14:00:25 <haleyb> #startmeeting networking 14:00:25 <opendevmeet> Meeting started Tue Oct 3 14:00:25 2023 UTC and is due to finish in 60 minutes. The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:25 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:25 <opendevmeet> The meeting name has been set to 'networking' 14:00:29 <obondarev> o/ 14:00:30 <mtomaska> o/ 14:00:31 <ralonsoh> hello 14:00:34 <rubasov> o/ 14:00:36 <haleyb> i got the right time now 14:01:18 <haleyb> #topic announcements 14:02:17 <ykarel> o/ 14:02:29 <haleyb> we tagged what should be the final RC of neutron for Bobcat, https://review.opendev.org/c/openstack/releases/+/896971 14:02:37 <bcafarel> o/ 14:02:56 <haleyb> i'm guessing that will merge soon as tomorrow is the expected release date 14:03:07 <ralonsoh> we missed https://review.opendev.org/c/openstack/neutron/+/896803, but we couldn't change the RC hash 14:03:17 <ralonsoh> so we'll need to propose it for the next bobcat release 14:03:21 <slaweq> o/ 14:04:20 <haleyb> ralonsoh: i thought you updated the review? 14:04:31 <ralonsoh> yes but I needed to revert it 14:04:44 <ralonsoh> we could only approve the RC version 14:05:09 <ralonsoh> it was not possible to add this hash without creating a new RC 14:05:14 <ralonsoh> and we were out of time 14:05:30 <haleyb> ralonsoh: ah, i somehow never got the update 14:06:08 <haleyb> we can just propose a new tag in a short while 14:06:14 <ralonsoh> yes 14:07:51 <haleyb> I would like to thank ralonsoh for all his work and guidance over the many cycles he was PTL! 14:08:25 <slaweq> ralonsoh++ thx for all Your work as PTL :) 14:08:32 <ralonsoh> thanks 14:08:38 <lajoskatona> +1 14:08:40 <ykarel> +1 14:09:21 <mtomaska> +1 14:10:17 <haleyb> the etherpad for Caracal vPTG is up, please add your name if you are attending 14:10:20 <haleyb> https://etherpad.opendev.org/p/oct2023-ptg-neutron 14:10:48 <haleyb> https://openinfra-ptg.eventbrite.com/ is signup 14:11:27 <haleyb> I will send out a doodle to vote on times people are available 14:11:29 <lajoskatona> thanks, I missed the registration mail 14:12:57 <haleyb> there is also a cycle highlights presentation this thursday, ralonsoh will give an update for neutron 14:13:45 <ralonsoh> at 13:30UTC 14:13:50 <ralonsoh> https://www.google.com/url?q=https://streamyard.com/ydfc48bmf2&sa=D&source=calendar&ust=1696751231543261&usg=AOvVaw11KOUAfAr1u2DAMbp5Lqcv 14:13:51 <haleyb> openinfra.dev/live is the place if you want to watch i believe 14:13:58 <haleyb> ah, that's a better link 14:14:05 <ralonsoh> shorter link: https://streamyard.com/ydfc48bmf2 14:14:54 <slaweq> 13:30 UTC? On the webpage it says 1400 14:15:11 <ralonsoh> I have the invitation starting at 1330 14:15:27 <ralonsoh> maybe this is just for the people presenting 14:15:34 <slaweq> ahh, but You are speaker and need to be earlier probably 14:15:44 <ralonsoh> yes, my bad 14:15:48 <slaweq> event will start 1400 UTC 14:15:57 <slaweq> no problem, I just wanted to make sure :) 14:17:18 <haleyb> any other announcements? 14:17:39 <haleyb> #topic bugs 14:17:45 <haleyb> last week's report is from lajoskatona 14:17:56 <haleyb> https://lists.openstack.org/pipermail/openstack-discuss/2023-October/035254.html 14:18:38 <haleyb> most bugs have owners and/or patches which is great, just a doc one is without 14:19:12 <haleyb> https://bugs.launchpad.net/neutron/+bug/2037263 (low hanging fruit if submittor does not propose change) 14:19:44 <lajoskatona> in comment the opener said they can push the fix to the doc 14:19:57 <haleyb> lajoskatona: ack 14:20:10 <haleyb> there were some high priority bugs 14:20:16 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2037596 14:20:24 <haleyb> Description Failed to invoke the API interface to obtain the address group list 14:20:50 <haleyb> #link https://review.opendev.org/c/openstack/neutron/+/896649 proposed 14:21:12 <haleyb> i will review after meeting see it has another +2 already 14:21:32 <haleyb> next one 14:21:42 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2037717 14:21:43 <haleyb> [OVN] ``PortBindingChassisEvent`` event is not executing the conditions check 14:21:58 <haleyb> #link https://review.opendev.org/c/openstack/neutron/+/896926 14:22:13 <haleyb> thanks for the test on that one ralonsoh 14:22:28 <ralonsoh> yw 14:23:29 <haleyb> there was also 3 ovsdbapp bugs filed, along with patches 14:23:49 <haleyb> #link https://bugs.launchpad.net/ovsdbapp/+bug/2037536 14:23:53 <haleyb> #link https://bugs.launchpad.net/ovsdbapp/+bug/2037573 14:24:09 <haleyb> #link https://bugs.launchpad.net/ovsdbapp/+bug/2037652 14:24:36 <lajoskatona> yes all ovsdb bugs are addressed 14:24:45 <haleyb> otherwiseguy seems to have already reviewed :) 14:25:07 * otherwiseguy jumps 14:25:41 <haleyb> sorry to scare you terry :) 14:26:00 <otherwiseguy> haleyb: yeah, i still have a bit more review to do. I'll get to it today. 14:26:10 <haleyb> next bug is a gate failure from last week that is still unsolved 14:26:15 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2037239 14:26:21 <haleyb> neutron-tempest-plugin-openvswitch-* jobs randomly failing in gate 14:27:07 <slaweq> I saw many failures in those scenario jobs this week :/ 14:27:14 <ralonsoh> this is the most serious issue we have right now 14:27:18 <lajoskatona> yes, many rechecks 14:27:22 <ralonsoh> and is blocking the CI almost 100% of the times 14:27:25 <slaweq> probably as a workaround temporary we can disable "l3_ha" in those jobs 14:27:41 <slaweq> so routers will be always active on the L3 agent as it will be legacy routers 14:27:42 <lajoskatona> last week I tried to find something in the logs but nothing which could be suspiciuo 14:27:53 <haleyb> i will spend a little time today looking, i can't tell if the OVS never plugs the port, or if it's just a little late and causes the l3-agent to fail 14:27:57 <slaweq> but we will lost coverage for L3ha then so I don't think it's good permanent solution 14:28:34 <ykarel> i too digged there but couldn't trace what exactly causing it, but seems the thread processing those events are just stuck 14:29:27 <ralonsoh> in the l3 agent? 14:29:32 <ralonsoh> or the API? 14:29:43 <ykarel> yes l3-agent basically keepalived-state-change 14:29:51 <ralonsoh> pffff ok 14:30:21 <haleyb> slaweq: is it easy enough to disable l3_ha in the yaml? as opposed to marking failed tests as unstable? 14:30:43 <haleyb> ykarel: so keepalived-state-change just sat there? 14:30:48 <ykarel> locally it reproduces in many runs in the vm, but in CI it's reproduces quite frequently 14:31:14 <slaweq> https://github.com/openstack/neutron-tempest-plugin/blob/master/zuul.d/master_jobs.yaml#L258 14:31:28 <slaweq> just change this to false and routers will be created as non-ha 14:31:44 <slaweq> so L3 agent will go through different path configuring them 14:32:04 <ykarel> haleyb, so it stuck differently sometimes in handle_initial_state, sometimes in ip_monitor thread around the calls to privileged methods 14:32:09 <slaweq> maybe that will workaround the issue and if so, it will also be clear pointer where the issue may be :) 14:32:45 <haleyb> slaweq: ack, i can send a change out after meeting 14:33:14 <haleyb> ykarel: can you add any notes to the bug? 14:33:32 <ykarel> haleyb, ack will do 14:34:01 <haleyb> ykarel: thanks 14:34:05 <haleyb> any other bugs to talk about? 14:34:51 <haleyb> #topic specs 14:35:55 <haleyb> #link https://review.opendev.org/c/openstack/neutron-specs/+/891204 is the only spec in review at the moment, looks like it needs an update based on reviews 14:36:42 <haleyb> ralonsoh: was there anything new from the drivers meeting last week? i was unable to attend and didn't look at notes yet 14:36:56 <ralonsoh> yes, I have one question 14:36:58 <ralonsoh> one sec 14:37:10 <ralonsoh> about https://bugs.launchpad.net/neutron/+bug/2037294 14:37:25 <ralonsoh> I presented this RFE in behalf of lucasagomes 14:37:40 <ralonsoh> but I thought a change in the oVN L3 scheduler was needed 14:37:55 <lucasagomes> o/ 14:37:55 <ralonsoh> lucasagomes, confirmed that the change is much more smaller and the scope 14:38:08 <ralonsoh> so I would propose not to request a spec for this rfe 14:38:11 <ralonsoh> just a patch 14:38:14 <lucasagomes> yeah, haleyb ralonsoh no L3 changes needed. The RFE is only about external ports 14:38:44 <lucasagomes> GW ports will continue to be scheduled as they are today (using the enable-chassis-as-gw flag) 14:39:26 <ralonsoh> so if you agree, now the scope is smaller, we can just accept a patch (with the needed documentation and reno) 14:39:26 <slaweq> ++ 14:39:27 <haleyb> lucasagomes: ack, will remove the rfe-approved tag then and look for the change 14:39:46 <lajoskatona> ack 14:39:50 <haleyb> or guess can leave since it is a slight change 14:39:51 <lucasagomes> thanks folks! 14:40:22 <haleyb> ralonsoh: regarding specs, we can merge this now - https://review.opendev.org/c/openstack/neutron-specs/+/891204 14:40:50 <ralonsoh> I would need to read this spec again 14:40:57 <haleyb> it's just creating the folder 14:41:02 <ralonsoh> but there are some questions there by dalvarez 14:41:17 <haleyb> ah, bad copy/paste 14:41:27 <haleyb> #link https://review.opendev.org/c/openstack/neutron-specs/+/891342 14:41:33 <ralonsoh> ah yes, sure 14:41:42 <ralonsoh> done 14:41:52 <opendevreview> yatin proposed openstack/neutron-tempest-plugin master: [DNM] Test l3_ha=False https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/897233 14:42:28 <haleyb> moving on 14:42:32 <haleyb> #topic community_goals 14:43:23 <haleyb> there are still just two open goals 14:43:30 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2026182 14:43:48 <haleyb> S-RBAC, https://review.opendev.org/c/openstack/neutron/+/886724 14:43:50 <slaweq> for that I still have patch https://review.opendev.org/c/openstack/neutron/+/886724 14:44:05 <slaweq> I will try to rebase it and address commnets there in next days 14:44:14 <slaweq> so we hopefully be able to merge it finally 14:44:20 <haleyb> slaweq: ack thanks 14:44:23 <opendevreview> Rodolfo Alonso proposed openstack/neutron master: WIP - Create a keepalived-state-change script in bash https://review.opendev.org/c/openstack/neutron/+/836140 14:45:05 <haleyb> second is https://bugs.launchpad.net/neutron/+bug/1999774 14:45:13 <haleyb> neutronclient deprecation 14:45:29 <haleyb> lajoskatona: is there anything that needs review there? 14:45:33 <lajoskatona> the usual etherpad: https://etherpad.opendev.org/p/python-neutronclient_deprecation 14:45:45 <lajoskatona> but no fresh reviews, I still have no time for this topic 14:46:45 <haleyb> lajoskatona: ack, i'll update the list in the meeting notes as some have merged 14:47:11 <haleyb> #topic on_demand 14:47:13 <lajoskatona> haleyb: thanks 14:47:25 <haleyb> is there anything else people want to discuss? 14:48:06 <ralonsoh> no thanks 14:48:29 <haleyb> oh, and i forgot during 'bugs' that slaweq is the deputy this week, i am next week 14:48:42 <slaweq> yeap, I'm on it :) 14:48:45 <slaweq> but thx for reminder 14:49:14 <haleyb> and CI meeting in :11 minutes for those interested 14:49:14 <ykarel> Also CI meeting will be in next 11 minutes, IRC this time 14:49:29 <lajoskatona> +1 14:49:44 <haleyb> ok, thanks everyone for attending, have a good week! 14:49:48 <haleyb> #endmeeting