00:00:16 #startmeeting Networking FWaaS 00:00:17 Meeting started Thu Feb 4 00:00:16 2016 UTC and is due to finish in 60 minutes. The chair is SridarK. Information about MeetBot at http://wiki.debian.org/MeetBot. 00:00:18 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 00:00:20 The meeting name has been set to 'networking_fwaas' 00:00:27 #chair xgerman SridarK 00:00:28 Current chairs: SridarK xgerman 00:00:55 #topic announcements 00:00:56 #topic FWaaS v2 patches 00:01:27 we can do a quick run thru of where we stand and things we may want to discuss 00:01:39 o/ 00:01:41 ok, sounds good 00:02:11 i am continuing with the plugin and adding some methods needed in the db patch and testing the plugin - db portion on devstack 00:02:38 I did some work on versioned objects + sat with an expert 00:02:50 xgerman: great i saw ur patch 00:03:04 yeah, slow progress 00:03:09 hopefully now we can work together to understand what is needed 00:03:15 +1 00:03:20 on the agent interface 00:03:25 yep 00:04:37 mickeys: how does the driver piece look ? Anything u want to bring up for discussion ? 00:05:18 also jwarendt posted a CLI patch 00:05:26 We have been distracted by other things. Made some progress understanding the existing security groups code flow, but have not worked out a design. 00:05:56 mickeys: ok understand 00:06:01 #link https://review.openstack.org/#/c/275473/1 <-CLI 00:06:50 jwarendt: thx 00:06:51 One issue we found is that the new iptables_manager that does a diff puts the rules right after the chains. We cannot have both security groups and fwaas apply that logic, or they would swap which set of rules comes first and cause a large diff every time. We have to work something out to order the rules differently. 00:07:23 oh yes 00:08:27 maybe some sort of feature specific thing needs to be done 00:09:04 Or deterministic ordering of chains, for example sorting? Still need to look into the details to understand whether there is an issue within chains. 00:09:36 ok 00:09:55 yeah, I am still a bit confused since I though we would have separate chains 00:10:15 Roey also had a UT patch out 00:10:29 yeah, saw that 00:10:36 great work 00:11:23 xgerman: may be next we can try to pull the plugin and agent pieces together to get some of the communication going 00:11:33 we can target next week 00:12:00 ok, let’s do that. Wish should be able to dedicate 25% to FwaaS soon, too 00:12:06 Aish 00:12:25 yes sigh i think we all have to juggle other things as well 00:12:35 that will be great 00:12:59 yep, things move slow at big places... 00:13:25 xgerman: i will ping u offline on the versioned objects to get a handle on that 00:13:41 ok, sounds good 00:13:57 i think that about does a quick run thru on the patches 00:14:05 yep 00:14:23 xgerman: did u want to bring some other things up 00:14:27 yep 00:14:35 #topic Open Discussion 00:14:50 There is a Neutron Midcycle in Rochester, soo 00:14:52 n 00:15:36 if you look there is an etherpad 00:16:14 big thing is “evolving the stadium” - look at that patch: https://review.openstack.org/#/c/275888/5 00:16:44 lastly, please do more reviews https://review.openstack.org/#/q/openstack/neutron-fwaas+status:open,n,z 00:16:47 #link https://review.openstack.org/#/q/openstack/neutron-fwaas+status:open,n,z 00:16:58 some nice tempest fixes, etc. sitting there :-) 00:17:16 that’s all from my soap box... 00:18:41 xgerman: one other thing - i need to follow up with Brocade folks on the observer hierarchy 00:18:50 oh, yes 00:19:08 they have taken this on (the other bharath) 00:19:26 i will find out how that is moving along 00:19:51 ask them to post some WIP so we can all comment ... 00:20:04 xgerman: yes exactly will do so 00:20:27 I did not have anything else to discuss 00:20:47 +1 00:20:49 others ? 00:20:59 I am reading about versioned objects too. xgerman SridarK Let me know when you discuss about that. 00:21:06 will do 00:21:22 Aish: yes will coordinate on that with both of u 00:21:29 ok, thanks. 00:22:12 ok if nothing else, xgerman: shall we close out the mtg and give some time back ? 00:22:20 yes 00:23:02 #endmeeting