13:00:15 <ralonsoh> #startmeeting networking 13:00:15 <opendevmeet> Meeting started Tue Aug 26 13:00:15 2025 UTC and is due to finish in 60 minutes. The chair is ralonsoh. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:15 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:15 <opendevmeet> The meeting name has been set to 'networking' 13:00:30 <ralonsoh> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, haleyb, ralonsoh, sahid 13:00:34 <mlavalle1> \o 13:00:37 <mtomaska> o/ 13:00:43 <ykarel> o/ 13:01:37 <slaweq> o/ 13:01:41 <bcafarel> o/ 13:01:59 <lajoskatona> o/ 13:02:12 <ralonsoh> Hello all! Let's start 13:02:21 <ralonsoh> Flamingo/2025.2 cycle schedule 13:02:27 <ralonsoh> #link https://releases.openstack.org/flamingo/schedule.html 13:02:44 <ralonsoh> we are in R-5, that is Flamingo-3 milestone 13:03:05 <ralonsoh> that also means feature freeze (no feature patches should be merged) 13:03:16 <ralonsoh> and final release for client libraries 13:03:24 <ralonsoh> #link https://review.opendev.org/q/topic:%22flamingo-final-non-client-libs%22 13:03:36 <ralonsoh> #link https://review.opendev.org/q/topic:%22flamingo-milestone-3%22 13:04:01 <ralonsoh> I think in the next days we'll see the networking projects proposal there 13:04:19 <ralonsoh> Next topic 13:04:20 <ralonsoh> Gazpacho/2026.1 cycle schedule 13:04:27 <ralonsoh> #link https://openinfra.org/ptg/ 13:04:45 <ralonsoh> the PTG will be virtual, during october 27-31 13:04:57 <ralonsoh> please register now to join us during this week 13:05:30 <ralonsoh> I don't think I'm missing anything here, am I? 13:06:11 <lajoskatona> looks complete, thanks 13:06:15 <ralonsoh> ok, let's move on 13:06:23 <ralonsoh> #topic bugs 13:07:04 <ralonsoh> one sec, my internet connection is not working 13:07:44 <ralonsoh> ok, yes, there is no bug mail this week... 13:07:52 <mtomaska> aaaaa I forgot to send my email 13:07:52 <ralonsoh> so let's review the existing ones, the list is short 13:07:59 <mtomaska> but it was short 13:08:15 <ralonsoh> no problem, let's review the list now 13:08:25 <ralonsoh> #link bugs.launchpad.net/neutron/+bug/2121346 13:08:30 <ralonsoh> DNS record is created for port even dns-publish-fixed-ip is not enable on subnet 13:08:46 <mtomaska> that is the only one ^ :) 13:09:04 <ralonsoh> I talked to frickler yesterday 13:09:16 <ralonsoh> this is related to a patch submitted some years ago 13:09:19 <ralonsoh> (let me find it) 13:10:16 <ralonsoh> #link https://review.opendev.org/c/openstack/neutron/+/662409 13:10:37 <ralonsoh> so the point is that we have an attribute that belongs to the subnet 13:11:10 <ralonsoh> but we don't find what IP address is assigned to the port and the subnet that belongs to 13:11:56 <ralonsoh> so here https://review.opendev.org/c/openstack/neutron/+/662409/43/neutron/plugins/ml2/extensions/dns_integration.py#376 we maybe should make this filter 13:12:07 <ralonsoh> mlavalle1, ^ maybe you know better this code 13:12:27 <mlavalle1> I can take a look 13:12:37 <ralonsoh> much appreciated 13:13:04 <ralonsoh> apart from this bug, there was an unassigned one without any patch proposed 13:13:09 <ralonsoh> #link https://bugs.launchpad.net/neutron/+bug/2120723 13:13:17 <ralonsoh> [OVN] Metadata agent errors when returning gzip-encoded data 13:13:31 <mlavalle1> I'll assign that bug to me 13:13:38 <ralonsoh> mlavalle1, yes, thanks! 13:14:03 <mlavalle1> I meant the DNS one 13:14:06 <ralonsoh> yes 13:14:31 <ralonsoh> so LP#2120723 looks legit but I don't know how to replicate it locally 13:14:45 <ralonsoh> because I don't know how to build a metadata server that compress the data 13:15:09 <ralonsoh> in any case, I've requested some information and I can propose a patch to be tested in the reporter's env 13:15:19 <ralonsoh> and that's all I have 13:15:23 <ralonsoh> any other bug? 13:15:39 <mtomaska> there was also this bug. but I closed it https://bugs.launchpad.net/neutron/+bug/2120777 13:15:59 <mtomaska> and that is all 13:16:34 <ralonsoh> yeah, doesn't look like a bug 13:17:01 <ralonsoh> so this week bcafarel is the deputy 13:17:09 <ralonsoh> next week is elvira 13:17:10 <mtomaska> no he was week before me 13:17:19 <elvira> I'm the deputy this week I think 13:17:26 <ralonsoh> ahh ok, so you swapped 13:17:37 <mtomaska> yep 13:17:38 <ralonsoh> so elvira is the deputy this weej 13:17:43 <ralonsoh> and slaweq is the next one 13:17:52 <slaweq> sure 13:17:53 <elvira> sure! o> 13:17:58 <ralonsoh> perfect, thanks folks 13:18:18 <ralonsoh> next topic 13:18:22 <ralonsoh> #topic specs 13:18:27 <ralonsoh> #link https://review.opendev.org/q/project:openstack%252Fneutron-specs+status:open 13:18:44 <ralonsoh> I'll be honest, I didn't spend too much time reviewing them 13:19:03 <ralonsoh> but we'll need to postpone the pending ones to the next cycle at this point 13:19:21 <ralonsoh> in any case, all of them are still valid, so please spend some time reviewing them 13:19:27 <ralonsoh> any comment on this? 13:20:30 <ralonsoh> ok, let's move on 13:20:42 <ralonsoh> #topic community_goals 13:20:52 <ralonsoh> first one, Neutron client deprecation 13:21:10 <lajoskatona> for that I hope the last horizon patch: https://review.opendev.org/c/openstack/horizon/+/953302 13:21:26 <lajoskatona> some work is necessary as seems some tests are failing in CI 13:21:54 <lajoskatona> And for I also worked a little on removing n-client: https://review.opendev.org/c/openstack/nova/+/928022 13:22:20 <ralonsoh> the first one looks almost ready, just two failing UI tests 13:22:27 <lajoskatona> it is wip, and I have to cut it to smaller chunks somehow, as it ibecome uglier and uglier with time.... 13:22:43 <lajoskatona> yes I hope I can find where I broke them :-) 13:23:02 <lajoskatona> that's it for this topic from me 13:23:16 <ralonsoh> ufff yes, maybe you need to break the nova one in small pieces, if possible 13:23:51 <lajoskatona> yes that will be the way for that 13:24:16 <ralonsoh> the next topic is the eventlet removal 13:24:28 <ralonsoh> lajoskatona, has pushed the fwwas patches 13:24:32 <ralonsoh> #link https://review.opendev.org/q/project:openstack/neutron-fwaas+status:open 13:24:54 <ralonsoh> just one consideration: the FTs are completely broken when using ML2/OVS and non-eventlet 13:25:01 <lajoskatona> I just started to cut it to smaller pieaces also, but I have to go back to functional 13:25:07 <ralonsoh> no no 13:25:22 <ralonsoh> functional are broken without eventlet and ML2/OVS 13:25:50 <lajoskatona> ok, so we have to skip that for now and find out how to fix it ? 13:25:51 <ralonsoh> is impossible, right now, to execute a OVS agent related test with a os-ken instance 13:26:05 <ralonsoh> I need to document it, along with the patch I'm working on 13:26:17 <ralonsoh> #link https://review.opendev.org/c/openstack/neutron/+/952258 13:26:30 <ralonsoh> ^^ in this patch I've skipped all OVS agent related patches 13:26:44 <ralonsoh> NOTE: OVS agent, not ovsdb related ones 13:27:44 <ralonsoh> as commented, I'll document what is happening, but as a quick summary: there are many threads that now cannot be killed and are not stopped, the os-ken app register some signals (now is not possible if running outside the main thread) 13:28:00 <ralonsoh> and the os-ken socket communication is buggy in the FTs 13:28:08 <ralonsoh> so, a drama... 13:28:34 <ralonsoh> anyway, this patch won't be merged in Flamingos 13:28:34 <lajoskatona> yeah, really bad news.... 13:28:46 <ralonsoh> the problem will be, for functional tests, in Gazpacho 13:29:01 <ralonsoh> anything else you want to add? 13:29:22 <ralonsoh> ok, next topic 13:29:27 <ralonsoh> #topic on_demand 13:29:32 <ralonsoh> anything you want to add? 13:29:49 <mlavalle1> not from me 13:30:08 <lajoskatona> no more from my side 13:30:16 <ralonsoh> thank you folks for attending, have a nice week! 13:30:19 <ralonsoh> #endmeeting