13:00:46 <haleyb> #startmeeting networking 13:00:46 <opendevmeet> Meeting started Tue Sep 2 13:00:46 2025 UTC and is due to finish in 60 minutes. The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:46 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:46 <opendevmeet> The meeting name has been set to 'networking' 13:00:53 <haleyb> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, haleyb, ralonsoh, sahid 13:00:54 <mlavalle> \o 13:00:57 <slaweq> o/ 13:00:58 <cbuggy> o/ 13:01:00 <mtomaska> o/ 13:01:10 <haleyb> hi everyone 13:01:32 <ralonsoh> hello 13:02:06 <lajoskatona> o/ 13:02:07 <haleyb> #announcements 13:02:29 <haleyb> We are currently in Week R-4 of Flamingo 13:02:47 <haleyb> Flamingo-3 was last week 13:03:07 <haleyb> I've been out and have not looked for any release emails 13:03:54 <haleyb> But it does mean we are at feature freeze, so should only be fixing bugs 13:03:55 <ralonsoh> The proposed patches are out 13:03:56 <ralonsoh> https://review.opendev.org/q/releases 13:04:01 <elvira> o/ 13:04:12 <ralonsoh> and last week we have the non client libraries, like n-lib or os-ken 13:04:20 <haleyb> ralonsoh: thanks, i'll take a look after meeting 13:04:56 <haleyb> yes, i did see n-lib and os-ken patches, do we need any updates there? i saw something in backtrace here but don't know 13:05:44 <ralonsoh> no that I'm aware 13:05:45 <haleyb> Final 2025.2 Flamingo release: October 3rd, 2025 13:05:46 <haleyb> #link https://releases.openstack.org/flamingo/schedule.html 13:05:51 <haleyb> ralonsoh: ack, thanks 13:06:15 <lajoskatona> for n-lib we have this: https://review.opendev.org/c/openstack/neutron-lib/+/958786 13:06:31 <ralonsoh> yes, that was for the on-demand section 13:06:34 <lajoskatona> yesterday we slightly discussed during the CI meeting 13:06:49 <lajoskatona> ahh, ok, sorry I didnt read the agenda :-) 13:06:49 <ralonsoh> so we enabled, by default, the hackings tests 13:07:04 <ralonsoh> so any project importing n-lib and using hacking, will inherit these checks 13:07:21 <ralonsoh> this can be disabled locally, this is not a hard stopper 13:07:46 <ralonsoh> but maybe we would like to propose a 3.21.2 release with this patch 13:07:55 <ralonsoh> and update the releases one: https://review.opendev.org/c/openstack/releases/+/959148 13:08:02 <ralonsoh> opinions? 13:08:04 <lajoskatona> yes some local changes in tox.ini I suppose can solve if projects depend on n-lib and failing due to this 13:08:38 <ralonsoh> the issue in the affected project was solved 13:08:51 <ralonsoh> and also n-t-p last week 13:09:06 <lajoskatona> so the patch from tkajinam is not necessary or urgent to be merged to n-lib? 13:09:19 <ralonsoh> I think it is not urgent 13:09:20 <haleyb> ok, that seems to be the patch mentioned previously, I guess if we have not tagged neutron-lib yet we can fix things 13:10:10 <haleyb> but i will defer to others that were here on what to do, if other repos are fine now can leave alone 13:10:37 <lajoskatona> +1 13:12:31 <haleyb> we can discuss more in on-demand, i'll continue 13:12:48 <haleyb> The next OpenInfra PTG will take place October 27-31, 2025 and registration for the event is now open 13:12:55 <haleyb> #link https://ptg.openinfra.dev/ 13:13:00 <haleyb> please sign-up if you haven't already 13:14:18 <haleyb> the PTG will be virtual, during october 27-31 13:16:38 <haleyb> for TC voting, most of you have probably received instructions on that, just a reminder to vote if you want to 13:17:12 <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 13:17:39 <haleyb> and sorry, i was out last week but had forgotten to cancel the meeting, not sure if we have topics for this week 13:18:21 <ralonsoh> one mine 13:18:26 <haleyb> sure 13:18:53 <haleyb> there might be another i can look and will send an email thursday 13:19:05 <mlavalle> Thanks 13:19:13 <haleyb> sorry, i was interrupted here but am back 13:19:26 <haleyb> any other announcements? 13:20:00 <haleyb> #topic bugs 13:20:17 <haleyb> elvira was the bug deputy last week, her report is at 13:20:21 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/MM52DUJJBKJSJH3WD3D2GY7KHMSUAYP7/ 13:20:28 <haleyb> thanks elvira! 13:21:20 <haleyb> first high was the flake8 issue in neutron-lib 13:21:46 <haleyb> https://review.opendev.org/c/openstack/neutron-lib/+/958786 13:22:01 <haleyb> ralonsoh: i saw you just +2'd should we just merge this now? 13:22:22 <ralonsoh> yes but no need for a new release for flamingos, IMO 13:22:36 <ralonsoh> then we'll need to manually add these tests to our project CIs 13:23:24 <haleyb> ok, so a tox.ini change for neutron, etc 13:23:46 <ralonsoh> for example: https://review.opendev.org/c/openstack/neutron/+/956350 13:24:07 <ralonsoh> inherit the n-lib library in additional_dependencies, and then add them in tox 13:26:03 <haleyb> so is that patch enough? just replicate for others like vpnaas, etc 13:26:14 <ralonsoh> yes 13:27:03 <haleyb> basically so we don't regress, got it 13:28:39 <haleyb> ok, we can work on that then 13:29:01 <haleyb> next bug is 13:29:05 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2121649 13:29:11 <haleyb> [n-t-p] Test "test_subport_connectivity" failing randomly in 2025.1 13:29:35 <haleyb> was this discussed at CI meeting yesterday? 13:29:39 <ralonsoh> yes 13:29:49 <ralonsoh> I saw it once but I decided to open a LP bug 13:29:54 <ralonsoh> just to keep an eye on it 13:30:07 <ralonsoh> if I don't see any other ocurrence, I'll close it 13:30:17 <haleyb> ack, sounds good 13:30:40 <haleyb> next bug 13:30:45 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2121605 13:30:52 <haleyb> nova and neutron pagination returns an http, not https href link 13:31:06 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/QEKXP5IYHYMX7PNSVWBJMYMKMJO6MLI2/ 13:31:10 <haleyb> email thread 13:32:08 <ralonsoh> I can check how we are creating these links 13:32:52 <lajoskatona> the and of mail thread concluded that HAproxy settings were behind it am I wrong? 13:32:56 <haleyb> sure. it was unclear from that thread if this was an issue with the load-balancer in front of the API 13:33:48 <ralonsoh> yeah, I think Takashi is right (last mail he sent) 13:34:56 <ralonsoh> in any case, I'll check how neutron is crafting these links and if we are forcing the "http" prefix always 13:35:06 <haleyb> ok, thanks 13:35:39 <haleyb> next is 13:35:47 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2121530 13:35:48 <haleyb> [SR-IOV] Trusted VF port MAC address can be changed from inside the VM 13:35:57 <ralonsoh> I pushed 2 doc patches 13:35:57 <haleyb> which i see ralonsoh picked-up and there is a patch 13:36:13 <ralonsoh> #link https://review.opendev.org/c/openstack/neutron/+/958877 13:36:22 <ralonsoh> #link https://review.opendev.org/c/openstack/neutron/+/958880 (related) 13:36:41 <haleyb> great, i'll take a look after meeting 13:37:08 <haleyb> and last bug is 13:37:12 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2121582 13:37:20 <haleyb> Source / Destination IP address is not respected in firewall group rule FWaaS v2 13:37:26 <haleyb> and i see slaweq picked this up 13:38:35 <haleyb> and this is ovs fwaas 13:38:54 <haleyb> that was all the bugs i had, any others to discuss? 13:39:16 <haleyb> this week slaweq is the deputy, next week is myself 13:40:16 <slaweq> I'm on it 13:40:23 <haleyb> great, thanks! 13:40:26 <slaweq> I mean on bug deputy 13:40:33 <slaweq> and I plan to check this week this issue with fwaas 13:41:07 <elvira> thanks for checking that!! 13:41:07 <haleyb> #topic specs 13:41:12 <haleyb> #link https://review.opendev.org/q/project:openstack/neutron-specs+status:open 13:41:27 <haleyb> so there are a number of open specs 13:41:47 <haleyb> and i see ralonsoh proposed 2026.1 folder and new one 13:41:49 <ralonsoh> I'm planning to actively work on https://review.opendev.org/c/openstack/neutron-specs/+/952166 13:42:07 <ralonsoh> yes, because we passed the 2025.2 mark now 13:42:10 <haleyb> ralonsoh: great 13:42:17 <ralonsoh> so I proposed the DHCP relay one on this new folder 13:42:31 <ralonsoh> in any case, for https://review.opendev.org/c/openstack/neutron-specs/+/952166 13:42:48 <haleyb> right, and i don't know what to do with these older ones - they really just needed updates to change to 2025.2 but no one has responded 13:42:49 <ralonsoh> I would need more information about the underlaying infrastructure 13:44:29 <haleyb> i will put comments in all the old ones asking to update to 2026.1 folder, but if there is no response by beginning of cycle i think we should abandon if no changes have been made 13:45:40 <haleyb> moving on 13:45:43 <haleyb> #topic community goals 13:45:56 <haleyb> lajoskatona: any updates on neutronclient? 13:46:23 <lajoskatona> yes, last Horizon patch is waiting to be merged: https://review.opendev.org/c/openstack/horizon/+/953302 13:46:39 <lajoskatona> it was topic on Horizon meeting soon so I am optimistic: https://meetings.opendev.org/meetings/horizon/2025/horizon.2025-08-27-15.00.log.html#l-18 13:47:33 <haleyb> great, i will try and take a look later this week when i am un-buried 13:47:37 <lajoskatona> other than that I am woring on the Nova patch to cut it to pieces, hope soon I can push a patch for changing networks calls to SDK, but lots of unit tests are failing so slow progress 13:47:48 <lajoskatona> thanks 13:48:16 <haleyb> lajoskatona: ack, and i'm assuming being this late in cycle they might wait until 2026.1 13:48:45 <lajoskatona> yes, 13:49:36 <haleyb> ralonsoh: and what is the latest on eventlet? 13:50:01 <ralonsoh> I almost have the FT/UT patch working 13:50:07 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/952258 13:50:18 <ralonsoh> the problem: more than 100 tests skipped 13:50:33 <ralonsoh> I'm documenting that in this same patch, in the eventlet_removal doc 13:50:51 <ralonsoh> as commented before, the OVS agent is not tested at all in functional tests 13:51:22 <ralonsoh> that's all, basically 13:51:29 <ralonsoh> I'll ping you to review the patch once working 13:52:24 <haleyb> ralonsoh: ack, thanks. are any of the patches in the relation chain ready to go? just to get them out of the way 13:52:45 <ralonsoh> no, this is the only one 13:54:42 <haleyb> #topic on-demand 13:55:06 <haleyb> anything else to discuss? think we already covered the flake8 issue in bugs 13:55:24 <lajoskatona> exactly 13:55:28 <lajoskatona> nothing from me 13:55:44 <ralonsoh> no thanks 13:56:08 <haleyb> ok, thanks for attending everyone and have a good week 13:56:12 <haleyb> #endmeeting