14:00:08 <slaweq> #startmeeting networking 14:00:09 <openstack> Meeting started Tue Sep 22 14:00:08 2020 UTC and is due to finish in 60 minutes. The chair is slaweq. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:10 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:12 <openstack> The meeting name has been set to 'networking' 14:00:26 <ralonsoh> hi 14:00:33 <slaweq> hi 14:00:47 <mlavalle> o/ 14:01:30 <lajoskatona> Hi 14:01:50 <bcafarel> o/ 14:02:11 <slaweq> ok, I think we can start 14:02:20 <slaweq> welcome everyone! 14:02:22 <slaweq> #topic Announcements 14:02:44 <slaweq> First of all: we are now in the Victoria RC-1 week 14:03:02 <slaweq> so at the end of the week we should have Victoria RC-1 done 14:03:46 <slaweq> I will sync with amotoki later this week and will check release patches for neutron 14:03:53 <slaweq> next 14:03:57 <slaweq> TC/PTL nomination starts 22.09 23:45 UTC (today): http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017306.html 14:04:31 <slaweq> so if You want to be Neutron PTL or join TC group, it's almost time to send Your nomination 14:05:50 <slaweq> I'm going to nominate myself for next cycle as neutron PTL 14:06:30 <ralonsoh> +1 14:06:30 <lajoskatona> Long live the King :-) 14:07:05 <slaweq> thx :) 14:07:06 <mlavalle> +1 14:07:28 <slaweq> ok, next one 14:07:43 <slaweq> also related to the end of the release cycle a bit 14:07:43 <bcafarel> late +1, go Hulk PTL! 14:07:46 <slaweq> Virtual PTG in October (26-30): http://lists.openstack.org/pipermail/openstack-discuss/2020-August/016424.html 14:07:58 <slaweq> etherpad: https://etherpad.opendev.org/p/neutron-wallaby-ptg - please add ideas of the topics there, it is still almost empty 14:08:37 <slaweq> we have scheduled with gibi one nova-neutron session for Wednesday 1400 UTC 14:08:50 <gibi> o/ 14:08:55 <slaweq> hi gibi :) 14:09:07 * gibi is lurking 14:10:14 <slaweq> and that's all announcements/reminders from me for today 14:10:22 <slaweq> anything else You want to share with the team? 14:11:36 <bcafarel> to confirm with amotoki (from this morning's discussion), but we will also probably make stable releases to go along rc-1 14:13:04 <slaweq> bcafarel++ 14:13:37 <slaweq> ok, lets move on 14:13:38 <slaweq> #topic Blueprints 14:13:56 <slaweq> I moved 3 BPs for RC-1 https://launchpad.net/neutron/+milestone/victoria-rc1 14:14:17 <slaweq> https://blueprints.launchpad.net/neutron/+spec/metadata-over-ipv6 is just missing some tempest scenario tests and patches are in the gate already 14:14:36 <slaweq> https://bugs.launchpad.net/neutron/+bug/1882804 and https://blueprints.launchpad.net/neutron/+spec/local-ip-prefix-in-metering-rules got FFE 14:14:37 <openstack> Launchpad bug 1882804 in neutron "RFE: allow replacing the QoS policy of bound port" [Wishlist,Confirmed] - Assigned to Lajos Katona (lajos-katona) 14:14:59 <slaweq> patches for both are close to merge IMO 14:15:07 <slaweq> please review them asap if You can: 14:15:12 <slaweq> https://review.opendev.org/#/q/topic:bug/1882804+(status:open+OR+status:merged) 14:15:17 <slaweq> https://review.opendev.org/#/q/topic:bug/1889431+(status:open+OR+status:merged) 14:15:48 <slaweq> other not finished BPs I moved to "neutron next" for now and I will schedule them for Wallaby cycle later 14:16:00 <slaweq> anything You want to add regarding BPs? 14:17:36 <mlavalle> \regarding the metadata over ipv6 test 14:17:45 <mlavalle> is anybody writing them? 14:18:04 <slaweq> mlavalle: yes 14:18:07 <mlavalle> ok 14:18:13 <slaweq> rubasov did them already, see https://review.opendev.org/#/q/topic:metadata-ipv6+(status:open+OR+status:merged) 14:18:26 <slaweq> there is neutron-tempest-plugin patch with those tests 14:19:36 <slaweq> btw. most important patches related to BPs scheduled for rc1 are in https://tinyurl.com/vezk6n6 14:19:48 <slaweq> please check this list and review those patches 14:20:02 <mlavalle> ok 14:20:16 <slaweq> I would like to have it merged before thursday so we can release rc1 with them already 14:21:33 <slaweq> ok, lets move on 14:21:36 <slaweq> #topic Community Goals 14:21:56 <slaweq> any updates about zuul v3 or migration to Ubuntu 20.04 ? 14:22:51 <tosky> I've only observed a few updates, but I'm not sure about the stauts 14:22:53 <tosky> status* 14:23:09 <tosky> (the 2 legacy grenade jobs left) 14:23:32 <lajoskatona> yeah, odl is one 14:23:38 <slaweq> for neutron-ovn-grenade job I'm almost done 14:23:42 <slaweq> patch is https://review.opendev.org/#/c/729591/ 14:23:47 <slaweq> and it works fine now 14:23:58 <slaweq> but it needs patch https://review.opendev.org/#/c/752412/ in devstack 14:24:01 <haleyb> slaweq: last i saw that was not installing ovs? no ovsdb-tool ? 14:24:04 <haleyb> ah 14:24:18 <slaweq> haleyb: yes, but now it works fine 14:24:43 <tosky> slaweq: isn't there a POST_FAILURE? 14:24:44 <slaweq> I think that we have another bug there as in fact ovn_agent module can't install ovn in other way than compiling it from source 14:24:56 <slaweq> tosky: damm, it is 14:25:03 <slaweq> but previous run was passing 14:25:23 <slaweq> so I'm not done and I will need to verify it again 14:25:43 <bcafarel> I commented on neutron-tempest-plugin review, it seems to be mostly OK (1 or 2 tests failing on recheck, but different each time) 14:26:40 <bcafarel> ( https://review.opendev.org/#/c/748367/ ) 14:27:54 <slaweq> tosky: I see that all tests passed in that grenade job 14:28:07 <slaweq> I'm not sure why it failed really at the end 14:28:24 <slaweq> can You maybe take a look also, as You probably have more experience with grenade jobs then me :) 14:28:28 <tosky> slaweq: it seems it failed while collecting the logs on the compute node: https://zuul.opendev.org/t/openstack/build/be55c09398164f03ad32e74fd06ddb98/console 14:28:54 <tosky> not sure why it happens, the base grenade job should be multinode-enabled 14:29:35 <slaweq> I rechecked it so lets see if that will happen again 14:29:46 <slaweq> as I said, I saw this job passing in previous run 14:30:17 <tosky> and it passed in revision 17, and failed in 18 and newer 14:30:43 <tosky> but there are no structural changes 14:30:46 <tosky> mmhhh 14:31:34 <slaweq> there are only logs added by haleyb https://review.opendev.org/#/c/729591/17..18/zuul.d/grenade.yaml 14:32:06 <haleyb> yes, copied those from another ovn review 14:32:31 <slaweq> if it will fail again, I will check without those logs to ensure that this isn't a culprit 14:34:24 <slaweq> ok, so I think we discussed zuulv3 migration 14:35:11 <slaweq> and speaking about focal, I think that as bcafarel mentioned in the comment to https://review.opendev.org/#/c/748367/ - it seems that it may be related to issue which nova have on focal 14:35:38 <slaweq> I think we will need to wait a bit until this will be solved and try then again 14:36:01 <slaweq> because now our scenario jobs aren't stable for sure on Ubuntu Focal 14:37:11 <slaweq> with that I think we can move on to the next topic 14:37:19 <slaweq> thx for updates on community goals 14:37:25 <slaweq> #topic Bugs 14:37:32 <lajoskatona> not sure what is the 'offcial' solution, but for bgpvpn I disabled those failing tests temprorary 14:38:00 <slaweq> lajoskatona: yes, but for neutron-tempest-plugin we would need to disable all scenario tests basically :) 14:38:46 <slaweq> ok, lets get back to bugs 14:38:54 <slaweq> hongbin was bug deputy last week 14:39:03 <slaweq> his report is at http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017432.html 14:39:24 <slaweq> I see there 4 bugs which aren't assigned to anyone 14:39:36 <slaweq> 2 ovn related things: 14:39:38 <slaweq> https://bugs.launchpad.net/neutron/+bug/1896205 14:39:39 <openstack> Launchpad bug 1896205 in neutron "AttributeError: 'NoneType' object has no attribute 'db_find_rows' during neutron-server startup" [Undecided,New] 14:39:40 <slaweq> and 14:39:44 <slaweq> https://bugs.launchpad.net/neutron/+bug/1896203 14:39:45 <openstack> Launchpad bug 1896203 in neutron "TypeError: Cannot look up record by empty string in check_for_igmp_snoop_support task" [Undecided,New] 14:40:03 <slaweq> maybe jlibosva You can take a look and triage them 14:40:08 * jlibosva looks 14:40:32 <slaweq> thx jlibosva :) 14:40:39 <jlibosva> I think the db_find_rows error happens when there is no active connection to OVN DBs 14:40:45 <jlibosva> we have a patch for that in ovsdbapp 14:40:59 <jlibosva> https://review.opendev.org/#/c/752092/ 14:41:00 <slaweq> would be great, so please just link patch there :) 14:41:48 <slaweq> we have also one related to L3HA https://bugs.launchpad.net/neutron/+bug/1895950 14:41:49 <openstack> Launchpad bug 1895950 in neutron "keepalived can't perform failover if the l3 agent is down" [Medium,New] 14:41:58 <slaweq> if there will be no other volunteers I will take a look into that 14:42:13 <slaweq> and I think that haleyb's proposal there is good way to solve this 14:42:48 <haleyb> slaweq: by migrating to OVN? :) 14:42:52 <slaweq> but if You have any other ideas, please write a comment (or assign it to Your self :)) 14:43:09 <slaweq> haleyb: yeah, that would also works but I think it's not that easy :P 14:43:56 <slaweq> and there is one more bug which I wanted to talk about 14:43:58 <slaweq> https://bugs.launchpad.net/neutron/+bug/1895933 14:43:59 <openstack> Launchpad bug 1895933 in neutron "Admin user can do anything without the control of policy.json" [Medium,Confirmed] 14:44:33 <slaweq> we have in Neutron everywhere hardcoded that if context.is_admin() then we are allowing everything 14:44:46 <slaweq> do You know why it is like that? 14:45:26 <ralonsoh> but this is the goal of the admin user 14:45:30 <ralonsoh> isn't it? 14:45:54 <slaweq> ralonsoh: I don't really know 14:46:33 <slaweq> because it can be specified in policy.json and You can theoretically forbid something for such user 14:46:47 <slaweq> at least that was my understanding of bug description 14:47:00 <slaweq> do You know how other projects works with that? 14:47:05 <ralonsoh> no idea 14:47:21 <ralonsoh> but we need context.is_admin() to perform some actions 14:47:30 <ralonsoh> to retrieve, for example, everything from the DB 14:47:56 <slaweq> ralonsoh: so maybe we should just close this LP as won't fix 14:48:07 <ralonsoh> I'll check it too 14:48:07 <slaweq> that's why I wanted to ask others during the meeting :) 14:48:29 <lajoskatona> amotoki can't give advice, usually He has better view of policies? 14:48:44 <slaweq> yes but I didn't saw him around recently 14:48:51 <slaweq> maybe he is on some pto, idk 14:49:49 <lajoskatona> ok 14:50:55 <slaweq> I will try to ping amotoki about that bug later this week 14:51:18 <slaweq> and that are all bugs from hongbin's report which I wanted to discuss today 14:51:26 <slaweq> this week our bug deputy is haleyb 14:51:30 <slaweq> and next week is lajoskatona's turn 14:51:59 <slaweq> haleyb: lajoskatona are You ok with that? 14:52:11 <lajoskatona> slaweq: yes 14:52:28 <haleyb> yes, good with that 14:52:30 <slaweq> great, thx 14:52:40 <slaweq> any other bugs anyone wants to discuss today? 14:53:06 <slaweq> or anything else You want to discuss with the team today as I don't have anything else 14:54:36 <slaweq> ok, I guess this silence means "no" so we can finish earlier today :) 14:54:42 <slaweq> thx for attending the meeting 14:54:46 <bcafarel> o/ 14:54:50 <lajoskatona> o/ 14:54:51 <slaweq> have a great week! 14:54:52 <slaweq> o/ 14:54:55 <slaweq> #endmeeting