| opendevreview | Brian Haley proposed openstack/neutron master: Bump pre-commit-config versions https://review.opendev.org/c/openstack/neutron/+/976562 | 00:04 |
|---|---|---|
| opendevreview | Dmitrii Shcherbakov proposed openstack/neutron master: Add support for remote-managed PFs https://review.opendev.org/c/openstack/neutron/+/976010 | 04:52 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2025.2: Fix ``ChassisAgentWriteEvent`` match condition https://review.opendev.org/c/openstack/neutron/+/976581 | 07:41 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2025.1: Fix ``ChassisAgentWriteEvent`` match condition https://review.opendev.org/c/openstack/neutron/+/976582 | 07:41 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2024.2: Fix ``ChassisAgentWriteEvent`` match condition https://review.opendev.org/c/openstack/neutron/+/976583 | 07:41 |
| ralonsoh | bcafarel, hello! if you have less than one min ^^^ these 3 patches | 07:41 |
| ralonsoh | thanks! | 07:41 |
| opendevreview | Merged openstack/neutron master: Bump pre-commit-config versions https://review.opendev.org/c/openstack/neutron/+/976562 | 10:09 |
| lucadelmonte | Hi! I've submitted a fix for bug 1966052 (ovn-octavia-provider associates to wrong logical router on LRP events). This bug has been open since 2022 and affects LB/FIP traffic when networks have multiple routers. CI is green. Would appreciate a review when you have time: https://review.opendev.org/c/openstack/ovn-octavia-provider/+/976236 - Thanks! haleyb, ralonsoh, froyo | 12:59 |
| opendevreview | Merged openstack/ovn-octavia-provider master: Update flake8-import-order version to fix checks https://review.opendev.org/c/openstack/ovn-octavia-provider/+/976120 | 13:26 |
| elvira | haleyb: Hi!! I did not see your message regarding PVLAN on the meeting, sorry! Yes it is ready for review. I left some questions in the spec itself for discussion | 13:32 |
| ralonsoh | lucadelmonte, let me check that patch | 14:09 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Make the ``logging_rpc`` member a class property https://review.opendev.org/c/openstack/neutron/+/976130 | 14:37 |
| ralonsoh | hi folks, if you have a couple of mins, please check https://review.opendev.org/c/openstack/neutron/+/973366 | 14:38 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Update ithe FDB age threshold default value to 300 https://review.opendev.org/c/openstack/neutron/+/975908 | 14:39 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: OVN ``FDB`` table is supported since v21.03 https://review.opendev.org/c/openstack/neutron/+/976638 | 14:58 |
| haleyb | ralonsoh: was wondering if you saw my comment in https://review.opendev.org/c/openstack/neutron/+/976213 - i saw the anti-affinity test fail yesterday with your patch | 15:01 |
| ralonsoh | haleyb, no, I'm reading it now | 15:05 |
| ralonsoh | I don't know if we should revert this optimization | 15:05 |
| ralonsoh | the new algorithm is not always delivering the best scheduling | 15:06 |
| ralonsoh | but that is in a few cases and this new code is much faster | 15:06 |
| haleyb | ralonsoh: so it's the algorithm not the test? | 15:19 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: WIP == [OVN] Make ML2/OVN compatible with AAP with IP+MAC https://review.opendev.org/c/openstack/neutron/+/976645 | 15:38 |
| ralonsoh | haleyb, yes, the algorithm doesn't always return the optimal solution, due to this: | 15:38 |
| ralonsoh | https://github.com/openstack/neutron/blob/master/neutron/scheduler/l3_ovn_scheduler.py#L232-L233 | 15:39 |
| ralonsoh | well, actually this: https://github.com/openstack/neutron/blob/master/neutron/scheduler/l3_ovn_scheduler.py#L235-L243 | 15:40 |
| ralonsoh | if the `leastloaded_by_priority` variable doesn't allow to find a selection of chassis, all of them unique, from highest prio to lowest prio | 15:40 |
| ralonsoh | what we are doing is just, from the bottom (the lowest prio chassis), adding the all of them | 15:41 |
| ralonsoh | haleyb, in a nutshell: the highest prio chassis are selected using the weights but it could be possible that that doesn't happen for the lowest ones | 15:42 |
| ralonsoh | to be fair, this is aceptable because of the speed gain | 15:42 |
| ralonsoh | of course, some AI could check this method and provide an accurate solution that always return the optimal chassis selection, even for the lowest proritu ones | 15:43 |
| haleyb | ralonsoh: sure, plug it into Claude, bleep bloop problem solved :) | 16:00 |
| opendevreview | Merged openstack/neutron master: bgp: Use port mapping for learned routes https://review.opendev.org/c/openstack/neutron/+/975674 | 19:09 |
| opendevreview | Merged openstack/neutron master: Revert "[OVN] Fix FDB table not registered in OvnSbIdl" https://review.opendev.org/c/openstack/neutron/+/975846 | 19:13 |
| opendevreview | Brian Haley proposed openstack/neutron stable/2025.1: Fix issues related to pkg_resources module https://review.opendev.org/c/openstack/neutron/+/976685 | 21:30 |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: Add neutron-id to DHCP_Options entries https://review.opendev.org/c/openstack/neutron/+/976692 | 22:28 |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: Add neutron-id to DHCP_Options entries https://review.opendev.org/c/openstack/neutron/+/976692 | 22:29 |
| opendevreview | Brian Haley proposed openstack/neutron stable/2025.1: Fix issues related to pkg_resources module https://review.opendev.org/c/openstack/neutron/+/976685 | 23:11 |
| opendevreview | JayGue Lee proposed openstack/ovn-octavia-provider master: ovn-octavia-provider: improve provider network detection for LB resources https://review.opendev.org/c/openstack/ovn-octavia-provider/+/976201 | 23:42 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!