14:00:22 <haleyb> #startmeeting networking 14:00:22 <opendevmeet> Meeting started Tue Jan 23 14:00:22 2024 UTC and is due to finish in 60 minutes. The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:22 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:22 <opendevmeet> The meeting name has been set to 'networking' 14:00:32 <haleyb> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, obondarev, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, amotoki 14:00:33 <mlavalle> \o 14:00:34 <frickler> o/ 14:00:45 <lajoskatona> o/ 14:00:47 <elvira> o/ 14:00:51 <obondarev> o/ 14:00:52 <slaweq> o/ 14:01:26 <ralonsoh> hi 14:01:30 <haleyb> wow, the gang is all here :) 14:01:47 <haleyb> #topic announcements 14:02:00 <ykarel> o/ 14:02:05 <haleyb> #link https://releases.openstack.org/caracal/schedule.html 14:02:10 <ihrachys_> o/ 14:02:46 <haleyb> We are in C-3, and that milestone is the week of Feb 26th, so one month 14:03:19 <haleyb> Also, the final release for non-client libraries is the week of Feb 19th 14:04:26 <haleyb> so just a reminder on code reviews :) 14:04:41 <lajoskatona> and ps-owners :-) 14:05:11 <bcafarel> late o/ 14:05:28 <haleyb> lajoskatona: yup 14:06:09 <haleyb> A name was chosen for D/2024.2 - Dalmation, and a proposed schedule is out 14:06:14 <haleyb> #link https://review.opendev.org/c/openstack/releases/+/906050 14:06:43 <haleyb> "2 October, 2024 as planned release date" 14:07:08 <haleyb> so if you were waiting for a dog-themed release to merge your code you now have a chance 14:07:38 <opendevreview> Merged openstack/neutron-tempest-plugin master: Fix network sorting in API tests https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/905853 14:07:39 * haleyb kids would have groaned at that joke 14:08:19 <lajoskatona> :-) 14:08:33 <lajoskatona> +1 for puppies 14:09:00 <lajoskatona> (we have cats, so not compatible, but perhaps there will be a cat release in the future) 14:09:09 <haleyb> regarding other branches, Ussuri is EOL, and all patches have been abandoned, X/W/V will eventually transition to unmaintained 14:09:42 <haleyb> lajoskatona: some dogs and cats are compatible, just not mine 14:10:09 <haleyb> Any other announcements? 14:10:13 <ralonsoh> PTG registration is open (but the link is incorrect): https://openinfra.dev/ptg/ 14:10:29 <ralonsoh> the button "register now" points to the survey, not the registrarion 14:11:51 <haleyb> ralonsoh: thanks, i will make sure to sign up neutron for rooms early this time 14:11:57 <slaweq> I think that link is correct and it's for leaders of the teams to sign teams up 14:12:33 <ralonsoh> but how can we register into the PTG then? 14:13:00 <slaweq> maybe that registration for users will be available later? I don't really know 14:13:17 <slaweq> now it says "Sign up your team to participate by February 18" 14:13:30 <slaweq> so I guess that after Feb 18 it will let users to register 14:13:43 <mlavalle> I think so 14:13:46 <haleyb> i will use that to sign-up Neutron after the meeting 14:14:00 <rubasov> late o/ 14:14:51 <ralonsoh> ok (but then it is weird that this survey is public) 14:14:58 <ralonsoh> anyway, I'll wait for that 14:15:36 <haleyb> ralonsoh: seems they trust everyone to not fill-out fake forms 14:16:23 <haleyb> ok, lets move on 14:16:26 <haleyb> #topic bugs 14:16:56 <haleyb> i was bug deputy last week, and of course, there were a lot of bugs (9) 14:17:05 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/BJUXN6KNZPIOMHP4HMENB766HOV36HIU/ 14:17:42 <mlavalle> lol 14:17:59 <haleyb> there are a few that do not have owners 14:18:00 <lajoskatona> :-) 14:18:25 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2049524 14:18:41 <haleyb> actually, looks like ralonsoh picked this up 14:18:41 <ralonsoh> this one is invalid 14:18:42 <haleyb> [Neutron/Nova] Need to fix attaching a vip port on VMs 14:18:52 <ralonsoh> is is working as expected with the patch in place 14:19:17 <ralonsoh> --> https://paste.opendev.org/show/bDMjxm7tePTpmwnAIijz/ (what is expected) 14:19:19 <haleyb> ralonsoh: i trust your testing was better than mine, i could have sworn it failed locally 14:20:14 <haleyb> we can wait for submitter to verify and/or close if no response 14:20:36 <haleyb> next one 14:20:41 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2049624 14:20:47 <haleyb> vpnaas problem --- strongswan on rocky 9 14:21:24 <haleyb> is there someone more closely working on vpnaas that can take this? 14:22:21 <lajoskatona> Christian Rohman commented on the bug, and there's an old patch he mentioned 14:22:23 <haleyb> noonedeadpunk ^^ ? 14:22:29 <lajoskatona> https://bugs.launchpad.net/neutron/+bug/2049624/comments/8 14:23:21 <haleyb> lajoskatona: ack, would just be good to have a name in "Assigned to" so it doesn't fall through the cracks 14:24:02 <lajoskatona> I try to check the guys, as the original issue is from Ericsson 14:24:10 <haleyb> lajoskatona: thanks 14:25:00 <haleyb> next one 14:25:11 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2049909 14:25:12 <haleyb> MLDv2 packets sent from L3 Agent managed networks cause backup routers to be preferred 14:26:01 <haleyb> This looks like a HA router issue we've seen before, but reported on linuxbridge 14:26:29 <haleyb> #link https://review.opendev.org/c/openstack/neutron/+/906114 14:26:53 <haleyb> i'm not sure i like using iptables to block mldv2 though, think we had tried that before 14:27:41 <ralonsoh> no, we don't link UP the standby interfaces 14:27:43 <ralonsoh> since c52029c39aa824a67095fbbf9e59eff769d92587 14:27:44 <ihrachys_> you mean iptables drop-all rule still leaked? 14:28:02 <slaweq> IIRC we had issues with MLDv2 packets send from HA routers in the past 14:28:22 <ralonsoh> this is the patch: https://review.opendev.org/c/openstack/neutron/+/707406 14:28:23 <haleyb> ihrachys_: no, i'm just not sure it's the best way to fix it 14:28:23 <slaweq> and due to that we were setting interfaces to DOWN in the backup routers 14:28:26 <ralonsoh> (and a partial revert) 14:28:33 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/834162 14:28:34 <andrewbonney> We see standby interfaces up all the time, and this is on 2023.1 14:28:48 <slaweq> maybe it was changed later again 14:29:15 <ralonsoh> btw, checking the related changes: https://review.opendev.org/c/openstack/neutron/+/839671/8 14:29:20 <ralonsoh> ^^ [L3-HA] Disable automatic link-local address assignment for HA routers 14:29:27 <ralonsoh> so we should not have this problem 14:30:16 <slaweq> thx ralonsoh :) 14:30:32 <andrewbonney> At least one of the MLDv2 packets I saw from the standby MAC was from the address '::' fwiw. I'd have to re-check if there were other link local ones or not 14:31:03 <ralonsoh> please check to what interface to mldv2 packet is going 14:31:22 <ralonsoh> because the HA interfaces should not have any ipc6 address and should not present this issue 14:31:46 <ralonsoh> andrewbonney, qq, related to the bug: why are you using linux bridge? 14:31:52 <andrewbonney> I'll double check as soon as possible, but I don't think the backup interface had any address, but the packets were still sent 14:32:00 <ralonsoh> any specific requirement? 14:32:14 <andrewbonney> Legacy reasons, we have done historically and it's a big effort to switch 14:32:23 <ralonsoh> ok (we can discuss this offline) 14:32:28 <andrewbonney> :) 14:33:37 <haleyb> i will skip the next bug on setting dns-name, since two of us suspect it was user-error in not enabling the correct dns extensions 14:33:47 <haleyb> any other bugs people want to talk about? 14:34:00 <ralonsoh> no thanks 14:34:10 <haleyb> This week amotoki is the bug deputy, next week will be mlavalle 14:35:06 <haleyb> Current bug count this week was: 760, up 2 from last week, so not much movement even with 9 new ones which is good 14:35:50 <haleyb> #topic specs 14:35:57 <haleyb> #link https://review.opendev.org/q/project:openstack%252Fneutron-specs+status:open 14:36:25 <haleyb> i was able to review the OVN-IC one yesterday, seems pretty close 14:37:50 <haleyb> i don't see racosta here, hopefully he will update this week 14:38:22 <haleyb> i will look at the other ones this week 14:39:02 <haleyb> #topic community_goals 14:39:04 <haleyb> #link https://review.opendev.org/c/openstack/horizon/+/891205 14:40:00 <haleyb> lajoskatona: that one test keeps failing :( 14:40:06 <lajoskatona> last week it was topic on the horizon weekly meeting, and horizon team checks the failing integration test 14:40:51 <lajoskatona> what makes it harder (for me at least) is that the team considers this test suite instable and there is a migration ongoing 14:41:56 <lajoskatona> and that's it for this topic 14:42:03 <haleyb> lajoskatona: thanks 14:42:16 <haleyb> #topic on_demand 14:42:26 <haleyb> ralonsoh: you had added an item 14:42:40 <ralonsoh> let me check 14:42:54 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2035332 14:42:56 <ralonsoh> ah yes https://bugs.launchpad.net/neutron/+bug/2035332 14:43:39 <ralonsoh> I didn't check the replies 14:44:06 <ralonsoh> ykarel foudn that this could be an issue with the MTUs beyween networks 14:44:35 <ralonsoh> (we don't handle the MTU fragmentation in the N->S direction) 14:45:01 <haleyb> right, we have another bug on that topic and added to gaps doc 14:45:10 <ralonsoh> in any case, I added this topic because the reporter asked me 14:45:37 <ralonsoh> if they can't reply to ykarel question, then we can consider this bug as incomplete for now 14:46:10 <ralonsoh> (that's all from my side) 14:46:26 <haleyb> ok, thanks 14:46:50 <haleyb> ykarel: CI meeting in :14, and video today? 14:47:00 <ykarel> yeap right 14:47:41 <haleyb> ok, if nothing else can end early 14:47:54 <haleyb> have a good week and keep reviewing! :) 14:47:59 <haleyb> #endmeeting