15:04:27 #startmeeting neutron_northbound 15:04:27 Meeting started Fri Mar 18 15:04:27 2016 UTC. The chair is yamahata. Information about MeetBot at http://ci.openstack.org/meetbot.html. 15:04:27 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:04:27 The meeting name has been set to 'neutron_northbound' 15:04:34 #topic agenda bashing and roll call 15:04:55 #info yamahata 15:05:22 #link https://wiki.opendaylight.org/view/NeutronNorthbound:Meetings agenda page 15:05:36 bug I haven't update the page. the agenda I have is usual one 15:05:53 any other agenda items? 15:07:31 seems nothing else. move on. 15:07:37 #topic Announcements 15:08:12 ODL boron M1 cycle ended. our project applyed for Boron simultenious release 15:08:27 The release plan is at https://wiki.opendaylight.org/view/NeutronNorthbound:Boron_Release_Plan 15:08:36 #link https://wiki.opendaylight.org/view/NeutronNorthbound:Boron_Release_Plan boron release plan(draft) 15:09:11 I put the main theme and priority as draft 15:09:53 code clean up(removing legacy I*Aware interfaces) 15:10:11 revise yang model. We found several bugs in its definitions 15:10:15 neutron HA support 15:10:22 improve port binding 15:10:36 providernet 15:10:43 lastly centralized SNAT support 15:11:16 The priority is mostly driven by the community. 15:11:38 i.e. if someone steps up for some feature, it will get priority. Otherwise low priority. 15:12:12 any comments/thoughts? 15:13:22 the next topic 15:13:28 #topic action items from last meeting 15:13:40 there is no action items from last week. 15:13:52 #topic bugs and patch review 15:14:08 bugs 15:14:24 #link https://bugs.opendaylight.org/show_bug.cgi?id=5478 Null values received on neutron security-group-rule-delete and security-group-delete 15:14:57 He reports two issues. One bug is addressed by already proposed patch. 15:15:08 another issue will be addressed soon. 15:15:25 On openstack side, We have one bug 15:15:47 #link https://bugs.launchpad.net/networking-odl/+bug/1557976 sg events callback: sg group deleted but the related rules are left 15:16:10 The issue is default sg rules aren't created on ODL side 15:16:24 and also on deletion of sg, the related rules aren't deleted. 15:16:31 Yalei is proposing patches. 15:16:44 #link https://review.openstack.org/#/c/294451/ 15:16:45 Cascading delete rules related to SG 15:17:07 #link https://review.openstack.org/#/c/294423/ Security group call back need cascading delete the related rules 15:17:43 any other bugs that need attention? 15:17:50 or patches? 15:19:14 I have also another patch review 15:19:16 https://review.openstack.org/#/c/281693/ 15:19:28 Pass the complete info in sg/rules db into PRECOMMIT_XXX callback 15:19:33 It's also needed for security group support 15:19:51 move on 15:20:01 #topic ML2 ODL driver rewrite 15:20:14 So far I have merged related patches. 15:20:23 The next thing is to test it. 15:21:07 And also full sync patch is WIP. https://review.openstack.org/#/c/275142/ 15:21:27 It would take a while. But we need it for Mitaka for feature parity with v1 driver 15:22:20 Given the Mitaka schedule, probably we should implemented limited version of full sync instead of whole functionality of fullsyn 15:22:32 any other update? 15:22:54 asomya: rcurran ? 15:23:20 sorry, on another window, no updates from me 15:23:28 yamahata: no updates from me 15:23:52 okay next topic 15:23:54 isaku on the full sync - it looks like you detect a full sync is needed by polling ODL to get the complete set of resources 15:24:09 and then compare against what neturon has? 15:24:41 is that the correct understanding of the spec? 15:25:07 Hmm no. It's not efficient and impractical if the number of resource (e..g ports) are large 15:25:22 For cold boot detection, dummy network will be used. 15:25:45 To detect random data breakage in ODL, I suppose hash or something would be needed as we discussed. 15:25:55 yes - I was worried about that 15:26:05 also the database is constantly changing 15:26:30 so you can sync everything but detection is limited to a cold boot case 15:26:38 Yes for Mitaka. 15:27:05 Ok - I need to study that more been busy on other things 15:27:06 healing random data breakage will be attacked for Newtorn and Boron 15:27:16 I do agree this is an important peice 15:27:41 ok thanks 15:28:50 I'll drop the item of release support from the agenda 15:29:01 for planning 15:29:22 #topic Beryllium release preparation and Boron planning 15:29:44 ODL boron planning is being done. and it was discussed at ODL developer design forum. 15:29:53 The outcome is the planning page 15:30:07 #link https://wiki.opendaylight.org/view/NeutronNorthbound:Boron_Release_Plan boron planning 15:31:05 For Newton planning, do we want some page? or etherpad? 15:31:24 openstack summit is the end of April, so we have a while. 15:32:14 There are several requrement from ODL side. 15:32:32 Basically it's for port binding and SNAT support. 15:33:16 Also we'd like to continue v2 driver effort. 15:33:26 any other requests/wishlist? 15:35:12 okay seems nothing. anyway I should prepair page for newton planning 15:35:21 #action yamahata prepare a page for newton planning 15:35:28 #topic open mike 15:35:38 anything to discuss? 15:36:57 seems nothing. 15:37:09 I'll give 20mins to you 15:37:14 thank you everyone 15:37:20 bye 15:37:20 #topic cookies 15:37:25 #endmeeting