| opendevreview | Merged openstack/neutron master: pre-commit: Migrate to ruff, enable autopep8 https://review.opendev.org/c/openstack/neutron/+/950580 | 02:29 |
|---|---|---|
| opendevreview | OpenStack Proposal Bot proposed openstack/neutron master: Imported Translations from Zanata https://review.opendev.org/c/openstack/neutron/+/954395 | 02:50 |
| opendevreview | OpenStack Proposal Bot proposed openstack/neutron-fwaas-dashboard stable/2024.1: Imported Translations from Zanata https://review.opendev.org/c/openstack/neutron-fwaas-dashboard/+/954397 | 02:55 |
| opendevreview | OpenStack Proposal Bot proposed openstack/neutron-fwaas-dashboard stable/2025.1: Imported Translations from Zanata https://review.opendev.org/c/openstack/neutron-fwaas-dashboard/+/954418 | 03:45 |
| opendevreview | OpenStack Proposal Bot proposed openstack/neutron-vpnaas-dashboard stable/2025.1: Imported Translations from Zanata https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/954421 | 03:52 |
| opendevreview | OpenStack Proposal Bot proposed openstack/neutron-vpnaas-dashboard stable/2024.1: Imported Translations from Zanata https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/954430 | 04:12 |
| CristiC | hi there, is there a compatibility matrix between neutron w/ ml2/ovn releases and OVN/OVS releases? thank you in advance. | 09:38 |
| elvira1 | CristiC: Not sure if this is a good way but you can check which default is used in devstack for each branch https://github.com/openstack/devstack/blob/56c6ba1d5f390e510ab28c2254bde2b2c1183097/lib/neutron_plugins/ovn_agent#L31 This is not a matrix though, just a suggestion of an OVN version that should work. Maybe someone else has a better way of looking into it | 10:03 |
| *** elvira1 is now known as elvira | 10:04 | |
| opendevreview | Slawek Kaplonski proposed openstack/neutron-lib master: Add ALLOWED_ADDRESS_PAIR to the callbacks resources https://review.opendev.org/c/openstack/neutron-lib/+/954444 | 10:50 |
| opendevreview | Merged openstack/neutron-lib master: Add "has_global_access" attribute to the context object https://review.opendev.org/c/openstack/neutron-lib/+/954054 | 11:19 |
| opendevreview | Sami Yessou proposed openstack/ovn-octavia-provider stable/2024.1: Patching Octavia LB pool removal issue and adding updated unit tests https://review.opendev.org/c/openstack/ovn-octavia-provider/+/954318 | 11:28 |
| ykarel | slaweq, ralonsoh if you can check https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/954240 and https://review.opendev.org/c/openstack/neutron/+/954222 | 11:54 |
| ralonsoh | sure, let me check | 11:57 |
| CristiC | elvira: thank you for your answer | 12:15 |
| ralonsoh | slaweq, haleyb I think we can release a new n-lib version, in particular because of the "has_global_access" flag in the context | 13:14 |
| opendevreview | Merged openstack/neutron master: [FT] Mark test_fip_connection_for_address_scope unstable https://review.opendev.org/c/openstack/neutron/+/954222 | 13:31 |
| opendevreview | Merged openstack/neutron-tempest-plugin master: Use infra mirrors for advanced image prepare v2 https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/954240 | 13:31 |
| haleyb | ralonsoh: sure. i think there's only one other n-lib change open, might as well merge it too, https://review.opendev.org/c/openstack/neutron-lib/+/954444 | 13:34 |
| ralonsoh | let me check | 13:35 |
| ralonsoh | ah yes, we can merge it too | 13:35 |
| haleyb | it was just opened today | 13:35 |
| ralonsoh | and https://review.opendev.org/c/openstack/neutron-lib/+/953198 is in the oven now | 13:35 |
| ralonsoh | let's wait for these 2 patches | 13:36 |
| opendevreview | Michal Nasiadka proposed openstack/neutron master: Introduce noop OVN Router scheduler https://review.opendev.org/c/openstack/neutron/+/953934 | 14:15 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider master: Increase coverage tests https://review.opendev.org/c/openstack/ovn-octavia-provider/+/954501 | 14:25 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider master: Increase coverage tests https://review.opendev.org/c/openstack/ovn-octavia-provider/+/954501 | 14:26 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Add deprecation warning for the OVN Metadata agent https://review.opendev.org/c/openstack/neutron/+/952319 | 14:34 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Add deprecation warning for the OVN Metadata agent https://review.opendev.org/c/openstack/neutron/+/952319 | 14:35 |
| haleyb | ralonsoh: do you remember a bug in neutron where we saw sqlalchemy QueuePool timeouts? I couldn't find one just seemed familiar | 15:02 |
| haleyb | sqlalchemy.exc.TimeoutError: QueuePool limit of size 5 overflow 50 reached, connection timed out, timeout 30.00 (Background on this error at: https://sqlalche.me/e/14/3o7r) | 15:02 |
| ralonsoh | haleyb, no, I don't remember this error | 15:03 |
| ralonsoh | sqlalchemy has been quite for a long time | 15:03 |
| ralonsoh | where did you find it? | 15:03 |
| haleyb | ralonsoh: ack, i see on yoga (old i know) it just sounded familiar so i started looking for old bugs but found nothing for neutron | 15:04 |
| haleyb | i will keep digging, nova had a similar issue and fix involved moving DB reader decorators. i'll have to see what happens on our caracal cloud | 15:07 |
| ralonsoh | https://bugs.launchpad.net/nova/+bug/2027755 | 15:07 |
| ralonsoh | yeah, I was going to paste the same bug | 15:07 |
| haleyb | yup, that one was easy to find | 15:07 |
| ralonsoh | so it could be possible that we don't have the correct context wrapping this operation | 15:08 |
| haleyb | in the neutron case it's a "get subnets" call i'm seeing, but too much has changed from yoga to today | 15:08 |
| ralonsoh | haleyb, let me check later today, I need to leave now. I'll ping you later (or maybe you find something) | 15:09 |
| haleyb | we will eventually move this workload to a caracal cloud and see if it happens there, but could be months away | 15:10 |
| haleyb | ralonsoh: ack, thanks. don't spend too much time | 15:11 |
| opendevreview | Merged openstack/neutron-lib master: Remove constants for LBaaS https://review.opendev.org/c/openstack/neutron-lib/+/953198 | 15:35 |
| opendevreview | Merged openstack/neutron-lib master: Add ALLOWED_ADDRESS_PAIR to the callbacks resources https://review.opendev.org/c/openstack/neutron-lib/+/954444 | 15:39 |
| opendevreview | Merged openstack/os-ken master: Fixing the BGP loop prevention https://review.opendev.org/c/openstack/os-ken/+/953631 | 16:01 |
| opendevreview | Merged openstack/neutron master: tox: Fix indentation https://review.opendev.org/c/openstack/neutron/+/950581 | 16:22 |
| opendevreview | Merged openstack/neutron master: docs: Remove crud from conf.py files https://review.opendev.org/c/openstack/neutron/+/953865 | 16:22 |
| opendevreview | Merged openstack/neutron master: Replace the OVN Metadata agent with the OVN agent https://review.opendev.org/c/openstack/neutron/+/953403 | 16:47 |
| opendevreview | Michal Nasiadka proposed openstack/neutron master: Introduce noop OVN Router scheduler https://review.opendev.org/c/openstack/neutron/+/953934 | 17:02 |
| opendevreview | Brian Haley proposed openstack/neutron master: OVS firewall: only remove security group when truly unused https://review.opendev.org/c/openstack/neutron/+/954359 | 17:04 |
| mnasiadka | Hello, I picked up vsaienko's patch (https://review.opendev.org/c/openstack/neutron/+/953934) - would be grateful for some insight if that's a good path forward for the gateway chassis vs ha chassis group allocation problem (cc: ralonsoh because you did post some comments on the original patch) | 20:06 |
| ralonsoh | mnasiadka, I already commented that before: there is a LP bug to migrate the schedulers from gateway_chassis to ha_chassis_group | 20:32 |
| ralonsoh | https://review.opendev.org/c/openstack/neutron/+/947317 | 20:32 |
| ralonsoh | https://review.opendev.org/q/topic:%22bug/2092271%22 | 20:32 |
| ralonsoh | right now we can't continue because of a dependency in octavia-ovn-provider | 20:32 |
| ralonsoh | but next cycle I'll finish this series of patches and both schedulers will use hcg instead of gc | 20:33 |
| mnasiadka | ralonsoh: that bug is there for so long, that I’ll probably need to think if I shouldn’t go the noop scheduler path downstream for now… I’m a bit fed up with other workarounds | 20:44 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!