18:14:26 #startmeeting networking_policy 18:14:27 Meeting started Thu Jun 19 18:14:26 2014 UTC and is due to finish in 60 minutes. The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:14:28 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:14:30 The meeting name has been set to 'networking_policy' 18:14:37 now it seems to be proper 18:14:47 #topic Mapping driver/data path patches 18:15:13 wow 18:15:16 ban hammer again! 18:15:35 SumitNaiksatam is turning bunch of people away everytime mapping driver is mentioned :-) 18:15:43 this is actually more fun; some get kicked out randomly 18:15:44 s3wong: hahaha 18:16:02 GPM API-1: #link https://review.openstack.org/#/c/100716/ 18:16:35 we just have to work through adversarial circumstances, i think we are used to it :-P 18:16:46 rkukura: did you get dropped and rejoin? 18:16:56 i think rkukura is back 18:17:08 got dropped 18:17:10 for some reason rkukura is always selected :-) 18:17:18 all the freenode rooms split 18:17:27 rkukura: you are on the spot as well 18:17:45 rkukura: that's right, can't blame OpenStack, happening to ODL also 18:17:46 rkukura: GPM-API-1 ^^^ 18:17:55 I think mandeep_ has been on the same side of the split as me, but doesn’t seem to be in this meeting 18:18:16 The patch SumitNaiksatam is in review 18:18:26 rkukura: is, great 18:18:40 The GPM-DB-1 patch is updated based on that and all Sumit’s recent updates in the GP-*-1 series 18:18:50 rkukura: okay 18:18:54 So I’m just finishing up the UTs and should post that for review today 18:19:02 rkukura: great thanks 18:19:11 I think the GPM-plugin-1 patch is trivial, so might get that out today too 18:19:30 Then the work on the drivers continues, so those should be in review by mid next week 18:19:54 rkukura: so (for the benefit of the folks here who havent looked at the GPM patch), the GPM API patch basically maps EP, EPG, L2/3 policy to the traditional neutron resources, right? 18:20:01 I’m traveling from 6/25 - 7/2, so my goal is to get the drivers into review beforehand 18:20:10 SumitNaiksatam: correct 18:20:28 rkukura: and we use attribute extension, right? 18:21:04 Thanks to support SumitNaiksatam added in his UTs, I was able to remove almost all UT code from GPM-API-1 18:21:41 The GPM-DB-1 patch will also reuse the GP-DB-1 UTs, but needs to add several additional UTs 18:22:13 Anyway, its been slow going, but I think the initial series should be very stable now, so less churn going forward 18:22:28 mapping RD->router, BD->subnet, etc? 18:23:06 LouisF: yes, though i made add RD is L3 policy now, and BD is L2 policy 18:23:10 LouisF: Yes, but RD is now L3P and BD is now L2P, and both use lists 18:23:14 made -> may 18:23:25 ok thx 18:23:45 EPG->subnets, EP->port 18:23:53 L2P->network 18:24:06 rkukura: thanks for the update (and also the reviews) 18:24:07 right 18:24:39 * SumitNaiksatam thinks rkukura is the master of three letter acronyms..ML2 was the first…now L2P, L3P :-P 18:24:49 So once this series is complete, the GP API will be usable to boot VMs, but all traffic will be allowed between EPGs in the same L3P 18:25:09 ;) 18:25:25 rkukura: okay good to know about the path forward 18:25:49 any other questions for rkukura on the mapping API patch? 18:25:54 Guys: getting out of the room for a moment, BRB 18:26:21 * SumitNaiksatam thinks s3wong is jumping out a strategic time ;-P 18:26:32 SumitNaiksatam: I don’t see it on the agenda, but are you getting started on the GP-*-2 series of patches? 18:26:52 rkukura: yes 18:27:08 rkukura: so that is the breadth series 18:27:43 the plan was to base that off the plugin-1 patch (until the mapping driver stuff comes along) 18:27:53 SumitNaiksatam: Makes sense 18:28:07 Aiming J-3 for those? 18:28:26 so breadth is something like API-1 -> DB-1 -> plugin-1 -> API-2 -> DB-2 -> plugin-3 18:28:39 banix: noooo…everything is aimed for J-2 :-) 18:28:56 we might not merge but at least lets try to get it out for review 18:29:06 We should aim to get at least GP-*-1 and GPM-*-1 merged before J-2 18:29:08 SumitNaiksatam: ok; i think that should be the goal for getting in Juno 18:29:09 banix: we should work towards the goal 18:29:25 banix rkukura: agree 18:29:35 CLI and Heat stuff is that for J2 too? 18:29:43 the thing is that the client and horizon patches have dependecies on the model patches 18:29:51 And getting as much of GP-*-2 and GPM-*-2 into review as possible by J-2? 18:29:52 and heat too, as prasadv points out 18:30:11 prasadv: lets discuss that in your part of the agenda 18:30:49 btw, correction from earlier, i meant to say: API-1 -> DB-1 -> plugin-1 -> API-2 -> DB-2 -> plugin-2 18:30:59 i think the GP-*-1s are ready; so we have to push for reviews to see if there are anything that need changing and move forward 18:31:07 SumitNaiksatam: What do you mean by “model patches”? 18:31:18 banix: totally 18:32:03 SumitNaiksatam, banix: Isn’t our strategy to get GPM-*-1 into review before pushing for cores to +2/+A GP-*-1? 18:32:15 rkukura: model patches being API/DB/Plugin-1/2/3 where 1 is for EP, EPG, L2P, L3P, 2 is for classifier, action and rules, and 3 is for Contracts and EPG/Contract association 18:32:30 rkukura: i think thats what banix meant 18:32:40 rkukura: in anticipation of your mapping patches landing 18:32:53 They are coming! 18:33:29 GP*-*-1 18:33:37 *-1 18:33:45 banix: :-) 18:34:10 we already covered UTs in the earlier update 18:34:45 i think we have agreement from the current set of reviewers regarding the current approach on the UTs 18:35:15 changes such as directly calling the DB layer are currently planned as an enhancement iteration 18:35:18 is s3wong back? 18:36:24 lets circle back to adv services in a bit 18:37:21 #topic CLI, Horizon update 18:37:48 prasadv: so any updates from hemanth on the CLI 18:37:58 prasadv: i know you are bit blocked my the API patches 18:38:11 no updates. We will get started on them soon 18:38:12 my -> by 18:38:18 prasadv: okay, np 18:38:31 prasadv: my understanding is that we are targeting J-2 even for this 18:38:41 is heat also planned for J2? 18:38:53 prasadv: we want to be able to give the reviewers all the tools to be able to review this in entirety 18:39:01 hemanth is in India. I will let him know 18:39:03 prasadv: may be we can go to J-3 for heat 18:39:14 is ronak here? 18:39:16 is there a patch for the heat work? 18:39:25 LouisF: yes 18:39:34 LouisF: old one 18:40:02 LouisF: #link https://review.openstack.org/93657 18:40:05 banix: true 18:40:13 thx 18:40:18 LouisF: this was written for the PoC 18:40:26 prasadv: so the plan is to build on this? 18:40:45 SumitNaiksatam: yes 18:41:08 susaant is not on the IRC. Will ask him start attending this 18:41:20 for heat 18:41:23 prasadv: ok 18:42:03 we might have heat for J2 but will confirm 18:42:27 prasadv: sweet 18:44:43 i dont think we have ronak here 18:45:09 ok moving on 18:45:22 is s3wong back? 18:45:46 Trouble in Texas! 18:45:57 may be banix can chime in 18:46:13 #topic Adv services integration 18:46:44 banix: have you heard for any progress? 18:46:58 banix: or rather any blockers? 18:47:05 Well, from yesterday meeting it looks like we will have flavors and hopefully steering moving forward 18:47:12 banix: okay 18:47:21 banix: but specifically for our GP integration 18:47:34 SumitNaiksatam: no 18:47:42 banix: we have the GP requirements on the agenda, but we are not able to bring it up 18:47:48 banix: ok 18:48:08 the only thing is that not sure about the status of service chaining 18:48:16 banix: ok 18:48:41 #action banix SumitNaiksatam to check on the status of service chaining support for GP 18:49:03 perhaps once we have the mapping driver framework in place, it will be easier to make progress with the services’ integration 18:49:37 * SumitNaiksatam thinks s3wong is busy guarding the Alamo or participating the victory parade :-P 18:49:54 we didnt get to talk about service insertion but the reviews seem to be progressing 18:50:01 banix: ok good 18:50:11 #topic Open Discussion 18:50:35 sorry my connection is a bit choppy, here so i am experience delays in the messages 18:50:46 anything else we want to discuss today? 18:51:18 or we can go back to catching up on the world cup! :-P 18:52:16 okay so be on the lookout for rkukura’s mappingpatches, they will land by tomorrow 18:52:30 and please also review my earlier patches 18:52:40 thanks all! 18:52:42 bye 18:52:45 #endmeeting