| opendevreview | Helen Chen proposed openstack/neutron master: Retry when NetlinkDumpInterrupted exception https://review.opendev.org/c/openstack/neutron/+/994338 | 02:20 |
|---|---|---|
| opendevreview | Brian Haley proposed openstack/neutron master: Fix consider-iterating-dictionary pylint warnings https://review.opendev.org/c/openstack/neutron/+/994390 | 03:10 |
| opendevreview | Merged openstack/neutron-fwaas master: Move helper methods from FirewallPluginDb to utils module https://review.opendev.org/c/openstack/neutron-fwaas/+/994418 | 03:53 |
| opendevreview | Brian Haley proposed openstack/neutron master: Fix consider-iterating-dictionary pylint warnings https://review.opendev.org/c/openstack/neutron/+/994390 | 03:57 |
| opendevreview | Brian Haley proposed openstack/neutron master: Fix consider-iterating-dictionary pylint warnings https://review.opendev.org/c/openstack/neutron/+/994390 | 04:20 |
| *** ykarel_ is now known as ykarel | 04:24 | |
| opendevreview | Brian Haley proposed openstack/neutron-lib master: Fix superfluous-parens pylint warnings https://review.opendev.org/c/openstack/neutron-lib/+/994588 | 05:47 |
| opendevreview | Brian Haley proposed openstack/neutron-lib master: Fix unnecessary-pass pylint warnings https://review.opendev.org/c/openstack/neutron-lib/+/994589 | 05:47 |
| opendevreview | Brian Haley proposed openstack/neutron-lib master: Fix no-else-return pylint warnings https://review.opendev.org/c/openstack/neutron-lib/+/994590 | 05:47 |
| opendevreview | Bharath M V proposed openstack/neutron-tempest-plugin master: [Tap-as-a-Service] Adding VM migration scenario test. https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/991062 | 06:26 |
| opendevreview | Eduardo Olivares proposed openstack/neutron master: Add upstream EVPN multinode Zuul job https://review.opendev.org/c/openstack/neutron/+/994257 | 07:00 |
| opendevreview | Eduardo Olivares proposed openstack/neutron master: Add upstream EVPN multinode Zuul job https://review.opendev.org/c/openstack/neutron/+/994257 | 07:05 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: doc: Add supported quota resources table to ops-quotas guide https://review.opendev.org/c/openstack/neutron/+/992854 | 07:12 |
| opendevreview | Merged openstack/neutron master: Retry when NetlinkDumpInterrupted exception https://review.opendev.org/c/openstack/neutron/+/994338 | 08:21 |
| opendevreview | Merged openstack/neutron master: evpn: delete EVPN HA Chassis Group on router deletion https://review.opendev.org/c/openstack/neutron/+/993325 | 09:22 |
| opendevreview | Felix Moebius proposed openstack/ovsdbapp master: Don't call frozen_row if there are no matching events https://review.opendev.org/c/openstack/ovsdbapp/+/994611 | 09:37 |
| opendevreview | Merged openstack/python-neutronclient master: Remove Dynamic Routing client code https://review.opendev.org/c/openstack/python-neutronclient/+/988466 | 10:12 |
| opendevreview | Merged openstack/neutron master: ovn_client: Reset only defined LRP.options https://review.opendev.org/c/openstack/neutron/+/993148 | 10:29 |
| opendevreview | Eduardo Olivares proposed openstack/neutron master: Add upstream EVPN multinode Zuul job https://review.opendev.org/c/openstack/neutron/+/994257 | 10:35 |
| opendevreview | Elvira García Ruiz proposed openstack/neutron master: Fix PVLAN multinode support https://review.opendev.org/c/openstack/neutron/+/993930 | 10:39 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron-fwaas master: Migrate FWaaS agent RPC to versioned objects https://review.opendev.org/c/openstack/neutron-fwaas/+/994674 | 10:47 |
| opendevreview | Stephen Finucane proposed openstack/os-vif master: pre-commit: Bump versions https://review.opendev.org/c/openstack/os-vif/+/994676 | 11:22 |
| opendevreview | Stephen Finucane proposed openstack/os-vif master: zuul: Use openstack-python3-next-jobs template https://review.opendev.org/c/openstack/os-vif/+/994677 | 11:22 |
| opendevreview | Stephen Finucane proposed openstack/os-ken master: pre-commit: Bump versions https://review.opendev.org/c/openstack/os-ken/+/994681 | 11:25 |
| opendevreview | Stephen Finucane proposed openstack/os-ken master: zuul: Use openstack-python3-next-jobs template https://review.opendev.org/c/openstack/os-ken/+/994682 | 11:25 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: pre-commit: Bump versions https://review.opendev.org/c/openstack/neutron-lib/+/994683 | 11:25 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: zuul: Use openstack-python3-next-jobs template https://review.opendev.org/c/openstack/neutron-lib/+/994684 | 11:25 |
| opendevreview | Bharath M V proposed openstack/neutron-tempest-plugin master: [Tap-as-a-Service] Adding test cases and improvements to existing cases. https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/991062 | 11:34 |
| opendevreview | Bharath M V proposed openstack/neutron-tempest-plugin master: [Tap-as-a-Service] Adding test cases and improvements to existing cases. https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/991062 | 11:39 |
| opendevreview | Merged openstack/os-vif master: pre-commit: Bump versions https://review.opendev.org/c/openstack/os-vif/+/994676 | 11:49 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: pre-commit: Bump versions https://review.opendev.org/c/openstack/neutron-lib/+/994683 | 12:55 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: zuul: Use openstack-python3-next-jobs template https://review.opendev.org/c/openstack/neutron-lib/+/994684 | 12:55 |
| haleyb | stephenfin: so question about ^^ - i know we still see random failures in py3.14 job sometimes, at least in neutron, has that stabilized recently where we can make that voting? | 13:12 |
| haleyb | i'm assuming there will be a neutron change eventually | 13:13 |
| stephenfin | That template only runs unit tests. I assume the failures are in functional tests? | 13:13 |
| ralonsoh | no, UTs in Neutron repo, with 3.14 | 13:14 |
| ralonsoh | https://zuul.opendev.org/t/openstack/builds?job_name=openstack-tox-py314&project=openstack%2Fneutron&result=FAILURE&skip=0 | 13:14 |
| ralonsoh | no so often, but almost every day | 13:14 |
| ralonsoh | I think we need to identify what tests are failing | 13:14 |
| ralonsoh | (go claude go...) | 13:15 |
| haleyb | right, and i didn't think 3.14 was supposed to be voting this cycle | 13:15 |
| haleyb | not being a luddite... | 13:16 |
| haleyb | i guess it's not failing in neutron-lib much though | 13:19 |
| haleyb | https://zuul.opendev.org/t/openstack/builds?job_name=openstack-tox-py314&project=openstack%252Fneutron-lib&result=FAILURE&skip=0 is just spinning looking | 13:20 |
| ralonsoh | haleyb, all py314 errors are random, no clear pattern | 13:20 |
| ralonsoh | at least for these 50 results | 13:20 |
| ralonsoh | I think we should have a weekly review in the CI meeting | 13:21 |
| ralonsoh | or the bug deputy should check all the py314 errors and report a bug if needed | 13:21 |
| ralonsoh | (sorry, I did that for Neutron, not n-lib) | 13:22 |
| haleyb | i think there was a garbage collection change causing strange failures that was maybe being reverted, but i haven't kept track of it | 13:22 |
| haleyb | in py3.14 that is | 13:22 |
| opendevreview | Eduardo Olivares proposed openstack/neutron master: Add upstream EVPN multinode Zuul job https://review.opendev.org/c/openstack/neutron/+/994257 | 13:23 |
| haleyb | ralonsoh: so probably safe merging the n-lib one as the past as there hasn't been a failure since april (and all the py3 jobs failed so was a bad patch) | 13:29 |
| haleyb | we need to find the intersection (in neutron) of when the py314 job fails and the others pass | 13:30 |
| ralonsoh | yes, we need to be alert on the py314 job | 13:31 |
| ralonsoh | actually, how often does it fail?? | 13:31 |
| ralonsoh | we can move to voting, just to make it more visible, if that doesn't affect the CI stability | 13:31 |
| haleyb | ralonsoh: yes, that should be OK, and as i mentioned, when it's the *only* failure in the bunch we can have Claude look deeper | 13:34 |
| ralonsoh | I'll push a patch | 13:34 |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: evpn: Add priority to EVPN callbacks https://review.opendev.org/c/openstack/neutron/+/994739 | 13:39 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Make ``openstack-tox-py314`` job voting https://review.opendev.org/c/openstack/neutron/+/994740 | 13:40 |
| haleyb | ralonsoh: if stephenfin pushes a -next patch for neutron it will do the same thing | 13:42 |
| ralonsoh | --> https://review.opendev.org/c/openstack/neutron/+/994740 | 13:43 |
| ralonsoh | right, now I understand | 13:43 |
| haleyb | ralonsoh: right, but that isn't needed with the new template | 13:43 |
| ralonsoh | ok, I can abandon it | 13:43 |
| haleyb | i'm assuming neutron is on his list :) | 13:43 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: ovn: Handle missing ``neutron:network_name`` in sibling LRPs https://review.opendev.org/c/openstack/neutron/+/993617 | 13:50 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Allow "service" role to create a default SG https://review.opendev.org/c/openstack/neutron/+/994744 | 13:58 |
| opendevreview | Merged openstack/neutron master: Add configurable HAProxy timeouts for metadata proxy https://review.opendev.org/c/openstack/neutron/+/990588 | 14:03 |
| stephenfin | haleyb: sorry, was lunching | 14:07 |
| stephenfin | You're correct that 3.14 isn't *required* this cycle, but I'm of the opinion that we benefit if all libraries and clients can adopt the latest Python versions as soon as they can. Any issues they're likely to face will likely be much simpler to debug than similar issues in services, and it helps make the adoption in services easier. That's the basis for that new template | 14:07 |
| stephenfin | However, if the py314 jobs are failing regularly then we definitely shouldn't merge that yet. It might be worth prioritising figuring out why though. I'm doing just that for futurist atm (which similarly fails more than it should on 3.14+) | 14:08 |
| ralonsoh | not so often and we see it failing, we'll open bugs faster to investigate | 14:10 |
| ralonsoh | of course, if py314 is failing too much, we'll mark it again as non-voting | 14:10 |
| ralonsoh | (but that should not happen) | 14:10 |
| stephenfin | indeed, all good points | 14:10 |
| stephenfin | I only know futurist is flakey on py314 because I attempted to make it voting. If It hadn't, I'd only learn that in 2027.1 | 14:11 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Make ``ovs_create_tap`` True by default https://review.opendev.org/c/openstack/neutron/+/994752 | 14:26 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: segments: Catch RP not found in ``_delete_nova_inventory`` https://review.opendev.org/c/openstack/neutron/+/993899 | 14:42 |
| opendevreview | Elvira García Ruiz proposed openstack/neutron master: Add PVLAN documentation https://review.opendev.org/c/openstack/neutron/+/994760 | 15:02 |
| opendevreview | Elvira García Ruiz proposed openstack/neutron master: Add PVLAN documentation https://review.opendev.org/c/openstack/neutron/+/994760 | 15:07 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider master: Add post-quantum cryptography readiness for TLS configuration https://review.opendev.org/c/openstack/ovn-octavia-provider/+/994763 | 15:10 |
| opendevreview | Eduardo Olivares proposed openstack/neutron master: Add upstream EVPN multinode Zuul job https://review.opendev.org/c/openstack/neutron/+/994257 | 15:58 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider master: Add post-quantum cryptography readiness for TLS configuration https://review.opendev.org/c/openstack/ovn-octavia-provider/+/994763 | 16:07 |
| opendevreview | Merged openstack/os-vif master: zuul: Use openstack-python3-next-jobs template https://review.opendev.org/c/openstack/os-vif/+/994677 | 16:37 |
| opendevreview | Merged openstack/neutron-lib master: pre-commit: Bump versions https://review.opendev.org/c/openstack/neutron-lib/+/994683 | 17:15 |
| opendevreview | Merged openstack/neutron-lib master: zuul: Use openstack-python3-next-jobs template https://review.opendev.org/c/openstack/neutron-lib/+/994684 | 17:28 |
| opendevreview | Felix Moebius proposed openstack/ovsdbapp master: Don't call frozen_row if there are no matching events https://review.opendev.org/c/openstack/ovsdbapp/+/994611 | 17:44 |
| opendevreview | Merged openstack/neutron master: Fix PVLAN multinode support https://review.opendev.org/c/openstack/neutron/+/993930 | 17:47 |
| *** haleyb is now known as haleyb|out | 18:11 | |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: evpn: Remove child_vxlan_port config option https://review.opendev.org/c/openstack/neutron/+/994820 | 20:18 |
| tbachman | Hello neutron devs! | 21:39 |
| tbachman | I have a question/concern about a patch that was merged recently - https://review.opendev.org/c/openstack/neutron/+/980637 | 21:40 |
| tbachman | That was created to address bug https://bugs.launchpad.net/neutron/+bug/2144501 | 21:40 |
| tbachman | The original behavior of the constructor gets the statically created segments for the network. | 21:41 |
| tbachman | A NetworkContext gets created as part of a PortContext used for port binding here: https://opendev.org/openstack/neutron/src/branch/master/neutron/plugins/ml2/plugin.py#L727-L730 | 21:42 |
| tbachman | Since the network passed in is a DB object and not a context, the else case gets used, which creates the NetworkContext using the network that was passed in: https://opendev.org/openstack/neutron/src/branch/master/neutron/plugins/ml2/driver_context.py#L137-L138 | 21:45 |
| tbachman | I believe this was intentional - the plugin then calls the mechanism manager's bind_port method with the context, which only includes the statically allocated network segments: | 21:47 |
| tbachman | https://opendev.org/openstack/neutron/src/branch/master/neutron/plugins/ml2/plugin.py#L734 | 21:47 |
| tbachman | Those segments are then used to as the first in a series of recursive calls to _bind_port_level with a level of 0: https://opendev.org/openstack/neutron/src/branch/master/neutron/plugins/ml2/managers.py#L872-L873 | 21:48 |
| tbachman | This means that the only segments considered are the statically assigned ones (i.e. level 0 segements): | 21:49 |
| tbachman | After binding at level 0, any dynamically created segments are returned in the context, and those are then used as segments for the next level down: | 21:51 |
| tbachman | https://opendev.org/openstack/neutron/src/branch/master/neutron/plugins/ml2/managers.py#L933 | 21:51 |
| tbachman | The point being that each call to _bind_port_level would only consider the segments created at that level | 21:52 |
| tbachman | With the change that was merged above, the initial constructor ends up pulling in *all* segments for the network (i.e. static and dynamic), regardless of any relationsihp to level | 21:52 |
| tbachman | This results in mechanism drivers binding ports for dynamically allocated segments at level 0, which breaks the original intent. | 21:53 |
| tbachman | I definitely can see the issue raised in the launchpad bug.... but I'm not sure this is the right fix. | 21:54 |
| tbachman | I was going to look into the delete_port_postcommit path to see if I can come up with other options here, but I wanted to bring this issue up with the greater neutron team to see what you thought. | 22:09 |
| opendevreview | Miguel Lavalle proposed openstack/neutron master: [OVN] Implement multiple segments per host for routed networks https://review.opendev.org/c/openstack/neutron/+/975111 | 22:13 |
| opendevreview | Miguel Lavalle proposed openstack/neutron master: [OVN] Agent extension for segment bridges https://review.opendev.org/c/openstack/neutron/+/982018 | 22:19 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!