15:00:13 <mlavalle> #startmeeting neutron_l3
15:00:18 <openstack> Meeting started Thu Aug 30 15:00:13 2018 UTC and is due to finish in 60 minutes.  The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:21 <openstack> The meeting name has been set to 'neutron_l3'
15:00:22 <haleyb> hi
15:00:24 <manjeets> hi
15:00:28 <mlavalle> hey haleyb
15:00:32 <mlavalle> hi manjeets
15:01:01 <munimeha1> hi
15:01:19 <mlavalle> welcome munimeha1
15:01:45 <mlavalle> munimeha1: do you have a topic you want to discuss during the meeting?
15:02:48 <mlavalle> #topic Announcements
15:02:52 <njohnston> o/
15:03:16 <mlavalle> Yesterday we released Rocky: https://review.openstack.org/#/c/597529/
15:04:15 <mlavalle> so for all practical purposes, we can think we are in the Stein cycle
15:04:25 <mlavalle> welcome to Stein!
15:04:48 <mlavalle> The PTG is a little more than a week away
15:05:26 <mlavalle> I am consolidating all the proposed topics in a coherent agenda: https://etherpad.openstack.org/p/neutron-stein-ptg
15:05:52 <mlavalle> manjeets: are you joininh the team dinner on Thursday?
15:06:17 <manjeets> mlavalle, yes
15:06:36 <mlavalle> manjeets: so add a Yes at the end of the line where your name is in the etherpad
15:07:08 <mlavalle> any other announcements from the team?
15:07:14 <manjeets> done
15:07:18 <mlavalle> :-)
15:07:45 <mlavalle> ok, let's mlove on
15:07:56 <mlavalle> yeah, let's love on
15:07:58 <mlavalle> LOL
15:08:00 <manjeets> lol
15:08:04 <mlavalle> #topic Bugs
15:08:46 <mlavalle> First I have for today is https://bugs.launchpad.net/neutron/+bug/1788006
15:08:46 <openstack> Launchpad bug 1788006 in neutron "neutron_tempest_plugin DNS integration tests fail with "Server [UUID] failed to reach ACTIVE status and task state "None" within the required time ([INTEGER] s). Current status: BUILD. Current task state: spawning."" [High,Confirmed] - Assigned to Miguel Lavalle (minsel)
15:09:26 <mlavalle> I investigated this one and couldn't find anything wrong on the Neutron side
15:09:52 <mlavalle> I traced it back to Nova. The compute receives the vif plugged event correctly from Neutron
15:10:27 <mlavalle> but that doesn't propagate to the Nova API, so the instance stays stuck in state "spawning"
15:10:37 <mlavalle> I opened a bug in the NOva side
15:11:12 <mlavalle> however, during the CI a meeting slaweq pointed out that it was weird that this is happening in DNS integration tests only
15:11:32 <mlavalle> so I am digging in other cases
15:12:02 <mlavalle> Next one is https://bugs.launchpad.net/neutron/+bug/1789434
15:12:02 <openstack> Launchpad bug 1789434 in neutron "neutron_tempest_plugin.scenario.test_migration.NetworkMigrationFromHA failing 100% times" [High,Confirmed] - Assigned to Miguel Lavalle (minsel)
15:12:22 <mlavalle> I had fight hard with haleyb to get ownership of this bug
15:12:38 <mlavalle> I finally prevailed ;-)
15:12:50 <njohnston> LOL
15:13:34 <mlavalle> as shared yesterday with manjeets and hongbin yesterday, I know have a vagrabt script where I can build dvr-ha environments in 50 minutes
15:14:00 <mlavalle> so I am going to attemopt to reporduce this issue locally in my environment today
15:15:10 <mlavalle> Finally, I am looking at https://bugs.launchpad.net/neutron/+bug/1787919
15:15:10 <openstack> Launchpad bug 1787919 in neutron "Upgrade router to L3 HA broke IPv6" [High,Confirmed]
15:15:26 <mlavalle> I just assigned it to myself
15:15:44 <mlavalle> and will also reproduce locally in my environment
15:15:49 <haleyb> that looks interesting
15:16:23 <mlavalle> in this case there are notes from hongbin on how to reproduce locally
15:16:31 <mlavalle> \so I'll follow the recipe
15:17:05 <haleyb> like baking a cake :)
15:18:03 <mlavalle> we have this critical one https://bugs.launchpad.net/neutron/+bug/1788185
15:18:03 <openstack> Launchpad bug 1788185 in neutron "[Stable/Queens] Functional tests neutron.tests.functional.agent.l3.test_ha_router failing 100% times " [Critical,Confirmed] - Assigned to Miguel Lavalle (minsel)
15:18:15 <mlavalle> really the assignee here is slaweq
15:18:32 <mlavalle> he has been working on the issue for the past few days
15:20:10 <mlavalle> fixes were fixed for Ocata and Queens. I will follow up at the end of this meeting with slaweq to see where we are
15:20:10 <haleyb> i think there's still an ocata backport?  but maybe it's still broken after reading the last comments?
15:20:20 <mlavalle> yeah
15:20:27 <mlavalle> that last comment gave me pause
15:20:41 <mlavalle> I'll check with slaweq to see what he thinks
15:21:11 <haleyb> i thought i saw a strange HA func test failure today, maybe that's related
15:21:40 <mlavalle> Those are the bugs I have at this point
15:21:59 <haleyb> there is another
15:21:59 <mlavalle> are there any other bugs we should discuss today?
15:22:08 <haleyb> https://bugs.launchpad.net/neutron/+bug/1789403
15:22:08 <openstack> Launchpad bug 1789403 in neutron "ip6tables rules for PD subnets not fully recreated at l3-agent restart" [High,Confirmed] - Assigned to Brian Haley (brian-haley)
15:22:40 <haleyb> posted https://review.openstack.org/#/c/597710/ - just got an update that it seems to fix the issue
15:22:56 <haleyb> that was the one that had the odd test failure
15:23:05 <mlavalle> ahhh, ok
15:23:12 <haleyb> http://logs.openstack.org/10/597710/1/check/neutron-functional/9d0610b/logs/testr_results.html.gz
15:23:20 <slaweq> hi, sorry for being late :)
15:23:36 <haleyb> mlavalle: so i didn't know if it was my code, or keepalived
15:24:18 <mlavalle> I can run that test locally. I'll let you know if it fails
15:24:37 <haleyb> mlavalle: thanks
15:25:14 <mlavalle> hi slaweq. I was worried that you migh have been eaten by the python ;-)
15:25:26 <slaweq> mlavalle: LOL, no
15:25:33 <slaweq> I eat python :D
15:25:41 <mlavalle> you do
15:26:01 <mlavalle> we were wondering about the status of https://bugs.launchpad.net/neutron/+bug/1788185
15:26:01 <openstack> Launchpad bug 1788185 in neutron "[Stable/Queens] Functional tests neutron.tests.functional.agent.l3.test_ha_router failing 100% times " [Critical,Confirmed] - Assigned to Miguel Lavalle (minsel)
15:26:31 <slaweq> workaround patches for Queens and Ocata are merged
15:26:38 <slaweq> there is still not merged patch for Pike: https://review.openstack.org/#/c/596559/
15:27:07 <slaweq> but yesterday frickler was playing a bit with it and he had some comments about keepalived version
15:27:29 <mlavalle> exactly, that is really my question
15:27:31 <slaweq> for him it looks like old keepalived was behaving wrong and we tested this wrong behaviour as good one
15:27:41 <slaweq> I will have to check that once again
15:28:03 <slaweq> and also strange thing is that on newer keepalived (1.3.9) test is fine
15:28:03 <mlavalle> ok
15:29:23 <mlavalle> is it ok if I assign the bug to you?
15:29:28 <slaweq> sure
15:29:51 <mlavalle> done
15:29:55 <mlavalle> thanks!
15:30:32 <mlavalle> any other bugs we should discuss today
15:30:39 <haleyb> https://bugs.launchpad.net/neutron/+bug/1786272
15:30:39 <openstack> Launchpad bug 1786272 in neutron "Connection between two virtual routers does not work with DVR" [Medium,In progress] - Assigned to Brian Haley (brian-haley)
15:30:59 <haleyb> it's only assigned to me since i updated the review, but i see new comments
15:31:47 <haleyb> slaweq made a change to schedule these shared-subnet DVR routers to allow traffic, but it requires more work
15:32:09 <slaweq> I'm trying to do scenario test to cover such case today
15:32:36 <mlavalle> so, should we change the assignee to slaweq?
15:32:37 <slaweq> with such scenario test would be easier to be sure that patch works
15:32:45 <haleyb> slaweq: looks like it's going to need an OF change ?
15:33:10 <slaweq> locally I didn't do any changes in OF
15:33:16 <slaweq> and tbh I don't think so
15:33:42 <slaweq> basically if there is vm from "second" subnet on host, qrouter- will be created now, and all will work then
15:33:48 <haleyb> oh, i thought from swami's comment it was needed
15:34:03 <haleyb> maybe that was without your change...
15:34:26 <slaweq> so if I will change logic of which qrouter namespaces should be created on node, it should do everything to be working
15:34:46 <slaweq> but let's do test and check then on CI :)
15:34:55 * haleyb was just confused with comments
15:35:27 <mlavalle> cool
15:35:32 <mlavalle> any other bugs?
15:35:41 <haleyb> we can re-assign, or just wait for slaweq to update and it will change to him :)
15:35:50 <mlavalle> ok
15:36:48 <mlavalle> ok, let's move to open agenda then
15:36:54 <mlavalle> #topic Open Agenda
15:37:59 <mlavalle> In building environment to test dvr-ha related bugs, I followed this recipe: https://docs.openstack.org/neutron/latest/admin/config-dvr-ha-snat.html
15:38:32 <mlavalle> and it referes in several place to br-ex
15:38:51 <mlavalle> that is obsolete, isn't it?
15:39:25 <davidsha> Not yet I think
15:39:58 <davidsha> The patch still hasn't merged?
15:40:29 <davidsha> https://review.openstack.org/#/c/567369/
15:41:40 <mlavalle> ok, but it is in it's way to be obsolete
15:41:46 <mlavalle> Thanks!
15:42:03 <davidsha> np!
15:42:21 <mlavalle> and that patch by the way fixes the doc
15:42:35 <mlavalle> so we are good
15:42:52 <mlavalle> any other topics we should discuss today?
15:43:45 <mlavalle> ok, thanks for attending
15:43:49 <mlavalle> #endmeeting