13:00:27 <haleyb> #startmeeting networking 13:00:27 <opendevmeet> Meeting started Tue Jul 15 13:00:27 2025 UTC and is due to finish in 60 minutes. The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:27 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:27 <opendevmeet> The meeting name has been set to 'networking' 13:00:37 <haleyb> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, haleyb, ralonsoh 13:00:37 <mlavalle> \o 13:00:42 <haleyb> o/ 13:00:46 <ralonsoh> hello 13:00:50 <elvira> o/ hi 13:00:59 <ykarel> o/ 13:01:18 <rubasov> o/ 13:01:34 <cbuggy> o/ 13:01:50 <bcafarel> o/ 13:01:56 <haleyb> #announcements 13:01:58 <haleyb> hi everyone 13:02:09 <haleyb> We are currently in Week R-11 of Flamingo 13:02:15 <haleyb> Our next milestone in this development cycle will be Flamingo-3, week of August 25th 13:02:22 <haleyb> Final 2025.2 Flamingo release: October 3rd, 2025 13:03:11 <haleyb> there were a number of project releases the past week, including neutron 27.0.0.0b1 13:04:01 <haleyb> so now other projects can update requirements if they need something in this cycle's neutron code 13:04:14 <haleyb> The next OpenInfra PTG will take place October 27-31, 2025 and registration for the event is now open 13:04:18 <haleyb> #link https://ptg.openinfra.dev/ 13:04:43 <haleyb> #link https://releases.openstack.org/flamingo/schedule.html (missed that) 13:04:52 <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 13:05:38 <haleyb> And just a reminder to check the priorities dashboard, and use the RP +1/+2 in changes that need attention 13:06:05 <haleyb> that was all the announcements i had, any others? 13:06:40 <haleyb> ok, moving on 13:06:46 <haleyb> #topic bugs 13:06:56 <haleyb> ralonsoh was the deputy last week, his report is at 13:07:03 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/AGG77SQFIKAIEUUXFSQYPB3KZC32OGVQ/ 13:07:56 <haleyb> all high/medium/low were taken and patches out 13:08:02 <haleyb> two are incomplete 13:08:35 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2116335 13:08:36 <haleyb> gateway_port: of NAT configured on logical router: is not a valid distributed gateway port on that router 13:08:44 <haleyb> ralonsoh asked for more information here 13:08:56 <ralonsoh> yes, I don't know what happened after the nodes reboot 13:09:09 <ralonsoh> so we need to know if the GW port were bound correctly 13:09:23 <ralonsoh> maybe there was a change in the chassis/chassis_private registers 13:09:33 <haleyb> ack, let's see if there is a response 13:10:12 <haleyb> next incomplete is 13:10:22 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2116723 13:10:23 <haleyb> VPN Perfect Forward Secrecy (PFS) is deactivated although configured as active 13:10:41 <haleyb> there was just a response 2 hours ago 13:10:58 <haleyb> i don't understand 'neutron version 2.0' though 13:11:02 <ralonsoh> yes but I'm not sure about this anwer 13:11:04 <ralonsoh> right 13:11:29 <ralonsoh> and he is also stating that "I don't use VPNaaS Plugin" 13:11:45 <ralonsoh> but the "openstack vpn" commands are from neutron-vpnaas 13:12:20 <ralonsoh> anyway, I'll try locally to reproduce this issue, but with master 13:13:20 <frickler> oh, osism is our downstream packaging, we might have added it by default there. I'll have someone have a look at this on our end 13:13:22 <haleyb> perhaps someone from vpnaas can take a look, bugs.rst shows Dongcan Ye as contact not sure if he's in the channel 13:13:56 <haleyb> frickler: great, thanks! 13:15:33 <haleyb> https://osism.tech/ nice 13:16:32 <haleyb> that was all the unassigned bug 13:16:41 <haleyb> any other bugs people want to discuss? 13:17:05 <haleyb> thanks for the report ralonsoh 13:17:13 <ralonsoh> yw 13:17:23 <haleyb> this week the deputy is sahid, next week is jlibosva 13:18:42 <haleyb> hopefull that works for both - sahid, let me know if that is ok 13:19:09 <haleyb> #topic specs 13:19:20 <haleyb> #link https://review.opendev.org/q/project:openstack/neutron-specs+status:open 13:20:02 <haleyb> we still have a backlog there, please review so we can move some forward 13:20:30 <haleyb> and if you have a spec that has a -1 because it was targeted at a past release, please update accordingly 13:21:59 <haleyb> #topic community goals 13:22:22 <haleyb> i do not see lajos here, but there is one neutronclient change in-flight 13:22:34 <haleyb> #link https://review.opendev.org/q/topic:%22bug/1999774%22+status:open+project:openstack/python-openstackclient+or+status:open+owner:self+project:openstack/openstacksdk 13:22:43 <haleyb> that is the dashboard for current ones 13:23:19 <haleyb> #link https://review.opendev.org/c/openstack/python-openstackclient/+/950494 actually has two +2's so is good to go 13:23:59 <haleyb> stephenfin: can you take a look at ^^ ? i do not have +W powers there 13:24:45 <haleyb> and finally eventlet 13:25:08 <stephenfin> haleyb done, though we should bump the minimum version of SDK in a follow-up 13:25:56 <haleyb> stephenfin: ack, unless you want to do there, up to you 13:26:43 <haleyb> i missed that comment 13:28:22 <stephenfin> ack, done that too 13:28:51 <haleyb> great, one more neutronclient change off the list 13:29:07 <haleyb> ralonsoh: ok, back to eventlet 13:29:21 <ralonsoh> very little progress 13:29:28 <ralonsoh> I'm finishing the UTs 13:29:36 <ralonsoh> but still stuck in the FT... 13:29:52 <ralonsoh> I'm starting with the OVS related ones, in particular with the OVS agent 13:30:12 <ralonsoh> because now we actually spawn certain threads (that I'm not aware) 13:30:23 <ralonsoh> and these threads are not stopped, the tests never end 13:30:36 <ralonsoh> that's all (summary: I'm blocked) 13:31:09 <haleyb> ralonsoh: spawned via the ThreadPollExecutor class or somesuch? 13:31:25 <ralonsoh> of just directly as threads 13:31:34 <ralonsoh> but yes, most of them vie threadpoolexecutor 13:31:50 <ralonsoh> but the problem is also how the ovs agent is spawned, using the os-ken hub 13:32:02 <ralonsoh> (all OVS agent is hijacked by os-ken...) 13:34:06 <haleyb> ralonsoh: i know there was an os-ken change to default to native threads, do we need to merge that this cycle? i think we over-ride the setting 13:34:10 <haleyb> #link https://review.opendev.org/c/openstack/os-ken/+/938337 13:34:20 <ralonsoh> no, this is not the problem (I think) 13:34:45 <ralonsoh> the problem is how all the ovs agent is spawned, using the os-ken hub as base executor 13:35:08 <ralonsoh> we are forcing this new "threading" module (I need to re-check that) 13:35:33 <ralonsoh> anyway, tomorrow, if I have time, I'll spend the whole day on this 13:35:41 <ralonsoh> we need to fix the FTs ASAP 13:35:52 <ralonsoh> ""fix"" 13:36:00 <haleyb> os.environ['OSKEN_HUB_TYPE'] = 'native' seems to be set in ovs agent code 13:36:32 <ralonsoh> yes 13:36:45 <haleyb> ralonsoh: ack, thanks for the work, let me know if you need any help 13:36:49 <ralonsoh> thanks 13:37:41 <haleyb> #topic on-demand 13:38:16 <haleyb> i had forgotten one thing in announcements - i will be out next week on vacation 13:38:30 <haleyb> so if someone could lead this meeting that would be great 13:38:46 <mlavalle> I can do it 13:38:46 <ralonsoh> for sure, I'll do 13:38:52 <ralonsoh> mlavalle, first 13:39:00 <mlavalle> lol 13:39:09 <haleyb> mlavalle: great, thanks! 13:39:56 <mlavalle> let's synch up on Friday to make sure we cover all the points you have in mind. I'll ping you 13:40:27 <haleyb> i would guess we can just cancel next friday's drivers meeting as well, if there is nothing on agenda by thursday for that can just cancel two weeks 13:40:34 <ralonsoh> ok 13:41:24 <haleyb> mlavalle: ack, at this point we are just a glider getting ready for final approach 13:42:06 <mlavalle> haleyb: lol that easy for me. I know about gliders 13:42:50 <haleyb> yes, i remember the stories :) 13:43:36 <haleyb> at this point, making progress on eventlet is the priority, and fixing the bugs :) 13:43:57 <opendevreview> Rodolfo Alonso proposed openstack/neutron master: [FT] Add a retry method when reading the ``Chassis_Private`` register https://review.opendev.org/c/openstack/neutron/+/955063 13:44:13 <haleyb> alright, any other topics? 13:44:24 <ralonsoh> not from me 13:44:39 <mlavalle> neither from me 13:44:55 <haleyb> ok, thanks for attending, have a good week! 13:44:57 <haleyb> #endmeeting