14:00:37 <haleyb> #startmeeting networking 14:00:37 <opendevmeet> Meeting started Tue Feb 25 14:00:37 2025 UTC and is due to finish in 60 minutes. The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:37 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:37 <opendevmeet> The meeting name has been set to 'networking' 14:00:39 <haleyb> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, obondarev, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, haleyb, ralonsoh 14:00:50 <lajoskatona> o/ 14:00:52 <ralonsoh> hello 14:00:54 <slaweq> o/ 14:01:11 <obondarev> o/ 14:01:18 <bcafarel> o/ 14:01:22 <mtomaska> o/ 14:01:43 <cbuggy> o/ 14:02:01 <haleyb> #announcements 14:02:09 <haleyb> We are in week R-5 of Epoxy 14:02:24 <haleyb> E-3 milestone is this week 14:02:31 <haleyb> #link https://releases.openstack.org/epoxy/schedule.html 14:03:14 <haleyb> So that means feature freeze and final client library release 14:03:16 <rubasov> o/ 14:03:50 <haleyb> so i will go through neutron-lib changes and make sure anything merges this week 14:04:09 <haleyb> i'm not sure there is much 14:04:24 <ralonsoh> haleyb, https://review.opendev.org/c/openstack/releases/+/942529 14:04:28 <ralonsoh> I proposed this patch 14:04:37 <ralonsoh> so maybe we would like to stop it for now 14:05:34 <haleyb> so looking, there is only one change 14:05:39 <haleyb> #link https://review.opendev.org/c/openstack/neutron-lib/+/941895 14:06:41 <haleyb> so i guess we might as well update the release change 14:07:03 <haleyb> once it merges that is, pretty simple one 14:07:56 <haleyb> The nomination period for the OpenStack PTL+TC elections closed last week 14:08:09 <haleyb> so it looks like you are stuck with me for another cycle :) 14:08:26 <slaweq> Congrats haleyb 14:08:28 <mtomaska> \o/ 14:08:37 <slaweq> And thank you 14:08:49 <ralonsoh> congrats 14:09:10 <lajoskatona> \o/ 14:09:11 <haleyb> and from the ML i don't think there will be voting for TC as the number of nominees matched open spots 14:09:30 <slaweq> Nope, no voting at all 14:09:55 <slaweq> We should conclude everything officially this week as voting is not needed 14:10:45 <haleyb> there is a new Nova PTL which is maybe the one change i remember 14:11:04 <haleyb> either way, will wait for the official results 14:11:10 <slaweq> There is also new Designate PTL 14:11:23 <slaweq> and new old PTL of Octavia :) 14:11:43 <slaweq> from the projects which are kind of close to Neutron 14:12:30 <haleyb> what's old is new again i guess 14:12:36 <haleyb> thanks for the info 14:13:29 <haleyb> i had one more announcement 14:14:02 <haleyb> Some of you might know, but Ihar has decided to step-down from being a core as he has a change in responsibilities 14:14:34 <haleyb> he let me know officially the other day 14:15:06 <haleyb> i wish him good luck, and who knows, maybe he will return (again) 14:15:40 <ralonsoh> best luck for Ihar 14:16:27 <haleyb> Amotoki has also not been active, and i was finally able to get a response from him that he should be removed from the core group as well 14:16:52 <lajoskatona> :-( 14:17:08 <slaweq> that's sad news :( 14:17:13 <ralonsoh> for sure... 14:17:22 <haleyb> he is doing ok so that is good 14:18:33 <haleyb> this will cause a one week shift in the bug deputy, i will update after this meeting 14:18:52 <lajoskatona> thanks for the news and for checking 14:19:12 <haleyb> that was all announcements i had, any others 14:19:19 <ralonsoh> 2 quick ones 14:19:25 <ralonsoh> 1) https://ptg2024.openinfra.dev/ is open now, please register 14:19:26 * haleyb will brb, dog barking, but con tinue 14:19:39 <ralonsoh> 2) haleyb can you create the 2025.2 etherpad? 14:19:41 <ralonsoh> that's all 14:20:06 <ralonsoh> (btw, the address 2024 is correct) 14:23:08 <haleyb> back 14:23:24 <haleyb> yes, i can create the etherpad, i did sign up neutron for the ptg 14:24:06 <haleyb> #topic bugs 14:24:15 <haleyb> slaweq was the deputy last week 14:24:19 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/2KVZ5ZVKMXLMO3USE3D2NCZRS46BFUYW/ 14:24:32 <haleyb> there were a lot of bugs 14:25:05 <slaweq> there were couple indeed :) 14:25:24 <slaweq> and I would like to ask someone to take a look and help with triage of the https://bugs.launchpad.net/neutron/+bug/2099675 14:26:50 <ralonsoh> I think there is something missing in this description 14:28:23 <ralonsoh> ok, please correct me if I'm wrong: is he using trunk ports with external ports? 14:28:32 <ralonsoh> I don't think we have ever tested that 14:28:57 <haleyb> that's what it looks like i guess 14:29:16 <ralonsoh> ok, I'll make some questions to OVN folks about this 14:29:25 <slaweq> thx ralonsoh 14:29:32 <ralonsoh> I'll take it, at least to know if that is supported by OVN 14:30:08 <haleyb> thanks 14:30:18 <slaweq> other than that, there are also some bugs reported for os-ken 14:30:28 <slaweq> but they are related to the CLI part of os-ken 14:30:40 <slaweq> I didn't even know that there is something like that 14:30:48 <ralonsoh> first one is solved, at least I pushed a patch 14:30:57 <haleyb> #link https://review.opendev.org/c/openstack/os-ken/+/942566 14:31:09 <slaweq> and the question is: should we keep it and try to fix those bugs, or maybe remove that CLI part from os-ken? 14:32:10 <ralonsoh> for us (neutron) we care only about the library part, not the CLI 14:32:18 <ralonsoh> that was imported from ryu initially 14:32:44 <ralonsoh> I'll check if we can solve easily https://bugs.launchpad.net/neutron/+bug/2099799 14:32:56 <slaweq> ralonsoh exactly, that's why I am asking if we should drop this CLI part as we don't use it really in neutron 14:32:59 <ralonsoh> but I don't think we should support it 14:33:06 <lajoskatona> +1 14:33:15 <slaweq> IMHO we should make os-ken as small as possible 14:33:32 <slaweq> we did it as a fork of ryu but we don't need everything what ryu provides really 14:33:46 <ralonsoh> exactly 14:34:26 <slaweq> I can open another bug in LP to track this and clean os-ken from the CLI part 14:34:26 <haleyb> i would agree, i also did not know there was a cli to it 14:34:32 <ralonsoh> slaweq, +1 14:34:35 <slaweq> if everyone is ok with it 14:34:43 <haleyb> i had a question on the other bug though, is it only related to cli? 14:34:47 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2099799 14:34:50 <ralonsoh> haleyb, me neither (and I wrote that in the first bug) 14:35:07 <ralonsoh> haleyb, I think this code is only used by the CLI 14:36:01 <haleyb> ack 14:36:45 <haleyb> i just pushed your other change into the gate, we can propose a new release once it merges 14:38:35 <haleyb> ralonsoh: did you want to discuss your rfe's on friday? 14:38:49 <ralonsoh> not really, no rush for now 14:39:00 <haleyb> ok 14:39:05 <ralonsoh> so if there is nothing else, let's cancel the meeting 14:39:20 <haleyb> there was one more that seemed simple, i just haven't thought it through yet 14:39:27 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2099818 14:40:19 <ralonsoh> we can discuss this RFE next friday 14:40:25 <ralonsoh> but what kills me is "where a shared OVN deployment is leveraged outside of OpenStack" 14:41:55 <haleyb> i think i've seen a few, but not very common (yet) 14:41:57 <slaweq> ralonsoh I agree, this may cause different issues also, like e.g. with conntrack, if I'm not wrong 14:42:27 <ralonsoh> let's take this conversation for the drivers meeting with the author 14:42:36 <haleyb> yes, will put on next weeks agenda 14:43:05 <haleyb> that was all the bugs i had 14:43:20 <haleyb> i am the deputy this week, rubasov is next week 14:43:25 <haleyb> is that ok for you rubasov ? 14:43:29 <rubasov> of course 14:43:37 <haleyb> great, thanks 14:43:59 <haleyb> #topic community goals 14:44:45 <haleyb> lajoskatona: i did not see any neutronclient patches merge, just more +1s i think 14:44:50 <haleyb> #link https://review.opendev.org/q/topic:%22bug/1999774%22+status:open 14:45:16 <lajoskatona> haleyb: yeah, no progress this week 14:45:18 <haleyb> hopefully we can get the ones in-flight merged this cycle 14:45:49 <haleyb> ralonsoh: and that leaves eventlet 14:45:54 <ralonsoh> yes 14:45:58 <ralonsoh> we have 3 fronts 14:46:04 <ralonsoh> 1) sahid's patches 14:46:06 <ralonsoh> https://review.opendev.org/q/owner:sahid.ferdjaoui@industrialdiscipline.com+status:open+project:openstack/neutron 14:46:20 <ralonsoh> (I'll continue checking them tomorrow) 14:46:25 <ralonsoh> please spend some time on them 14:46:32 <ralonsoh> 2) DHCP 14:46:36 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/942393 14:46:40 <ralonsoh> and the upper patch 14:46:50 <ralonsoh> these are removing eventlet from DHCP agent 14:47:06 <ralonsoh> 3) https://bugs.launchpad.net/neutron/+bug/2099896 14:47:18 <ralonsoh> I forgot this one when I created the eventlet-removal bugs 14:47:27 <ralonsoh> I'm working on this, tomorrow I'll send a first patch 14:47:29 <ralonsoh> that's all 14:47:47 <haleyb> ack, was going to ask if you needed help with this last one 14:48:16 <ralonsoh> I'm going to re-use the ovn metadata agent implementation (that is also use in the ovn agent) 14:48:24 <ralonsoh> so that should be faster (and should work) 14:48:31 <haleyb> right 14:48:50 <haleyb> did you also push an update for the l3-agent? i can't keep track 14:49:01 <ralonsoh> I have one pending 14:49:08 <ralonsoh> because of the comments I received 14:49:15 <ralonsoh> I'm not going to remove the concurrency 14:49:25 <ralonsoh> I have a patch in progress locally to remove eventlet 14:49:28 <ralonsoh> one sec 14:49:30 <haleyb> understood 14:49:46 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/938411 14:49:49 <ralonsoh> (not working yet) 14:49:52 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/938541 14:49:59 <ralonsoh> this one is ready for review 14:50:30 <haleyb> i will look after meetings 14:50:40 <ralonsoh> thanks! 14:51:04 <haleyb> moving on 14:51:07 <haleyb> #topic on-demand 14:51:14 <haleyb> any other topics to discuss? 14:51:35 <ralonsoh> no thanks 14:51:55 <lajoskatona> nothing from me 14:52:29 <haleyb> ok, if nothing else will end 14:52:32 <haleyb> have a good week everyone! 14:52:36 <haleyb> #endmeeting