14:00:22 <haleyb> #startmeeting networking 14:00:23 <opendevmeet> Meeting started Tue Nov 26 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. 14:00:23 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:23 <opendevmeet> The meeting name has been set to 'networking' 14:00:26 <haleyb> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, obondarev, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, amotoki, haleyb, ralonsoh 14:00:29 <ralonsoh> hello 14:00:34 <obondarev> o/ 14:00:37 <ykarel> o/ 14:00:39 <lajoskatona> o/ 14:00:42 <s3rj1k> hi all 14:00:44 <jlibosva> o/ 14:00:46 <slaweq> o/ 14:00:47 <ralonsoh> (I was going to ask if we were going to have this meeting today) 14:01:05 <ralonsoh> we don't have the drivers meeting this friday, right? 14:01:12 <bcafarel> o/ 14:01:30 <haleyb> ralonsoh: i will not be here friday, so i was going to cancel the drivers meeting 14:01:36 <ralonsoh> perfect for me 14:02:00 <haleyb> this is actually my last work day of the week 14:02:31 <haleyb> we have this big thing with turkeys, etc 14:02:51 <ralonsoh> happy thanksgiving day 14:03:01 <haleyb> that's the one :) 14:03:07 <slaweq> happy Thanksgiving :) 14:03:23 <slaweq> and good luck in Black Friday too :) 14:03:27 <ralonsoh> hahahaha 14:04:14 <haleyb> i have already bought my gift, you will not find me at the stores on Friday, craziness 14:04:39 <lajoskatona> :-) 14:04:45 <haleyb> i guess we should get started... 14:04:50 <haleyb> #announcements 14:05:04 <haleyb> #link https://releases.openstack.org/epoxy/schedule.html 14:05:10 <haleyb> We are currently in week R-18 14:05:15 <haleyb> Epoxy-2 milestone R-12 (week of Jan 6th) 14:06:07 <haleyb> Thanksgiving holiday in US this week, so as I said ^^ I will cancel the drivers meeting this week 14:07:00 <haleyb> and i will be out the rest of the week 14:07:59 <haleyb> Let's continue to use the priorities dashboard for patches in the "ready to merge" state (link in wiki), still seems to be working well 14:08:00 <lajoskatona> have a good rest (without stores) 14:08:35 <haleyb> lajoskatona: thanks, my plan is to not spend time at a keyboard 14:09:28 <haleyb> any other announcements? that's all i had 14:10:00 <haleyb> #topic bugs 14:10:11 <haleyb> i was the deputy last week, my report is at 14:10:18 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/W4BA5XBNHEUMTKLH4FTY3KTOZRKEJUC5/ 14:11:13 <haleyb> All the neutron bugs have owners and patches 14:11:45 <haleyb> there were two vpnaas bugs, the submitter was going to take a look at one 14:12:06 <haleyb> i did see some discussions on IRC regarding the other one 14:12:14 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2089250 14:13:02 <haleyb> failover not working with OVN driver 14:13:22 <ralonsoh> I think I was talking to the reporter last week 14:13:37 <lajoskatona> for vpnaas bugs, I sent one older to Bodo (bpeterman I think) and he promised to check it ,so I hope we are saved :-) 14:13:40 <ralonsoh> I think there is a problem on how the agent is retrieved, because OVN is patching this method 14:14:36 <ralonsoh> in any case, this is a matter of checking how a dragent is registering itself in the Neutron API 14:14:38 <haleyb> ralonsoh: can you just add a note to the bug with anything that might help, i'm not expecting you to fix it :) 14:14:45 <ralonsoh> sure 14:15:02 <haleyb> thanks 14:15:48 <haleyb> since everything is done or in progress with those, are there any other bugs people want to discuss? 14:16:18 <haleyb> ok 14:16:23 <haleyb> This week rubasov is the bug deputy, next week will be mlavalle 14:16:37 <haleyb> is that still ok with both of you? 14:17:06 <rubasov> yes 14:18:08 <haleyb> mlavalle might be out chasing his turkey 14:18:39 <haleyb> alright, moving on 14:18:46 <haleyb> #topic community goals 14:19:26 <ralonsoh> should I start? 14:19:36 <ralonsoh> about eventlet deprecation, nothing new this week 14:19:51 <haleyb> sure, i was typing slowly 14:19:57 <ralonsoh> there is a forum/team to study the ASGI migration (or uWSGI migration) 14:20:03 <ralonsoh> but nothing defined yet 14:20:23 <ralonsoh> and I'll start the migration of the tcp server for the metadata 14:20:28 <ralonsoh> that is based on eventlet 14:20:32 <ralonsoh> that's all 14:21:12 <haleyb> ralonsoh: i wanted to ask if we should encourage people to pick-up a task/bug you identified 14:21:26 <ralonsoh> for sure, I'll take the metadata agent now 14:21:36 <ralonsoh> but anyone is more than welcome to take any of them 14:21:44 <ralonsoh> sriov agent, ovs agent, etc 14:21:58 <ralonsoh> and be aware of the oslo.services patch 14:22:14 <ralonsoh> that is removing the eventlet stuff 14:22:22 <haleyb> #link https://bugs.launchpad.net/neutron/+bugs?field.tag=eventlet-deprecation 14:23:17 <haleyb> ralonsoh: do you have a link to the oslo.services patch? 14:23:35 <ralonsoh> this is still in the oven 14:23:41 <ralonsoh> that was just a heads-up 14:23:52 <haleyb> ah, ok 14:24:37 <haleyb> thanks for the update 14:25:13 <haleyb> the other community goal is deprecating the neutron client 14:25:25 <haleyb> lajoskatona has updated his Horizon patch for routers 14:25:31 <haleyb> #link https://review.opendev.org/c/openstack/horizon/+/934457 14:25:47 <haleyb> thanks for keeping that work going 14:25:47 <lajoskatona> For the SDK thing: https://review.opendev.org/q/topic:%22bug/1999774%22+status:open the router API patch is ready, the other one I have to still check the tests 14:26:36 <haleyb> great, thanks, will review the routers one today 14:27:13 <haleyb> that was all for community goals 14:27:18 <opendevreview> yatin proposed openstack/neutron-lib master: Switch to neutron-functional job https://review.opendev.org/c/openstack/neutron-lib/+/936265 14:27:23 <haleyb> #topic on-demand 14:27:33 <haleyb> there was one item i left for carry-over 14:28:11 <haleyb> had to do with the migration of jobs (or a job) to noble 14:28:31 <haleyb> but we are still blocked on tempest patch 14:28:36 <ralonsoh> there are several patches proposed 14:28:43 <ralonsoh> yes, but the Neutron one is now working 14:28:49 <haleyb> #link https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/933277 14:28:54 <haleyb> #link https://review.opendev.org/c/openstack/tempest/+/932156 14:28:55 <ralonsoh> yes 14:29:09 <haleyb> #link https://review.opendev.org/c/openstack/neutron/+/932920 14:29:25 <ralonsoh> yes, this one too 14:29:45 <ralonsoh> I think ykarel was checking the issues with pip install 14:29:50 <haleyb> do we still need to disable the mysql performance setting for devstack? 14:29:55 <ralonsoh> related to functional and fullstack 14:30:07 <haleyb> ah, i see it in the patch now 14:30:18 <ralonsoh> https://review.opendev.org/c/openstack/neutron/+/936022 14:30:31 <haleyb> my guess is on Noble we will need to use pipx 14:30:43 <ykarel> no not needed to disable, i am doing other tests in https://review.opendev.org/c/openstack/neutron/+/935847 14:31:02 <ykarel> fullstack have some issues related to systemd, so checking those 14:31:10 <ykarel> functional one passed there 14:31:39 <ralonsoh> isn't a way to pip install using system packages? 14:31:45 <ralonsoh> well, inside a venv... 14:31:58 <ralonsoh> that's the main problem 14:32:44 <ykarel> you mean pip install outside venv? within venv it works 14:33:04 <frickler> "python -m venv --system-site-packages" creates a venv that uses system pkgs 14:34:13 <haleyb> we (i) use pipx when I'm on Noble, which usually works 14:34:22 <haleyb> i'm assuming there is some devstack piece to all of this? 14:34:30 <ralonsoh> ok, no, the error was different in the FT job 14:34:47 <ralonsoh> frickler, yes, we use it in our CI 14:36:39 <haleyb> alright, at least we are trying to move this forward 14:36:45 <ykarel> so devstack jobs use global venv(with --system-site-packaegs) so those are not impacted, functional/fullstack were not relying on that so /me trying that in above patch 14:37:09 <ykarel> newer pip don't allow by default to install modules system wide 14:37:36 <ykarel> though there is option to do overrite that with --break-system-packages 14:37:55 <ykarel> and that should be avoided in normal scenarios 14:38:22 <haleyb> and that's why we recommend pipx :) but i'm a broken record now 14:39:08 <frickler> haleyb: "we" = ubuntu/canonical? 14:39:25 <opendevreview> Merged openstack/neutron master: Bump psutil to 6.1.0 https://review.opendev.org/c/openstack/neutron/+/936144 14:39:50 <haleyb> frickler: well, at least me - anything our team does tries to use pipx 14:40:13 <frickler> would it make sense for devstack to look into switching to that instead of the global venv it currently uses? 14:41:29 <frickler> (sorry if that's a bit offtopic, we can discuss after the meeting, too) 14:41:40 <haleyb> they could try, i could try to see if there is some source of truth here 14:42:01 <ralonsoh> ykarel, we can sync after the meeting about what packages are broken in this installation 14:42:07 <frickler> unrelated just a note that I merged the long pending netaddr bump earlier today. testing looked fine but there might be unexpected issues still https://review.opendev.org/c/openstack/requirements/+/909300 14:42:16 <ykarel> ralonsoh, ack 14:42:38 <haleyb> alright, were there any other topics? 14:42:45 <ralonsoh> frickler, ack, from 0.10.1 to 1.3.0... a long jump 14:43:01 <frickler> yes, it was a long time in the making, too ;) 14:43:25 <lajoskatona> frickler: cool, thanks for the update on that 14:44:09 <haleyb> everyone have a good week! 14:44:10 <haleyb> #endmeeting