14:04:10 #startmeeting neutron_l3 14:04:11 Meeting started Wed Mar 13 14:04:10 2019 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:04:12 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:04:16 The meeting name has been set to 'neutron_l3' 14:04:22 hi there! 14:04:23 this daylight savings is messing me up 14:04:32 Hey! 14:06:22 yes, in betwween the change of time and being in California, I'm also messed up 14:06:54 #chair haleyb 14:06:55 Current chairs: haleyb mlavalle 14:07:32 #chair liuyulong_ 14:07:33 Current chairs: haleyb liuyulong_ mlavalle 14:07:44 haleyb: I have to leave 15 minutes before the hour. In case you need to continue the meeting ^^^^ 14:07:54 ack 14:08:10 hi 14:08:15 hi 14:08:30 Hi 14:09:19 #topic Bugs 14:09:24 #undo 14:09:25 Removing item from minutes: #topic Bugs 14:09:34 #topic Announcements 14:10:47 We are in the RC-1 period until the end of next week 14:11:29 This is what we are targeting for RC-1: https://launchpad.net/neutron/+milestone/stein-rc1 14:13:30 Also the PTL nomination period ended yesterday 14:13:55 and we are in the voting period: 14:13:59 #link https://governance.openstack.org/election/ 14:15:12 any other annoucements? 14:16:31 ok, let's move on 14:16:42 #topic Bugs 14:17:31 we have a critical bug assigned to slaweq: https://bugs.launchpad.net/neutron/+bug/1818614 14:17:32 Launchpad bug 1818614 in neutron "Various L3HA functional tests fails often" [Critical,In progress] - Assigned to Slawek Kaplonski (slaweq) 14:18:02 I have patch for it https://review.openstack.org/#/c/642295/ 14:18:51 but I wanted to check another approach also and maybe try to modify ExternalProcess monitor class that it would only return that neutron-keepalived-state-change process is active if "ip monitor" would be active also 14:19:31 but if You can take a look at current approach and You think that it's enough, that it's fine for me too :) 14:20:13 slaweq: ahh ok, we'll take a look 14:20:26 thx mlavalle 14:20:28 haleyb, liuyulong_: ^^^^ 14:20:47 i never knew you could pass options via AGENT-$option 14:21:07 it's like that because option is defined to be in AGENT group 14:21:51 Yes, slaweq and I have talked about this, we are still working on this. 14:22:24 is there another change checking ip monitor? i hadn't seen it 14:22:26 liuyulong_: I think You are talking about https://bugs.launchpad.net/neutron/+bug/1819160 now 14:22:27 Launchpad bug 1819160 in neutron "Functional tests for dvr ha routers are broken" [High,In progress] - Assigned to Slawek Kaplonski (slaweq) 14:22:37 haleyb: no, there isn't yet 14:22:50 I want to try it locally and then push patch 14:23:12 but for now I don't know if it will be easy to do and really will fix the problem 14:23:38 so I just want to explore such another approach to fix this bug 14:23:39 the other patch is really experimental, for now, right? 14:23:51 right mlavalle 14:23:52 i think what's there should fix it, will review again 14:24:09 haleyb: yes, this one is for sure fixing the issue 14:24:34 You can even check in journal.log of functional tests that it happens couple of times that this helped :) 14:26:28 ok slaweq we will review the current patch 14:26:33 Thanks! 14:26:36 mlavalle: thx 14:27:37 slaweq, sorry, my bad memory... they have similar tittle... 14:27:49 Next one is https://bugs.launchpad.net/neutron/+bug/1795870 14:27:50 Launchpad bug 1795870 in neutron "Trunk scenario test test_trunk_subport_lifecycle fails from time to time" [High,In progress] - Assigned to Miguel Lavalle (minsel) 14:29:03 after deep exploration on how to avoid kill filters for python interpreter, we decided it is not possible without change to oslo.rootwrap 14:29:51 so for the time being we are going with https://review.openstack.org/#/c/636710/ 14:30:03 slaweq: I just addressed your comment 14:30:31 thx mlavalle 14:30:54 so hopefully it is good to go today 14:32:08 next one is https://bugs.launchpad.net/neutron/+bug/1717302 14:32:09 Launchpad bug 1717302 in neutron "Tempest floatingip scenario tests failing on DVR Multinode setup with HA" [High,In progress] - Assigned to Slawek Kaplonski (slaweq) 14:33:04 slaweq: this is fixed isn't it? 14:33:15 my patch is merged 14:33:18 so it should be better 14:33:40 but I'm not sure if there is no other issues which cause similar problems 14:33:50 I will mark it fixed then 14:34:20 ok, slaweq? 14:34:31 ok 14:34:42 if there will be other issues, we can open new bug always :) 14:35:54 yeap 14:35:59 thanks! 14:36:05 next one is https://bugs.launchpad.net/neutron/+bug/1789434 14:36:06 Launchpad bug 1789434 in neutron "neutron_tempest_plugin.scenario.test_migration.NetworkMigrationFromHA failing 100% times" [High,Confirmed] - Assigned to Manjeet Singh Bhatia (manjeet-s-bhatia) 14:36:23 IIRC this one should be fixed by Your patch mentioned above 14:36:25 I'm assigning it back to me 14:36:48 I was recently checking it locally and with changes in rootwrap filters it was working fine for me 14:36:48 for the time being, assuming that the filters patch fixes it 14:37:11 I just want to confirm 14:37:18 so You can send follow up patch to unmark those tests as unstable 14:37:36 slaweq: yes, I'll do that 14:37:39 thx mlavalle 14:37:47 mlavalle: should you update the commit message to include that bug? 14:38:17 haleyb: good point. I'll do it in a bit 14:38:43 haleyb++ 14:39:04 Next one is https://bugs.launchpad.net/neutron/+bug/1818334 14:39:06 Launchpad bug 1818334 in neutron "Functional test test_concurrent_create_port_forwarding_update_port is failing" [High,Confirmed] - Assigned to LIU Yulong (dragon889) 14:39:15 hi 14:39:25 Let me output some updates about this bug. 14:39:51 I can not reproduce this locally in 100+ times run. 14:40:24 But I can indeed meet the issue when run two concurrent functional tests. 14:40:58 It may be related to the thread pool which is simulating concurrent updating port ip and creating fip port forwarding for this port. 14:41:01 3 times in last 15days. 14:41:58 Maybe we can lower the priority for now. 14:42:20 will you continue working on it? 14:42:48 Yes 14:42:57 ok, lets lower to medium 14:44:04 last one in my list is https://bugs.launchpad.net/neutron/+bug/1819160 14:44:05 Launchpad bug 1819160 in neutron "Functional tests for dvr ha routers are broken" [High,In progress] - Assigned to Slawek Kaplonski (slaweq) 14:44:15 this is also fixed, isn't it? 14:44:26 No, just marked as unstable 14:44:29 no, this one is what liuyulong_ mentioned before 14:44:46 ah ok, so let's keep it alive 14:44:56 those are all the bugs I had for today 14:44:58 basically IMO those dvr ha functional tests are not testing what they should test 14:45:15 ok, understood 14:45:26 because routers created there aren't "one router on 2 nodes" but "2 different routes on 2 nodes" 14:45:37 and liuyulong_ is checking how to fix it :) 14:45:54 thanks slaweq, liuyulong_ ! 14:45:55 Here are my try: https://review.openstack.org/#/c/642220/, not the root cause for now IMO 14:46:07 I have to leave now, please continue the meeting 14:46:15 I don't have any more bugs 14:46:20 o/ 14:46:38 o/ 14:49:13 hello? 14:49:20 Hey 14:49:37 Does anyone have anymore bugs to discuss? 14:50:20 oh, mlavalle must have dropped 14:50:40 He had to leave at XX:45 14:51:11 haleyb, you or liuyulong_ should finish the meeting 14:51:22 hi 14:51:37 14:54:29 14:51:41 doh, 14:51:47 #topic On demand agenda 14:52:14 We have a question about ovs dvr 14:52:43 sure 14:53:36 We're working on converting it to backend rather than another agent-mode, I was just wondering is there any documentation specificly about what each mode supports individually, so are there any that shouldn't support Floating IP for example? 14:54:53 No floating IP for L3 ? 14:55:15 dvr_no_external has no external network, so no floating IP in that one mode 14:55:28 liuyulong_, just using it as an example 14:57:33 haleyb: kk, we'll keep that one in mind. 14:58:52 any other topics? 2 minutes 14:59:19 none from me 14:59:42 liuyulong_ 14:59:48 ok, thanks for attending everyone 14:59:53 bye 14:59:55 #endmeeting