18:00:37 #startmeeting networking_ovn 18:00:38 Meeting started Thu Sep 29 18:00:37 2016 UTC and is due to finish in 60 minutes. The chair is rtheis. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:40 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:42 The meeting name has been set to 'networking_ovn' 18:00:50 hi 18:00:59 #topic: Announcements 18:01:03 hi russellb 18:01:12 Open vSwitch 2.6.0 released 18:01:19 #link http://openvswitch.org/pipermail/discuss/2016-September/022786.html 18:01:33 networking-ovn Newton RC2 released 18:01:40 #link http://lists.openstack.org/pipermail/openstack-dev/2016-September/104673.html and https://review.openstack.org/#/c/374820/ 18:01:55 \o/ 18:02:01 I think we are in good shape for newton 18:02:14 yep, thanks for staying on top of things rtheis 18:02:19 yw 18:02:34 I don't have any other announcements 18:03:00 #topic: Bugs 18:03:12 #link https://bugs.launchpad.net/networking-ovn/+bug/1627810 18:03:13 Launchpad bug 1627810 in networking-ovn "gate-grenade-dsvm-networking-ovn-nv [Fail] Couldn't ping server " [Undecided,In progress] - Assigned to Richard Theis (rtheis) 18:03:25 grenade is broken 18:03:49 Babu and I have been looking at it but nothing solid to go on for a fix yet 18:04:22 i suggested to him that if after some time (a day or whatever) it still wasn't too clear how to fix it, we might just want to forget about it and wait until we can drop the l3 agent in that job 18:04:26 and then worry about getting it working again 18:04:57 sounds good 18:05:40 I'll take another look if I have time ... one more idea to try out 18:05:51 then maybe defer it 18:06:06 #link https://bugs.launchpad.net/networking-ovn/+bug/1621627 18:06:08 Launchpad bug 1621627 in networking-ovn "Create multi-node CI job for networking-ovn" [High,In progress] - Assigned to Richard Theis (rtheis) 18:06:20 ok 18:06:20 there is also the multi-node CI job 18:06:24 how's that going? 18:06:26 very exciting 18:06:35 it works 18:06:38 https://review.openstack.org/#/c/376968/ 18:07:15 but we aren't doing the heavy lifting yet with floating IPs until l3 support merges 18:07:43 but we can handle that as a change later 18:08:18 once https://review.openstack.org/#/c/376968/ goes in, https://review.openstack.org/#/c/378032/ will move the job to non-voting 18:09:02 And final bug that I have ... 18:09:15 no launchpad bug yet ... 18:09:27 the functional test is broken with ovs 2.6.0 18:09:35 and thus our gate isn't happy 18:09:53 I've triaged it down to ovs transaction mutate support 18:10:21 I think this may be a bug in ovs 18:10:43 * azbiswas ears perk up 18:11:04 running 2 commands that both mutate a column works, but ... 18:11:28 using the idl to get the column data in first command return full set 18:11:30 rtheis: Do you have no details on this? 18:11:30 more details 18:11:47 using the idl to get the column data in second commend returns an empty list 18:12:32 end result is that if n-ovn does any comparisons with the column data, it may be inaccurate since the mutate blasted it for subequent commands under the same transaction 18:13:01 unfortunately, there are a couple hits in n-ovn 18:13:15 We can try running it with the master branch of OVS and see if that fails as well 18:13:17 I have an idea how to fix on n-ovn side and will open a bug 18:14:08 hm, i wonder if we want to make the CI jobs for stable/newton use branch-2.6 of ovs vs master ... 18:14:15 azbiswas: has there been a lot of changes between master and ovs 2.6.0 just released on pypi 18:14:21 That sounds like a bug in idl. 18:14:45 Not in the idl part 18:15:48 russellb: that's a good idea 18:16:39 that's all I have on this bug for now 18:16:54 I'll finish my investigation and open a bug report with the details 18:17:18 any other bugs? 18:17:49 #topic Reviews 18:18:08 how about reviews? 18:18:25 yes! 18:18:48 we still have similar list of new features and unit test coverage improvements in the queue 18:19:01 I'll try doing more reviews once the gate is fixed 18:19:12 was hoping to have more time now that newton rc2 is out 18:19:25 is there a gate fix in the queue? 18:19:31 no 18:19:43 i'm also interested in the status of the SNAT / Floating IP patch 18:19:55 looks like it got rebased yesterday 18:20:09 chandrav ^^^^ 18:20:20 i was proposing that we should try to merge it soon since tests are passing 18:20:29 and address any cleanup or bugs we come across as follow-ups 18:20:58 sounds good to me 18:21:25 chandrav: thanks for sticking through with this one! 18:21:31 The only part remaining from implementation is the sync part 18:21:47 russellb: Also, there is 1 tempest test failing "tempest.scenario.test_network_v6" 18:22:01 you have it disabled though right? 18:22:07 I'll raise 2 bugs for sync code and the tempest failure 18:22:07 filing a bug / following up on that is OK with me 18:22:12 yes, i disabled it 18:23:14 chandrav: once the bugs are opened, please reference them in the code todos 18:23:16 thanks 18:23:38 will do 18:23:59 anything else on reviews ? 18:24:29 #topic Conferences 18:24:43 We have a room scheduled for OpenStack summit 18:24:51 Workroom setup scheduled on Wednesday from 3:05 to 3:45 18:24:56 #link https://docs.google.com/spreadsheets/d/1TQ-RSlbiBBEclkonIbfUP7R1ExZSJylF1uiEKV2G_Cw/pubhtml?gid=1107826458&single=true 18:25:26 add topics to etherpad https://etherpad.openstack.org/p/networking-ovn-ocata-summit 18:25:53 Also there is a roundtable setup open all day Friday if needed 18:26:20 anything else on conferences? 18:26:54 #topic Open Discussion 18:27:08 I don't have any specific topics 18:27:35 If nothing, we can end early 18:28:51 okay, back to the gate fix and reviews ... 18:29:01 thanks everyone 18:29:02 #endmeeting