| opendevreview | Rodolfo Alonso proposed x/whitebox-neutron-tempest-plugin master: Physical network max-bw QoS is applied on the localnet port https://review.opendev.org/c/x/whitebox-neutron-tempest-plugin/+/942567 | 06:24 |
|---|---|---|
| opendevreview | Dmitriy Chubinidze proposed openstack/neutron master: Update CIDR notation for loopback address according to actual RFC (https://datatracker.ietf.org/doc/html/rfc5735) https://review.opendev.org/c/openstack/neutron/+/942763 | 08:49 |
| opendevreview | Dmitriy Chubinidze proposed openstack/neutron master: Update CIDR notation for loopback address according to actual RFC (https://datatracker.ietf.org/doc/html/rfc5735) https://review.opendev.org/c/openstack/neutron/+/942763 | 08:52 |
| ralonsoh | hi folks, qq to confirm that | 08:59 |
| ralonsoh | If I want to specify a version to checkout for a project in the CI | 09:00 |
| ralonsoh | actually a hash | 09:00 |
| ralonsoh | e.g.: https://review.opendev.org/c/openstack/neutron/+/942718/1/zuul.d/base.yaml | 09:00 |
| ralonsoh | can I add the hash in override-checkout? | 09:00 |
| ralonsoh | slaweq, ^ | 09:03 |
| slaweq | I never tried it but that should works I think | 09:11 |
| ralonsoh | cool, I'm going to test that because I can't find locally the issue with pyroute2 master | 09:12 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942695 | 09:47 |
| ralonsoh | slaweq, if you have 1 min: https://review.opendev.org/c/openstack/neutron/+/938541 | 09:49 |
| ralonsoh | and https://review.opendev.org/c/openstack/python-openstackclient/+/942175 | 09:51 |
| opendevreview | Merged openstack/os-ken unmaintained/2023.1: Update .gitreview for unmaintained/2023.1 https://review.opendev.org/c/openstack/os-ken/+/935105 | 10:17 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-tempest-plugin master: Redefine ``configure_vlan_transparent`` method again https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/942776 | 10:27 |
| lucadelmonte | hello, i was having some issue with the creation of external ports in Neutron/OVN of type baremetal, the port was created fine but the HA_Chassi_Group member were changing priority at every port creation, unbinding from the ex-master chassis to be bound to the new one ad infinitum | 11:40 |
| lucadelmonte | i think i found the culprit in this code path https://github.com/openstack/neutron/blob/99f7f13403b84cf463889ce0174ad37a76fcd965/neutron/common/ovn/utils.py#L1121-L1126, the ha_ch_grp_cmd since it is in a transaction does not have the result even if the ha_group is already present | 11:42 |
| opendevreview | Stefan Hoffmann proposed openstack/os-vif master: [ovs] Add method to get external_ids from Interface https://review.opendev.org/c/openstack/os-vif/+/942715 | 11:43 |
| lucadelmonte | i managed to fix it adding a txn.commit() before the check of ha_ch_grp_cmd.result, i wanted to ask if you think this could create some issue and what i am suspecting makes sense. | 11:43 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942695 | 12:12 |
| ralonsoh | lucadelmonte, this should be tested in https://review.opendev.org/c/openstack/neutron/+/909194/36/neutron/tests/functional/plugins/ml2/drivers/ovn/mech_driver/ovsdb/test_ovn_client.py | 12:15 |
| ralonsoh | lucadelmonte, can you propose a patch, for functional tests, that probes that the current code is broken? | 12:15 |
| ralonsoh | proves* | 12:15 |
| ralonsoh | although this test is not creating the resources in the same transaction | 12:16 |
| ralonsoh | that could be something to test | 12:17 |
| lucadelmonte | ralonsoh i can try, this is my first time being involed in neutron/opestack code, i will try to make a test proving it and get back to you | 12:17 |
| lucadelmonte | yeah i saw a commit by you that removed exactly the get of the HA group and grouped it in the transaction | 12:17 |
| lucadelmonte | i suspect this is the source of the issue | 12:17 |
| ralonsoh | lucadelmonte, it could be as easy as changing how this test is executed | 12:17 |
| lucadelmonte | https://github.com/openstack/neutron/commit/25a1809964f4845d98d9b3d05148b764a65c9458 | 12:18 |
| lucadelmonte | you can see that before you change the retrival of the ha_group was done ouside of the transaction | 12:18 |
| ralonsoh | but that doesn't work if this ``_sync_ha_chassis_group`` method is called inside a txn | 12:18 |
| ralonsoh | so we need that and fix it if that is broken | 12:19 |
| lucadelmonte | i am pretty sure because i added some nasty debug messages in my local installation and confirmed that the result is None even if the ha_group is there, while if i manually commit the transacction the result is there, i will work on the test and get back to you, | 12:19 |
| opendevreview | Vasyl Saienko proposed openstack/neutron master: Enable atop by for dsvm jobs https://review.opendev.org/c/openstack/neutron/+/942792 | 13:24 |
| opendevreview | Elod Illes proposed openstack/networking-bagpipe unmaintained/2023.1: [CI] Remove periodic stable jobs from unmaintained branch https://review.opendev.org/c/openstack/networking-bagpipe/+/942584 | 13:28 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [FT] Remove unnecesary wait in ``TestNeutronServer._check_active`` https://review.opendev.org/c/openstack/neutron/+/942798 | 13:54 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional`` https://review.opendev.org/c/openstack/neutron/+/942696 | 13:54 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942695 | 14:14 |
| danfai | hi, we were recently wondering about our placement entries for segmented networks. It seems the `used` field for the resource provider in a segment is never updated, is this by design or are we missing some entry to report that? Looking at the code, neutron does not care that much about that field, only about the reserved one | 14:25 |
| ralonsoh | danfai, can you point to the code? Placement in Neutron is used for QoS, for BW assignement | 14:36 |
| ralonsoh | what are you expecting from placement API in Neutron? | 14:37 |
| danfai | ralonsoh: in https://opendev.org/openstack/neutron/src/tag/20.0.0/neutron/services/segments/plugin.py | 14:37 |
| ralonsoh | what is the use case? what are you expecting? some details will help | 14:39 |
| danfai | in general it looks like it will update the reserved and total number of IPs for the subnets in a segment, whenever segment, subnet or port are updated. | 14:40 |
| danfai | What I was expecting is, that the used field is updated whenever fixed IPs in a subnet are used up. e.g. the number of IPs available are represented in placement | 14:40 |
| danfai | not sure this would be useful, since ip availability is also there, rather wondering if a resource provider would not be taken into account when it is full. But in that case the resource request also might not have IPV4_ADDRESS, so it probably doesn't block either there | 14:42 |
| danfai | (e.g. resource request when creating a new VM in a proider network) | 14:43 |
| ralonsoh | danfai, please open a Launchpad bug with this description | 14:44 |
| danfai | ack, thanks | 14:44 |
| ralonsoh | next week Miguel Lavalle, who created this module, will be here | 14:44 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-tempest-plugin master: Redefine ``configure_vlan_transparent`` method again https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/942776 | 15:02 |
| opendevreview | Bodo Petermann proposed openstack/python-neutronclient master: vpnaas: add support for more ciphers (auth, encryption, pfs modes) https://review.opendev.org/c/openstack/python-neutronclient/+/942815 | 15:02 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942695 | 15:40 |
| noonedeadpunk | hey folks! do you know anybody using IPv6 with OVN? | 15:58 |
| noonedeadpunk | as it looks somehow raw and untested | 16:00 |
| danfai | hi noonedeadpunk, we have OVN and IPv6 running for provider networks | 16:10 |
| noonedeadpunk | danfai: aha, okay, thanks for raising the hand :) | 16:12 |
| noonedeadpunk | I just catched/submitted 3 weird bugs on testing adding IPv6 support, which made me a bit o_O about state of it overall | 16:13 |
| danfai | good to know, we'll keep an eye :D | 16:20 |
| noonedeadpunk | danfai: worst one was that adding IPv6 to an existing provider network breaks all IPv4 FIPs | 16:32 |
| danfai | good thing we don't have FIPs :), I just looked through your bug reports o_o | 16:33 |
| noonedeadpunk | so at least keep an eye on this one... rest 2 were actually operations which are very rare... Like removing IPv6 subnet or updating it (ie - updating gateway) | 16:33 |
| noonedeadpunk | ah, lucky you then | 16:34 |
| lucadelmonte | hi, i was chatting earlier today with ralonsoh about a bug i found, he asked me to commit a fix to a functional test to prove the bug, i managed to have the test failing, since i never did it is there some guide i can follow? not sure if i have to follow some naming covention for the branch | 16:34 |
| noonedeadpunk | lucadelmonte: so branches are not thing really, as we're using gerrit. So eventually you need to have a `git-review` plugin | 16:35 |
| noonedeadpunk | then you can do a commit to any branch, including master, and do `git review` command | 16:35 |
| noonedeadpunk | but you need to have an account confiugred first at https://review.opendev.org/ | 16:35 |
| lucadelmonte | oh i see, thanks, i will create and configure gerrit and get back to you when i am done | 16:36 |
| noonedeadpunk | lucadelmonte: there's some guide here: https://docs.openstack.org/contributors/en_GB/code-and-documentation/using-gerrit.html | 16:38 |
| opendevreview | Vasyl Saienko proposed openstack/neutron master: WIP: Inherit parent port vif|vnic_type for subports https://review.opendev.org/c/openstack/neutron/+/942837 | 17:44 |
| opendevreview | Merged openstack/networking-bagpipe master: Drop unused horizon from requirements https://review.opendev.org/c/openstack/networking-bagpipe/+/933529 | 20:19 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942695 | 21:18 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [FT] Refactor the ``TestNeutronServer._check_active`` method. https://review.opendev.org/c/openstack/neutron/+/942798 | 21:39 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional`` https://review.opendev.org/c/openstack/neutron/+/942696 | 21:39 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942859 | 21:57 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!