Tuesday, 2026-04-07

opendevreviewElvira García Ruiz proposed openstack/neutron master: Add PVLAN service plugin  https://review.opendev.org/c/openstack/neutron/+/97879807:44
opendevreviewRodolfo Alonso proposed openstack/neutron-specs master: Add spec for L3 Agent Scheduler API support in ML2/OVN  https://review.opendev.org/c/openstack/neutron-specs/+/98274307:49
opendevreviewRodolfo Alonso proposed openstack/neutron master: WIP == Add ``l3-agent-scheduler-ha-priority`` extension for ML2/OVN  https://review.opendev.org/c/openstack/neutron/+/98279208:08
opendevreviewRodolfo Alonso proposed openstack/neutron master: [OVN] Update the parent hostname in the virtual ports  https://review.opendev.org/c/openstack/neutron/+/98154009:17
opendevreviewElvira García Ruiz proposed openstack/neutron master: Add PVLAN service plugin  https://review.opendev.org/c/openstack/neutron/+/97879809:23
stephenfinralonsoh: regarding the mypy issues you're seeing, some of those are due to a new mypy but a lot of them are due to (a) typed oslotest and (b) typed testtools, both of which came in via a recent u-c10:18
stephenfinI had a version of https://review.opendev.org/c/openstack/openstacksdk/+/983429/ locally but I'm waiting on the testtools 2.9.0 bump to get into u-c. I've proposed that here https://review.opendev.org/c/openstack/requirements/+/98352410:18
stephenfinthat will unblock https://review.opendev.org/c/openstack/oslotest/+/983353 which you've also seen10:19
ralonsohstephenfin, thanks! if you want, I can abandon this patch10:37
stephenfinyou mean https://review.opendev.org/c/openstack/openstacksdk/+/983429? Probably best10:39
opendevreviewRodolfo Alonso proposed openstack/neutron-lib master: Revert "Add VIF_TYPE_VIRTUAL constant to portbindings"  https://review.opendev.org/c/openstack/neutron-lib/+/98305811:12
haleyb#startmeeting networking13:00
opendevmeetMeeting started Tue Apr  7 13:00:14 2026 UTC and is due to finish in 60 minutes.  The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot.13:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.13:00
opendevmeetThe meeting name has been set to 'networking'13:00
haleybPing list: bcafarel, elvira, frickler, mlavalle, mtomaska, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, haleyb, ralonsoh13:00
mlavalle\o13:00
lajoskatonao/13:00
haleybmaybe others will show up, but i'll get started13:01
haleyb#topic announcements13:01
ralonsohhi13:02
haleybGazpacho is done and shipped :)13:02
haleybthanks for all the work on it13:02
ralonsohcongrats!13:02
lajoskatonayeah :-)13:02
rubasovo/13:03
haleybthe PTG is in a couple of weeks13:03
haleybHibiscus PTG April 20-24, 202613:03
haleyb#link https://ptg.openinfra.org/13:03
haleyb#link https://etherpad.opendev.org/p/apr2026-ptg-neutron13:04
haleybthanks for all the topics so far, please keep adding if you think of anything13:04
haleybi will only schedule meeting times for Tues/Wed/Thurs from 13:00-16:0013:05
elvirao/13:06
haleybthe current schedule might show more i'll remove when i get to making the schedule13:06
haleybalso, i sent a ML reminder that we will not have any drivers meeting this or next week, can just add to PTG topics13:06
haleybJust a reminder to use the priority dashboard for anything neutron/networking related13:07
haleyb#link https://tinyurl.com/59z278km13:07
haleybRP +1 for "ready to merge" changes, RP +2 for gate blockers and similar13:08
haleybwe did have a gate blocker yesterday, thanks everyone for helping fix13:08
haleybi added a related topic regarding it to the PTG list as part was due to eventlet deprecation leftovers13:08
haleybwell, the underlying was a u-c bump13:09
haleybthat was all i had, any other announcements?13:09
haleybok, moving on13:10
haleyb#topic bugs13:10
haleyblajoskatona was the bug deputy last week, his report is at13:10
haleyb#link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/THWLXHOME4ZIA3YQXDN7ASEVRIYM7C7R/13:10
lajoskatonaI hope it is correct, as launchpad was in terrible state this morning (timeouts mostly)13:11
lajoskatonathere was some unassigned13:12
haleybif i only knew a guy that helped with that infra :-o13:12
haleybfirst unassigned13:12
haleyb#link https://bugs.launchpad.net/neutron/+bug/214687013:12
lajoskatonaone is for s-g creation performance: https://bugs.launchpad.net/neutron/+bug/2146870 ( [OVN] Performance13:12
lajoskatonadegradation in Security Group Rules creation when network logging is enabled)13:12
haleyboh it's assigned now13:12
ralonsohI'm checking this one13:12
haleyb[OVN] Performance degradation in Security Group Rules creation when network logging is enabled13:13
lajoskatonaoh, cool13:13
ralonsohbtw, the issue is legit13:13
lajoskatonathanks for checking13:13
haleybso it is on the neutron side just creating the ACLs or on OVN side?13:14
ralonsohis how we iterate over the ACLs13:14
ralonsoheach rule will fire a AFTER_CREATE event13:14
ralonsohand each event calls `_set_acls_log`, that iterated and re-updates the ACLs13:15
ralonsohthat's unnecessary 13:15
haleybah, ok, thanks for looking and taking the bug13:16
lajoskatonaOk, the next unassigned is an RFE:13:16
lajoskatonahttps://bugs.launchpad.net/neutron/+bug/2147189 ([RFE] Support multiple non-contiguous ports in a single security group rule for ML2/OVN backend )13:16
lajoskatonaShall we have an RFE/spec session during the PTG? or wait for the coming drivers meeting with these?13:17
ralonsohwe need at least an owner first13:17
ralonsohwill Kyuyeong take it?13:18
lajoskatonayeah :-) I hope the reporter will show up and that will be solved13:18
ralonsohanother qq, do we have the same for ML2/OVS?13:18
ralonsohbecause we have a common API for all backends13:18
lajoskatonaI hope at least, and not just a random reporter who waits for somebody to fix it13:18
haleybi think he proposed a patch for his other bug/RFE13:19
haleybat a high level, does this at least seem useful to others?13:19
lajoskatonaralonsoh: true, that could be common change for all backends (ovs/ovn currently)13:19
slaweqralonsoh yeah, I don't want to do it for one backend only13:19
slaweqif we are going to change API in that way, it should probably be supported by other backends too13:20
slaweqbut we can discuss that during PTG I guess13:20
ralonsohfor sure, the best place for this discussion13:20
lajoskatona+1, I add these to the etherpad to be listed there13:20
lajoskatonaAnd there's one more unassigned: https://bugs.launchpad.net/neutron/+bug/2147305 (neutron cannot handle forks due to ovsdbapp)13:21
haleybsure, will add to the list13:21
ralonsohabout the ovsdbapp one, this seems to me more a RFE13:22
lajoskatonaralonsoh already commented on it and I have a feeling that perhaps this one is more an RFE13:22
lajoskatonaralonsoh: yeah, thanks13:22
opendevreviewBence Romsics proposed openstack/neutron-lib master: Reject multicast addresses in AAP: n-lib part  https://review.opendev.org/c/openstack/neutron-lib/+/98320313:22
ralonsohI would like to see some numbers: that optimization really provides any performance?13:23
ralonsohor is just for error catching, as commented in the bug?13:23
lajoskatonagood questions, but true some numbers for such change are helpful13:24
haleybagreed would be good to see impact13:24
lajoskatonaI think that's it for the unassigned bugs, all other bugs have owner13:25
haleybgreat, thanks for the update lajoskatona 13:25
haleybany other bugs we should discuss?13:26
haleybi see ralonsoh is the deputy this week, ichen is next week13:26
haleybis that still ok for both?13:26
haleybi realize i need to update the wiki13:26
ralonsohok13:26
ichenYes, works for me.13:27
haleybgreat, thanks!13:27
haleyb#topic specs13:27
haleyb#link https://review.opendev.org/q/project:openstack/neutron-specs+status:open13:28
ichenIs this a good time to bring up https://review.opendev.org/c/openstack/neutron-specs/+/982256 ?13:28
ralonsohthe best one13:28
ralonsohI'll check this spec tomorrow morning, word13:28
ichenI submitted a revision.  Please review.13:29
ralonsohI've updated mine too: https://review.opendev.org/c/openstack/neutron-specs/+/98274313:29
mlavalleI'll look at it this week also13:29
ralonsohI've addressed jakub's comments: the API should never reorder the chassis priorities13:29
ralonsohthat makes sense, the user must do this (this is why we are providing an API)13:30
haleybi'll try to review both again this week13:30
haleybany that was still targeting 2026.1 i added a comment to13:31
haleybralonsoh: i'm not sure if you are still working on the OVN DHCP relay ?13:32
ralonsohnot right now, I'll update it after the PTG13:32
ralonsohwe already discussed it and approved the RFE13:32
ralonsohbut this is not a high priority feature13:32
haleyback, only asking since the reviews went stale13:33
haleyb#topic community goals13:33
haleyblajoskatona: any update on sdk patches?13:33
haleyb#link https://review.opendev.org/q/topic:%22sdk_for_neutron%2213:34
lajoskatonanot much13:34
lajoskatonaI updated some heat patches13:34
lajoskatonaother than the nova ones13:34
lajoskatonahttps://review.opendev.org/q/topic:%22neutron_openstacksdk%2213:34
lajoskatonaI have a good feeling that this cycle I can find somebody to review them :-)13:35
lajoskatonathat's it13:35
haleyback, but it on my list13:36
haleybthe other series that lajoskatona and slaweq are working on is the neutronclient migration to OSC13:37
haleyb#link https://review.opendev.org/q/topic:%22migrate_stadium_osc%2213:37
lajoskatonayes, I have to go back to that but last week was short and crowded with downstream things...13:37
haleyback, thanks13:38
haleyband i guess i should have added a topic for the tenant_id to project_id migration, but i've been putting most of those on our priority dashboard when they're ready13:39
haleyb#topic on-demand13:40
haleybi did not see anything new in on-demand list, just some stale items i forgot to remove13:41
haleyblajoskatona: i had one question regarding https://review.opendev.org/c/openstack/os-ken/+/98238013:41
haleybwere you going to add a release note for that?13:42
lajoskatonagood question, perhaps I can13:42
haleybwould be good to get merged, still time for it to make our LTS13:43
lajoskatonaack, I will add a re-no13:44
haleybthat was all i had, any other topics?13:44
haleybok, thanks for attending everyone, have a great week! and just a reminder no drivers meeting this week13:45
haleyb#endmeeting13:45
opendevmeetMeeting ended Tue Apr  7 13:45:52 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)13:45
opendevmeetMinutes:        https://meetings.opendev.org/meetings/networking/2026/networking.2026-04-07-13.00.html13:45
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/networking/2026/networking.2026-04-07-13.00.txt13:45
opendevmeetLog:            https://meetings.opendev.org/meetings/networking/2026/networking.2026-04-07-13.00.log.html13:45
lajoskatonao/13:46
mlavalle\o13:46
ralonsohbye13:46
rubasovo/13:47
slaweqo/13:49
lajoskatonahaleyb, slaweq: question: do you perhaps know if we migrate to matrix from IRC? As opendev migrated few moths back I beleived that slowly everybody will migrate, but there's silence so perhaps we will remain on IRC for years13:56
haleyblajoskatona: i haven't heard, but would vote to stay on IRC myself :)14:04
lajoskatonahaleyb: I am fine with anything that works, and only one tool for all communication in the community :-)14:09
lajoskatonahaleyb: just to say it loud and keep my self safe: even a tox -epy312 --- something or tox -ereleasenotes  (and of course git pull) fails 14:13
haleyblajoskatona: in the gate?14:14
lajoskatonaas I see on the matrix opendev channel infra struggles to keep things working14:14
lajoskatonagerrit and zuul seems to be working, opendev and launchpad are the bottlenecks today it seems14:14
haleyboh, i wonder if that's launchpad or opendev or github?14:15
lajoskatonaI see messages on opendev matrix channel like this one:14:15
lajoskatonafungi14:15
lajoskatona15:4814:15
lajoskatonalooks like another crawler botnet hitting every commit url, but this one's using random user agent strings claiming to be android or iphone14:15
haleyblet me look what's going on locally14:15
fungiyeah, word in launchpad's matrix room is they're overrun with new crawlers again too14:17
lajoskatonafungi: thanks, this morning it was hard to collect the few bugs from last week for the weekly deputy report :-)14:30
fungii can imagine. i'm struggling with lp as well (in parallel with trying to get opendev's gitea responsive again)14:31
lajoskatonafungi: yeah, I suppose that is why I can't run tox -epy312 or any tox where it tries to fetch upper constraints (I just try to use github link instead of https://releases.openstack.org/constraints/upper/master )14:46
lajoskatonafungi: thanks for fighting on the frontiers14:46
lajoskatonaI am not sure that this "side effect" of LLMs/AI is visible enough for the public behind the fantastic stories of AI reviews and wibe coding etc....14:47
fungiyes, i've tried to bring this up everywhere i can. all the open source communities i'm involved in (not just opendev/openstack/openinfra ones) are spending already precious maintainer time trying to combat this14:56
fungiin particular it's hitting the least-well-staffed roles in communities like infrastructure management, vulnerability handling, code review...14:56
fungithings society had already started to come around to identifying as in dire need of help well before the "ai revolution" started14:57
fungiso now all of that has just gotten 10x worse14:57
opendevreviewTakashi Kajinami proposed openstack/neutron master: Use constant for ha_failover_strategy value  https://review.opendev.org/c/openstack/neutron/+/98337116:15
opendevreviewTakashi Kajinami proposed openstack/neutron master: Do not explicitly mention supported values  https://review.opendev.org/c/openstack/neutron/+/98337216:15
-opendevstatus- NOTICE: Load on the opendev.org Gitea backends is under control again for now, if any Zuul jobs failed with SSL errors or disconnects reaching the service prior to 16:15 UTC they can be safely rechecked17:04
opendevreviewLajos Katona proposed openstack/os-ken master: Add native threading LoopingCall to BGP  https://review.opendev.org/c/openstack/os-ken/+/98238017:24
*** dmellado471 is now known as dmellado4717:48
opendevreviewBrian Haley proposed openstack/os-ken master: Add native threading LoopingCall to BGP  https://review.opendev.org/c/openstack/os-ken/+/98238020:48

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!