. 15:29:59 <Swami> Is murali in here. 15:29:59 <Swami> ping murali 15:30:00 <mrsmith> FTO = vacation :) 15:30:25 <Swami> mrsmith: Thanks for the clarification 15:31:00 <Swami> Yes Yi in here 15:31:40 <Swami> rajeev: hi 15:32:00 <Rajeev> Swami: Hello 15:32:41 <Swami> rajeev: mrsmith just gave an update on the L3 Agent. 15:32:52 <Swami> Do you have any update to share. 15:33:45 <Rajeev> csnat is coming along well 15:34:17 <Rajeev> when is east-west expected t be in by? 15:36:05 <Rajeev> we can then start getting the n/s items in too 15:37:56 <mrsmith> did we lose Swami ? 15:38:02 <mrsmith> ping Swami 15:38:08 <chuckC> seems so 15:41:19 <chuckC> shift to open discussion until swami returns? 15:43:16 <mrsmith> swami is telling me he is having probs re-joining 15:43:23 <mrsmith> looking back at the agenda 15:43:29 <mrsmith> I think l3 extension was next 15:43:34 <mrsmith> but I think we already hit that 15:43:39 <mrsmith> right armax? 15:44:00 <armax> yes 15:44:04 <Swami> folks back on 15:44:04 <mrsmith> bam - there he is 15:44:16 <Swami> sorry guys I got disconnected. 15:44:20 <mrsmith> we hadn't gone far without you Swami :) 15:44:48 <Swami> mrsmith: Thanks for communicating. 15:45:02 <Swami> Next topic I wanted to discuss is the Services. 15:45:11 <Swami> Is 'Yi' in here. 15:45:43 <Swami> we were discussing with the FWaaS team about addressing the FWaaS service with the DVR 15:46:03 <Swami> There is a mailing thread that we started with the team. 15:46:29 <Swami> Rajeev: Did you hear anything back from Yi or Sumit on the FWaaS 15:46:57 <Rajeev> Swami: no, I haven't 15:47:21 <Rajeev> not sure I am on the thread 15:47:22 <Swami> ok, I will check with the FWaaS team tomorrow. 15:47:47 <Swami> rajeev: Thanks 15:48:36 <Swami> I will also check with "yi' on this. 15:49:08 <Swami> rajeev: before I got disconnected I was stating that we had plans to push the East-West first and move the North-South. 15:50:07 <Rajeev> Swami: that makes sense. Tentative timeline you can share ? 15:50:18 <Swami> If have the csnat work coming up, then we can think about pushing both East-West and North-South 15:56:28 <Rajeev> sure. Any thoughts on how the scheduler is to be brought in? 15:57:20 <chuckC> swami is gone again 16:00:36 <Sukhdev> Hello ML'ers 16:00:37 <Rajeev> Thanks for the meeting 16:01:01 <shivharis> hi all, DVR meeting is still in progress... 16:01:12 <chuckC> Sukhdev: previous meeting leader disconnected, so no #endmeeting 16:01:17 <emagana> ah ok.. 16:01:19 <banix> hi there 16:01:23 <emagana> I am here anyway! 16:01:30 <rkukura> hi 16:01:31 <Sukhdev> chuckC: opps 16:01:53 <chuckC> Sukhdev: opps? 16:01:54 <banix> #endmeeting 16:02:04 <Sukhdev> chuckC: can you type #endmeeting with the meeting name 16:02:25 <shivharis> hi all 16:02:32 <chuckC> #endmeeting distributed_virtual_router 16:02:51 <Swami> #endmeeting