13:00:24 <haleyb> #startmeeting networking
13:00:24 <opendevmeet> Meeting started Tue Apr 15 13:00:24 2025 UTC and is due to finish in 60 minutes.  The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:00:25 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:00:25 <opendevmeet> The meeting name has been set to 'networking'
13:00:36 <haleyb> hopefully everyone remembered we moved this an hour earlier
13:00:50 <mlavalle> I did
13:01:12 <lajoskatona> o/
13:01:15 <sahid> o/
13:01:48 <bcafarel> I did not but I was checking IRC so good timing :)
13:01:54 <haleyb> #announcements
13:02:27 <haleyb> Just wanted to thank everyone that attended the PTG last week
13:03:01 <haleyb> #link https://etherpad.opendev.org/p/apr2025-ptg-neutron
13:03:04 <ralonsoh> hello
13:03:31 <haleyb> you can find all the other etherpads at
13:03:38 <haleyb> #link https://ptg.opendev.org/etherpads.html
13:04:26 <haleyb> So we are now in Flamingo, schedule is at
13:04:29 <haleyb> #link https://releases.openstack.org/flamingo/schedule.html
13:04:36 <haleyb> We are currently in Week R-24
13:04:44 <haleyb> Final 2025.2 Flamingo release: October 3rd, 2025
13:05:28 <haleyb> Reminder: If you have a topic for the drivers meeting on Friday, please add it to the wiki @ https://wiki.openstack.org/wiki/Meetings/NeutronDrivers
13:06:01 <haleyb> I had no other announcements, did anyone else?
13:06:41 <haleyb> ok, we can move on to bugs
13:06:42 <haleyb> #topic bugs
13:07:23 <haleyb> sahid was the bug deputy last week and ralonsoh the week before, we never officially went through that list
13:07:27 <haleyb> we can start with last week
13:07:35 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/WYDAWQVUJ72I6PS26H5YGESBBMN3VSLW/
13:08:32 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2106374/
13:08:39 <haleyb> neutron-server error with ml2_distributed_port_bindings
13:09:09 <haleyb> this is ml2/ovs
13:09:50 <haleyb> oh, and it's AlmaLinux, openstack-neutron-20.3.1-1.el9s
13:10:22 <haleyb> what release is that?
13:11:15 <lajoskatona> yoga perhasp
13:11:31 <lajoskatona> https://releases.openstack.org/teams/neutron.html#team-yoga-neutron
13:11:45 <haleyb> yes, looks like yoga
13:12:28 <haleyb> and being a distro package i'm not sure how much time we should spend
13:12:34 <lajoskatona> without reproduction steps hard to say anything....
13:12:51 <haleyb> i will add a comment to try and reproduce on something much newer
13:12:59 <sahid> looks like some sort of eadge case difficult to reproduce, I imagine we can try to at leat handle the exception and correctly log something
13:13:32 <sahid> that is why I was initially proposed a low hanging fruit
13:13:35 <haleyb> sahid: is the code still the same?
13:13:55 <sahid> no idea I have not checked
13:15:03 <ralonsoh> the code is the same since 10 years ago
13:15:07 <ralonsoh> the logs are legit
13:15:11 <haleyb> either way i think we should ask to reproduce on newer release as yoga is unmaintained
13:15:15 <ralonsoh> this could be a leftover in the DB
13:15:32 <ralonsoh> because the table expects the {port,host} to be unique
13:15:41 <sahid> they specifically mention they are not able to reproduce on stagging
13:15:55 <sahid> something probably related with the load
13:15:56 <ralonsoh> that is something to be checked in this user DB
13:17:25 <haleyb> let's add a comment there to try and help them along
13:17:38 <sahid> should i mark it as wont fix? and ask report to try with newer release?
13:18:17 <ralonsoh> no, we need first to triage it
13:18:53 <haleyb> sahid: ask them if they can reproduce on something that is supported upstream unless we can reproduce on master ourselves
13:19:12 <sahid> ack
13:19:38 <haleyb> next bug
13:19:43 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2106357/
13:19:51 <haleyb> [ovn-octavia-provider] Loadbalancer member list show operating status as ONLINE when member is created for a SHUTOFF state vm
13:20:00 <haleyb> on release 8.0.0 which is epoxy
13:21:17 <haleyb> is froyo around?
13:21:27 <froyo> o/
13:21:54 <haleyb> hey froyo - can you take a look at that bug? ovn-octavia-provider related
13:21:57 <froyo> I'll take a look to it!
13:22:06 <haleyb> thanks!
13:22:10 <froyo> yw
13:22:49 <haleyb> next one
13:22:54 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2106690/
13:23:01 <haleyb> ovn-sb leader was changes frequently
13:23:49 <haleyb> i'm not sure if they have something broken in their cluster
13:24:05 <haleyb> WARN|tcp:10.230.186.15:34182: receive error: Connection reset by peer
13:24:10 <haleyb> INFO|current entry eid 24736b60-3f1b-4f13-b3c4-21ad1c20e500 does not match prerequisite 0e84e4da-58a9-45ed-93c2-29fd9ca64469 in execute_command_request
13:24:25 <haleyb> i don't exactly know what that second one is
13:24:47 <ralonsoh> https://mail.openvswitch.org/pipermail/ovs-discuss/2020-October/050758.html
13:24:53 <ralonsoh> "This is expected in cluster mode when there are concurrent transactions
13:24:53 <ralonsoh> initiated from different cluster nodes. It is recommended to connect to the
13:24:53 <ralonsoh> leader for write-heavy clients. Occasional conflict is ok and the
13:24:53 <ralonsoh> transaction will be resubmitted."
13:26:54 <haleyb> i think neutron defaults to leader-only for writes, i'm remembering an old ihar change that made it possible to not do that, but there are ovsdb implications from waht i remember
13:27:58 <haleyb> i think they need to debug their OVN setup
13:28:53 <haleyb> i'll leave a comment with some info
13:28:56 <haleyb> next bug
13:29:09 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2106793
13:29:19 <haleyb> after running ``os-nova-install.yml`` lost instances connectivity
13:29:46 <haleyb> is this a heat template?
13:31:00 <lajoskatona> perhaps opaenstack-ansible, but no idea
13:31:16 <ralonsoh> I would like to know what this yaml file is (ansible role, etc)
13:31:20 <haleyb> i don't think the issue here is with neutron, but with whatever service they are using not restarting things correctly
13:31:59 <ralonsoh> I don't think they could change the cpu libvirt configuration like this
13:32:29 <ralonsoh> it could be useful the nova compute logs too
13:32:38 <ralonsoh> anyway, more information
13:32:58 <haleyb> yes, i'll ask some questions
13:33:01 <haleyb> next one
13:33:18 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2106776
13:33:26 <haleyb> Unable to enable/disable src_nat per external gateway
13:34:05 <haleyb> i remember noonedeadpunk mentioning this last week but i had no time to look further
13:34:34 <ralonsoh> that looks like a legit bug of the multi GW feature
13:34:50 <noonedeadpunk> o/
13:35:10 <noonedeadpunk> I also did not spend any time on digging into the code
13:37:41 <haleyb> sorry, had a local question :)
13:37:56 <haleyb> like ralonsoh said, seems a legit bug maybe a corner case
13:38:54 <haleyb> will just need an owner to dig into it
13:39:19 <ralonsoh> I can but not this week
13:40:10 <haleyb> at least it has a reproducer :)
13:40:33 <haleyb> that was the last bug on sahid's list
13:40:59 <haleyb> ralonsoh: did you have any from the previous week that need discussion? as we never did look at them in a meeting
13:41:22 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/message/437AQKV6NFINO7RPDL7SEZJ7NIBZ24TB/
13:41:41 <ralonsoh> no, all addressed
13:41:46 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2105977
13:41:54 <haleyb> Random error in ``test_packets_of_any_connection_state_can_reach_dest`` test
13:42:10 <haleyb> you filed that bug unassigned
13:42:14 <ralonsoh> I'm checking this oen
13:42:21 <ralonsoh> I didn't assign it, my bad
13:42:34 <haleyb> np, the others are all assigned
13:44:25 <haleyb> Our bug count is down to 753 (epoxy releasing closed a bunch :)
13:44:38 <haleyb> this week the deputy is lajoskatona, next week is ykarel
13:44:42 <haleyb> is that good for both?
13:45:20 <lajoskatona> I am on it
13:45:34 <haleyb> ack, thanks
13:45:39 <haleyb> we can move on
13:45:43 <haleyb> #topic community goals
13:46:10 <haleyb> having just had the ptg, not sure how much we need to discuss
13:46:25 <haleyb> are there any patches that need review and can move forward on eventlet?
13:46:32 * haleyb has not looked today
13:46:33 <lajoskatona> nothing from my side
13:46:41 <ralonsoh> yeah, one sec
13:46:52 <sahid> still have some in mine
13:47:05 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/946992 and the upper 4
13:47:35 <haleyb> ack, will look
13:48:14 <haleyb> #topic on-demand
13:48:25 <haleyb> any additional topics?
13:48:34 <ralonsoh> no thanks, all good
13:48:44 <lajoskatona> nothing from my side, thanks
13:48:51 <haleyb> ok, thanks for attending everyone, have a good week
13:48:53 <haleyb> #endmeeting