15:00:28 #startmeeting neutron_l3 15:00:29 Meeting started Thu Dec 15 15:00:28 2016 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:30 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:32 The meeting name has been set to 'neutron_l3' 15:00:48 o/ 15:01:05 hi 15:01:25 hey guys 15:01:46 As usual, agenda is here: 15:01:55 #link https://etherpad.openstack.org/p/neutron-l3-subteam 15:02:09 #topic Annoucements 15:02:47 Well, we made it to Ocata-2 15:02:54 #link https://releases.openstack.org/ocata/schedule.html 15:03:09 This cycle is definitely running faster than previous ones 15:04:01 And a lot of folks will be taking time off between now and O-3 15:04:41 yeap. are you or heleyb among them? 15:05:14 this is my last day until next year :-o 15:05:45 haleyb: but that means you coming back in early January, right? 15:05:49 of course i'll be watching the review queue here and there 15:06:12 mlavalle: I'm away from Dec 23 to Jan 9 15:06:14 mlavalle: yes, i'll be back :) 15:06:29 mlavalle: You? 15:06:42 I'll be here through the Holidays 15:07:10 I'll take a few days off 15:07:44 mlavalle: you'll be un-packing, that's not a day off 15:07:44 I also had a chat last week with tidwellr. I pinged him to know if he wanted to continue co-chairing this meeting with me 15:08:38 He has a few conflicts between now and the end of January. But after that, he want to continue being coc-chair 15:08:51 So you will see us rotating the role 15:09:18 Any other annoucements? 15:10:21 ok 15:10:25 #topic Bugs 15:11:30 First one is https://bugs.launchpad.net/neutron/+bug/1610483 15:11:31 Launchpad bug 1610483 in neutron "Pluggable IPAM rollback mechanism is not robust" [High,In progress] - Assigned to Aliaksandr Dziarkach (aliaksandr-dziarkach) 15:11:53 There is a fix proposed: 15:11:59 #link https://review.openstack.org/#/c/390594/ 15:12:21 I have been slowly moving through it and made a few comments, that I haven't published yet 15:12:31 I'll continue going through it 15:12:49 I encourage other team members to do the same 15:12:58 Any other comments? 15:13:57 Ok, next one: https://bugs.launchpad.net/neutron/+bug/1570122 15:13:58 Launchpad bug 1570122 in neutron "ipv6 prefix delegated subnets are not accessable external of the router they are attached." [High,In progress] - Assigned to John Davidge (john-davidge) 15:14:45 Any other feedback from the reporter? 15:15:03 Latest version of the WIP patch has positive feedback from promethianfire (the bug reporter). Next I'll add some tests and it should be good to go 15:15:11 #link https://review.openstack.org/#/c/407025/ 15:15:19 yeah, I am just reading it 15:15:27 Great! 15:16:14 Thanks! 15:17:08 Next one is https://bugs.launchpad.net/neutron/+bug/1627424 15:17:09 Launchpad bug 1627424 in neutron "FlushError on IPAllocation" [High,Confirmed] - Assigned to Miguel Lavalle (minsel) 15:17:33 First, thanks to haleyb for the notes you left there. Very useful 15:17:44 my eyes were spinning 15:17:59 I haven't had much time this week to follow up because I am the bugs deputy this week 15:18:49 I'll continue gojng after it next week 15:18:58 mlavalle: we might want to ping kevinbenton to see if he can help, he's good at the DB stuff 15:19:20 haleyb: ok, will do that 15:20:04 Next one is https://bugs.launchpad.net/neutron/+bug/1627480 15:20:05 Launchpad bug 1627480 in neutron "create_port can succeed without returning fixed_ips on all requested subnets" [High,Confirmed] 15:20:39 I talked to tidwellr about this one. He is interested on it, so he agreed to help with it 15:21:07 any other comments? 15:22:22 Next one is https://bugs.launchpad.net/neutron/+bug/1636963 15:22:23 Launchpad bug 1636963 in neutron "Service Subnets - DHCP port creation fails" [High,In progress] - Assigned to John Davidge (john-davidge) 15:22:43 Updated the patch to only work for dhcp-enabled subnets based on our conversation in the last meeting, and it now has a +2 from haleyb. 15:22:47 #link https://review.openstack.org/#/c/392183/ 15:23:00 Thanks haleyb and baoli for the reviews 15:23:13 john-davidge: you are welcome 15:23:32 lgtm, thanks john-davidge 15:23:46 yeah, thanks john-davidge 15:23:48 john-davidge: good job 15:24:16 Thanks all :) 15:24:45 Any other bugs I may have missed? 15:25:38 ok, in that case 15:25:53 #topic Prefix Delegation 15:25:59 mlavalle: thank you for looking at https://review.openstack.org/#/c/408306/. The remaining CI failures don't seem to be related to the patch, 15:26:41 baoli: yeah, I spent a good amount time looking (and learning) at it 15:26:59 very hard to argue agains it. It looks pretty good 15:27:29 I'll wait for the result of that re-check and will take another look 15:27:36 the Tracebacks in gate-neutron-dsvm-functional-ubuntu-xenial seems to be happening in other patches as well. So it might be a bug that need to be looked at. 15:28:51 http://logs.openstack.org/06/408306/3/check/gate-neutron-dsvm-functional-ubuntu-xenial/8afbeaf/logs/dsvm-functional-logs/neutron.tests.functional.agent.l3.test_ha_router.LinuxBridgeL3HATestCase.test_ha_router_lifecycle.txt.gz 15:29:10 there is an error: cannot open network namespace. 15:29:36 While looking at it, I wondered if we need a tempest scenario test 15:29:56 do you think that is doable, baoli? 15:30:34 mlavalle: i'm not quite familiar with the scenario test yet. I wonder if it can be done separately? 15:30:59 baoli: yes, I don't mean for you to include a scenario test in this patchset 15:31:35 my point is that even with our unit tests, the thing didn't really work and we didn't realize it until you tested it 15:32:07 a scenario test in tempest allows you to test at the user level 15:32:08 mlavalle: I agree that we should have such a test. 15:32:42 ok, let's think about that 15:33:18 anything else you want to share with the team baoli? 15:33:23 mlavalle: yes 15:34:21 mlavalle: regarding the PD HA feature, I've had the code to make it work. Checking the RFE page, it seems that I shouldn't open a RFE bug anymore. 15:35:19 So I wonder if I just open a regular bug. What I'm missing now is the unit tests that I need to develop. And again, it seems that some sort of scenario tests are also in order 15:36:36 baoli: i would agree there is a hole in the PD tests 15:36:38 baoli: The bug report wouldn't need to be very detailed, but I believe it would be appropriate to tage it with RFE 15:36:48 baoli: Not something that will need a spec I imagine 15:37:23 haleyb: baoli: Yup, some PD testing definitely needs attention 15:37:43 haleyb: yeap, things like this is hard to catch in unit tests 15:38:44 john-davidge, if I read the RFE page correctly, if there is immediate code available, then it shouldn't be tagged with RFE 15:39:04 baoli: I have experience with tempest scenario. we can discuss it 15:39:43 baoli: Ah ok, wasn't aware of that condition. Should be fine then I guess 15:39:54 mlavalle: sure. I can also see if there is someone in my team that can help. 15:40:04 ok 15:40:43 so for PD HA, shall I just open a regular bug, and post a WIP patch (without unit tests yet) 15:40:51 baoli: anything else? 15:41:08 mlavalle: ^^ 15:41:28 baoli: yeah go ahead and do that, please 15:41:42 mlavalle: ok, cool. 15:41:56 malvalle: I'm all set. 15:42:09 baoli: thanks for working on this :-) 15:42:22 mlavalle: sure thing 15:42:33 #topic Routed Networks 15:42:55 In this front I have continued working on the integration with the Nova placement API 15:43:05 I am almost done with the unit tests 15:43:34 and after that I only need to move my placement client to Neutron 15:43:48 I expect to be done by year's end 15:44:16 Any questions? 15:44:54 ok 15:45:06 I don't think tidwellr is on line 15:45:17 so we will skip BGP dynamic routing 15:45:25 #topic Open agenda 15:45:51 Does anyone want to share something with the team? 15:46:18 mlavalle: i am in the process of updating the subnet service type doc for DVR, doing final RST changes 15:46:43 so i didn't completely forget :) 15:46:47 haleyb: thanks.... I didn't want to be annoying with that. I know you won't forget 15:46:48 haleyb: Cool! Thanks 15:47:27 anything else? 15:47:58 ok haleyb enjoy your time off 15:48:36 Thanks everybody for attending 15:48:42 #endmeeting