18:35:39 #startmeeting Networking FWaaS 18:35:41 Meeting started Wed Nov 4 18:35:39 2015 UTC and is due to finish in 60 minutes. The chair is SridarK. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:35:42 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:35:43 Hi 18:35:44 The meeting name has been set to 'networking_fwaas' 18:35:57 thanks! 18:36:02 #chair xgerman sc68cal SridarK 18:36:03 Current chairs: SridarK sc68cal xgerman 18:36:31 #topic Bugs 18:37:29 Not sure if anything new came up - not on my radar 18:37:47 bharathm: we did get some comments on the FIP NS bug 18:37:47 well, still need some +2 on https://review.openstack.org/#/c/231246/ 18:38:03 xgerman: my bad - will get it in 18:38:09 thanks 18:38:24 SridarK: Yes and I replied more or less with the same info as your comment 18:38:48 bharathm: yes - we shd not be making changes in l3 agent 18:38:56 +! 18:39:02 and i agree - we shd plan for the refactor 18:39:27 i think brocade guys also offered to help, but u wanted to work on this as well correct ? 18:39:44 bharathm: if so i can work with u 18:39:54 I think we need to divvy up the work better 18:39:57 SridarK: yeah.. Is there any IRC/etherpad discussing the observer hierarchy 18:40:16 bharathm: no but pc_m has done this for vpn 18:40:29 xgerman: yes i agree 18:41:40 xgerman: IMO, bharathm: came up with this so he has first shot - i kind of forgot during our summit discussions 18:41:50 yep 18:41:59 Then I shall check the implementation of this in VPN ... 18:42:22 bharathm: surely - i had discussed this with pc_m a while back and we can pull in some context 18:42:28 well, there is also the QoS plugin mechanism we were looking at 18:42:34 Cool.. 18:43:12 * pc_m ears perked 18:43:34 nice :-) 18:43:38 xgerman: sorry is this regarding the classifier stuff ? 18:43:55 sorry, I forgot about daylight savings 18:44:08 sc68cal: hi 18:44:11 my calendar didn't shift the start time 18:44:13 nope, classifier is a different beast 18:44:25 xgerman: so which thing on QoS ? 18:44:44 how to get ourselves notified of neutron router changes 18:45:11 but I didn’t check that in detail - sc68cal might have more insight 18:45:14 xgerman: ok i was not aware of that, but vpn will work 18:45:28 *vpn model will work as an example 18:45:32 yeah, QoS solves the same problems we have ;-) 18:46:09 xgerman: ok good 18:47:37 I think QoS works on L2… so might not be as applicable for an L3 firewall 18:48:07 xgerman: yes we basically plug in to L3 agent notifications 18:48:42 #action bharathm and SridarK to sync and possibly bug pc_m if needed 18:49:00 yeah, let’s do that before we get a better handle what QoS does 18:49:07 ok 18:49:19 wasn't following the discussion, but if you need notifications of events from neutron, it's easy. 18:49:26 I can provide info as needed. 18:50:25 pc_m: thanks 18:50:48 pc_m: Thats great.. thanks :-) 18:50:54 Any other bugs we need to discuss ? 18:50:55 someone should put something together in devref 18:51:17 for registering for notifications 18:51:36 +1 18:52:06 agreed, pc_m: do u know if something exists ? 18:53:09 ok lets move on 18:53:26 #topic Summit Updates 18:53:38 For notification yes there is. 18:53:54 We can change offline or in neutron channel, if you want. 18:54:01 pc_m: +1 18:54:15 * pc_m juggling phone call and IRC :) 18:54:22 sc68cal: would u like to drive ? 18:54:45 First everyone needs to watch our talk ;-) 18:54:50 I'm currently on mobile, I thought I had an hour 18:54:53 :-) 18:55:15 sc68cal: ok no worries, will update u can chime in 18:55:27 sounds good, sorry all 18:55:55 #link https://www.openstack.org/summit/tokyo-2015/videos/presentation/openstack-neutron-fwaas-roadmap 18:56:06 Overall Summit Discussions quite fruitful, also met with potential new contributers which is always good 18:56:35 yeah, a lot of people showed interest in refactoring the FWaaS API and functionality a bit 18:56:52 mickeys kept good notes on https://etherpad.openstack.org/p/fwaas-api-evolution-spec 18:56:58 #link https://etherpad.openstack.org/p/fwaas-api-evolution-spec 18:57:14 Priorities: API - we close on overall model and we break out into sub blueprints and start pushing things for Mitaka 18:58:18 yeah, we need to get a spec into the neutron-specs so we can start iterating 18:58:54 Also badveli: on SG - some updates - caution from armax on being careful to keep our focus on API. Also some comments from Ihar on overlap with Classifier model 18:59:25 yeah, sc68cal wrote some classifier code: #link https://github.com/sc68cal/neutron-classifier 18:59:41 badveli: sc68cal replied on thread - so that is something to consider as well in terms of SG overlap with classifier and look for synergy 19:00:09 +1 19:00:55 SG - i meant Service Groups just to be clear 19:01:07 yep :-) 19:01:27 we also had some good chat with the SFC people. That will be important for our future... 19:01:53 and we are completely aligned on moving from router to the Neutron ports 19:01:59 that was a high level view, xgerman: , sc68cal: pls add 19:02:02 xgerman: +1 19:02:24 good summary 19:02:52 we also discussed some grouping mechanisms (address groups, Port Groups) 19:03:25 also here is the etherapd from the design session #link: https://etherpad.openstack.org/p/mitaka-neutron-next-adv-services 19:04:23 mostly with the new port focus we will leave DVR as is and provide east-west once we can attach to vm ports 19:05:10 there was some talk on the logging API but the first talk was in japanese so I skipped 19:05:23 Some of the DVR folks were also happy to hear that :-) 19:05:44 no kidding :-) 19:05:52 indeed 19:09:05 ok, if there are no questions let’s move on 19:09:23 #topic Open Discussion 19:09:48 Next week we will probab have all folks back 19:09:56 yeah, I hope so 19:10:36 probably others got bit by daylight savings and their calendars not being set to UTC 19:10:45 Google really makes it difficult 19:10:51 sc68cal: i also think so 19:12:49 I think we start thinking abt specs and build off the things we have 19:13:11 +1 19:13:15 +1 19:13:18 +1 19:13:29 +1 19:13:30 I think once we have that in the specs directory it will be easier to iterate 19:13:43 agreed 19:14:55 ok if there is nothing else we can end early and give some time back 19:15:16 +! 19:15:18 +1 19:15:43 #endmeeting