opendevreview | Zhoneym proposed openstack/kolla-ansible master: Limit bcrypt version to <=4.0.1 for compatibility https://review.opendev.org/c/openstack/kolla-ansible/+/943184 | 00:23 |
---|---|---|
opendevreview | Zhoneym proposed openstack/kolla-ansible master: Limit bcrypt version to == 4.0.1 for compatibility https://review.opendev.org/c/openstack/kolla-ansible/+/943184 | 00:24 |
opendevreview | Zhoneym proposed openstack/kolla-ansible master: Limit bcrypt version to == 4.0.1 for compatibility https://review.opendev.org/c/openstack/kolla-ansible/+/943184 | 00:24 |
opendevreview | Michael Still proposed openstack/kolla-ansible master: Fix docs rendering errors. https://review.opendev.org/c/openstack/kolla-ansible/+/944184 | 01:37 |
opendevreview | Merged openstack/kolla-ansible master: Fix typo in prechecks command https://review.opendev.org/c/openstack/kolla-ansible/+/944117 | 05:34 |
frickler | getting "Unsupported server version: '10.11.10-MariaDB-ubu2204-log'." when trying to run mariadb_backup against proxysql, does that ring a bell for anyone? | 06:40 |
frickler | mnasiadka: I have held a node now to look at the podman issues, your debug output looks like the container isn't even running? | 06:44 |
frickler | mnasiadka: https://review.opendev.org/c/openstack/neutron/+/942916 , the agent really no longer uses the port we are healthchecking. now the big bonus question is: why is this only breaking podman? are our docker healthchecks bogus? | 06:54 |
frickler | I must admit I also do not understand why we are checking the rpc port | 06:57 |
mnasiadka | frickler: well, it seems that metadata agent is connecting to rabbitmq (hence the rpc port), but have no clue yet why it fails only on podman - will have time to look in around an hour :) | 07:05 |
frickler | ah, maybe this is only for OVS and we run docker with OVN? | 07:06 |
opendevreview | Dr. Jens Harbott proposed openstack/kolla-ansible master: DNM: disable neutron-metadata-agent healthcheck https://review.opendev.org/c/openstack/kolla-ansible/+/944194 | 07:10 |
mnasiadka | frickler: nope, we run ovs everywhere and ovn only in -ovn jobs | 07:22 |
tafkamax | Hi I have a question regarding hw_offload for ovs. I have Mellanox CX5 and CX6 NIC-s. Should I enable the feature to take strain off the cpu? | 08:06 |
tafkamax | https://docs.openstack.org/neutron/2024.2/admin/config-sriov.html | 08:06 |
tafkamax | https://docs.openstack.org/neutron/2024.2/admin/config-ovs-offload.html | 08:06 |
tafkamax | looking at the docs, there are some limitations | 08:06 |
tafkamax | As I understand the Neutron security groups wont work? | 08:07 |
tafkamax | Also to use the HW offloading I need to specify it when creating a vm or port at CLI or API level? As I understand it will not be a default option, e.g. all VM-s created via UI or CLI by default will have hw offload support? | 08:15 |
tafkamax | Also, isn't SRIOV per VM? So if I create 4 virtual interfaces in the mellanox card, then only 4 vm-s can use them? | 08:16 |
tafkamax | or am I misunderstanding something | 08:17 |
tafkamax | or are they shared? the CX5 cards have max of 8 virtual functions. | 08:22 |
frickler | tafkamax: that doesn't sound kolla related at all | 08:22 |
tafkamax | sorry, you guys are just very active | 08:22 |
tafkamax | will take it to another channel | 08:22 |
frickler | mnasiadka: do we want to merge my disable patch joined with your debugging patch? seems most jobs passed already | 08:23 |
frickler | (only as CI fix, finding a proper new healthcheck still needs to be done afterwards) | 08:24 |
mnasiadka | frickler: I guess so - we should not block merging other patches on a broken healthcheck ;-) | 09:33 |
opendevreview | Merged openstack/kayobe master: CI: Make kayobe-seed-images-rocky9-podman experimental https://review.opendev.org/c/openstack/kayobe/+/944076 | 10:00 |
mnasiadka | frickler: but I guess we should remove DNM from the commit message :) | 10:02 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: neutron-metadata-agent: Disable healthcheck https://review.opendev.org/c/openstack/kolla-ansible/+/944194 | 10:03 |
frickler | mnasiadka: I think we'll also need to squash the changes in order to make the CI pass? or swap the order at least? | 10:07 |
mnasiadka | yeah, I'll rebase my CI ,,fix'' on top of your patch | 10:07 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Wait until containers exit health starting state https://review.opendev.org/c/openstack/kolla-ansible/+/944104 | 10:07 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Wait until containers exit health starting state https://review.opendev.org/c/openstack/kolla-ansible/+/944104 | 10:08 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: neutron-metadata-agent: Disable healthcheck https://review.opendev.org/c/openstack/kolla-ansible/+/944194 | 10:08 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Wait until containers exit health starting state https://review.opendev.org/c/openstack/kolla-ansible/+/944104 | 10:08 |
mnasiadka | now should be good | 10:08 |
mnasiadka | (some misclicks) :D | 10:08 |
frickler | you missed the backport-candidate flag ;) also I'll enqueue it right into gate now, hoping we don't have too much other CI flakiness | 10:20 |
opendevreview | Michal Nasiadka proposed openstack/kolla stable/2024.1: Added systemd-standalone-sysusers package for Debuntu https://review.opendev.org/c/openstack/kolla/+/944100 | 10:27 |
opendevreview | Normo proposed openstack/kolla-ansible master: Fix variable name in globals.yml for enabling the prometheus memcached_exporter https://review.opendev.org/c/openstack/kolla-ansible/+/944205 | 10:28 |
opendevreview | Normo proposed openstack/kolla-ansible master: Fix variable name in globals.yml for enabling the memcached_exporter https://review.opendev.org/c/openstack/kolla-ansible/+/944205 | 10:29 |
opendevreview | Maksim Malchuk proposed openstack/kolla-ansible stable/2023.1: Revert "Skyline: Fix incorrect keystone port" https://review.opendev.org/c/openstack/kolla-ansible/+/944206 | 10:47 |
opendevreview | Pierre Riteau proposed openstack/kayobe master: Remove remaining mentions of Ubuntu Jammy https://review.opendev.org/c/openstack/kayobe/+/944075 | 11:17 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Wait until containers exit health starting state https://review.opendev.org/c/openstack/kolla-ansible/+/944104 | 11:55 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Wait until containers exit health starting state https://review.opendev.org/c/openstack/kolla-ansible/+/944104 | 11:55 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Fix boolean representation in all configurations https://review.opendev.org/c/openstack/kolla-ansible/+/944124 | 12:02 |
opendevreview | Dr. Jens Harbott proposed openstack/kolla-ansible master: Cinder: move to block-storage in service catalog https://review.opendev.org/c/openstack/kolla-ansible/+/937038 | 12:02 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible stable/2024.1: Add support for Ubuntu Noble Numbat (24.04 LTS) https://review.opendev.org/c/openstack/kolla-ansible/+/932539 | 12:06 |
opendevreview | Verification of a change to openstack/kolla-ansible master failed: neutron-metadata-agent: Disable healthcheck https://review.opendev.org/c/openstack/kolla-ansible/+/944194 | 12:29 |
opendevreview | Merged openstack/kolla-ansible master: Fix variable name in globals.yml for enabling the memcached_exporter https://review.opendev.org/c/openstack/kolla-ansible/+/944205 | 12:41 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Wait until containers exit health starting state https://review.opendev.org/c/openstack/kolla-ansible/+/944104 | 13:03 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible stable/2024.1: Add support for Ubuntu Noble Numbat (24.04 LTS) https://review.opendev.org/c/openstack/kolla-ansible/+/932539 | 13:16 |
mnasiadka | frickler: maybe I'm an ignorant, but why are we using catalog_info for Cinder only? :) | 13:17 |
frickler | mnasiadka: I asked that question in the review, too, tbh I have no idea, just wanted to keep the existing stuff in place as far as possible | 13:21 |
frickler | maybe back in the past the other options weren't there yet? | 13:22 |
frickler | but I could also agree if you would prefer the "interface = internal" variant | 13:23 |
mnasiadka | Well, maybe that should be a separate patch that we don't backport. | 13:26 |
frickler | mnasiadka: I still wouldn't backport either patch, since that could disrupt existing deployments, couldn't it? | 13:50 |
mnasiadka | I guess it could. | 13:50 |
mnasiadka | I don't want you to go back and forth... but maybe let's change all occurences of catalog_info to interface? ;-) | 13:50 |
frickler | oh, there's also this thing that was completely missed it seems: ansible/roles/glance/templates/glance-api.conf.j2:cinder_catalog_info = volume:cinder:internalURL | 13:54 |
frickler | ok, I'll do yet another revision, just hoping we'll finally have consensus, then ;) | 13:56 |
frickler | I'm also confused because that glance usage can't possibly work in our current setup where we only have cinderv3?!? | 13:57 |
frickler | ah, iiuc that is because that option is in [DEFAULT], which it would belong into [cinder], so I guess we are lucky and it just gets ignored | 14:01 |
frickler | humm, but also of course glance doesn't support the "interface" option, at least it isn't mentioned in the sample glance-api.conf | 14:04 |
frickler | mnasiadka: ^^ a) just drop the broken cinder_catalog_info option from glance? b) move it to the [cinder] section and switch it like for nova? c) (optional) nag glance people about the "interface" option? | 14:05 |
mnasiadka | nagging glance is not going to be easy or successful I assume, so let's rule that out. | 14:09 |
mnasiadka | I would do a) I think | 14:09 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Fix boolean representation in all configurations https://review.opendev.org/c/openstack/kolla-ansible/+/944124 | 14:09 |
opendevreview | Michal Nasiadka proposed openstack/kolla master: libvirt: Rename nova-libvirt to libvirt https://review.opendev.org/c/openstack/kolla/+/938325 | 14:13 |
opendevreview | Michal Nasiadka proposed openstack/kolla master: libvirt: Rename nova-libvirt to libvirt https://review.opendev.org/c/openstack/kolla/+/938325 | 14:13 |
opendevreview | Michal Nasiadka proposed openstack/kolla master: libvirt: Rename nova-libvirt to libvirt https://review.opendev.org/c/openstack/kolla/+/938325 | 14:13 |
opendevreview | Mickael Razzouk proposed openstack/kolla master: Added libvirt dependency to nova-libvirt debian based containers https://review.opendev.org/c/openstack/kolla/+/934952 | 14:19 |
opendevreview | Grzegorz Koper proposed openstack/kolla master: [release] Use RDO's Epoxy packages https://review.opendev.org/c/openstack/kolla/+/944263 | 14:33 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Drop Cinder HNAS support https://review.opendev.org/c/openstack/kolla-ansible/+/874381 | 14:35 |
opendevreview | Dr. Jens Harbott proposed openstack/kolla-ansible master: Cinder: move to block-storage in service catalog https://review.opendev.org/c/openstack/kolla-ansible/+/937038 | 14:39 |
mnasiadka | frickler: booked 2x1h slot for Kolla (Tue and Wed), it needs to be enough this time :) | 14:51 |
mnasiadka | (PTG) | 14:51 |
opendevreview | Verification of a change to openstack/kolla-ansible master failed: neutron-metadata-agent: Disable healthcheck https://review.opendev.org/c/openstack/kolla-ansible/+/944194 | 14:51 |
opendevreview | Christian Berendt proposed openstack/kolla-ansible master: cinder: add cinder_file_datadir_volume parameter https://review.opendev.org/c/openstack/kolla-ansible/+/944267 | 14:53 |
test_ | ls | 14:56 |
test_ | ls | 14:56 |
test_ | ?????????? | 14:56 |
frickler | mnasiadka: ack, we could do some followup a couple of weeks later if needed | 14:57 |
opendevreview | Michal Nasiadka proposed openstack/kolla master: CI: Add qemu-user-static back https://review.opendev.org/c/openstack/kolla/+/944273 | 15:01 |
mnasiadka | frickler: makes sense | 15:01 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Fix boolean representation in all configurations https://review.opendev.org/c/openstack/kolla-ansible/+/944124 | 15:19 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ceilometer scenario https://review.opendev.org/c/openstack/kolla-ansible/+/944276 | 15:21 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: gnocchi: Add support for using uWSGI https://review.opendev.org/c/openstack/kolla-ansible/+/940101 | 15:23 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Fix boolean representation in all configurations https://review.opendev.org/c/openstack/kolla-ansible/+/944124 | 15:23 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ceilometer scenario https://review.opendev.org/c/openstack/kolla-ansible/+/944276 | 15:33 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: gnocchi: Add support for using uWSGI https://review.opendev.org/c/openstack/kolla-ansible/+/940101 | 15:33 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Run ML2/OVS agents processes in separate containers https://review.opendev.org/c/openstack/kolla-ansible/+/864780 | 15:34 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Run ML2/OVS agents processes in separate containers https://review.opendev.org/c/openstack/kolla-ansible/+/864780 | 15:38 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Run ML2/OVS agents processes in separate containers https://review.opendev.org/c/openstack/kolla-ansible/+/864780 | 15:38 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Run ML2/OVS agents processes in separate containers https://review.opendev.org/c/openstack/kolla-ansible/+/864780 | 15:39 |
opendevreview | Michal Nasiadka proposed openstack/kolla master: Add podman/docker clients for ML2/OVS improvement https://review.opendev.org/c/openstack/kolla/+/913473 | 15:47 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Run ML2/OVS agents processes in separate containers https://review.opendev.org/c/openstack/kolla-ansible/+/864780 | 15:49 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Bump ansible-lint to <25.2 https://review.opendev.org/c/openstack/kolla-ansible/+/937242 | 16:01 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Make CI roles passable with ansible-lint shared profile https://review.opendev.org/c/openstack/kolla-ansible/+/920159 | 16:01 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Bump ansible-lint to <25.2 https://review.opendev.org/c/openstack/kolla-ansible/+/937242 | 16:01 |
kevko | frickler: I know what is your issue with backup through the proxysql | 16:03 |
kevko | or - maybe I know | 16:03 |
frickler | kevko: I already found https://review.opendev.org/c/openstack/kolla/+/934346 but somehow this is still broken | 16:50 |
kevko | frickler: well, check if script is working ..it should return the backup host | 16:54 |
kevko | frickler: mariabackup is tested in CI :/ | 16:54 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Fix Redis Sentinel authentication for octavia's jobboard HA https://review.opendev.org/c/openstack/kolla-ansible/+/942799 | 17:02 |
opendevreview | Verification of a change to openstack/kolla-ansible master failed: neutron-metadata-agent: Disable healthcheck https://review.opendev.org/c/openstack/kolla-ansible/+/944194 | 17:14 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: neutron-metadata-agent: Disable healthcheck https://review.opendev.org/c/openstack/kolla-ansible/+/944194 | 17:19 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Fix boolean representation in all configurations https://review.opendev.org/c/openstack/kolla-ansible/+/944124 | 17:19 |
frickler | kevko: I didn't mention that I'm trying this on 2024.1, seems we need to backport https://review.opendev.org/c/openstack/kolla-ansible/+/913724 downstream | 17:52 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add oslo.messaging Queue Manager https://review.opendev.org/c/openstack/kolla-ansible/+/924623 | 18:52 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Allow ignoring missing containers in service-stop https://review.opendev.org/c/openstack/kolla-ansible/+/939657 | 18:52 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Rabbitmq: enable quorum for transient queues https://review.opendev.org/c/openstack/kolla-ansible/+/927497 | 18:52 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: rabbitmq: Add support for using stream queues for fanout https://review.opendev.org/c/openstack/kolla-ansible/+/916911 | 18:52 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Remove om_enable_rabbitmq_high_availability https://review.opendev.org/c/openstack/kolla-ansible/+/940690 | 18:52 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Don't allow quorum queues to be disabled https://review.opendev.org/c/openstack/kolla-ansible/+/940691 | 18:52 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Upgrade RabbitMQ docs for the Epoxy release https://review.opendev.org/c/openstack/kolla-ansible/+/940692 | 18:52 |
kevko | frickler: aaaa, yeah ..you are right ... I wanted ..but it was considered as feature :D | 18:53 |
kevko | frickler: but i am in .. you have my vote :) | 18:53 |
frickler | kevko: I'll try to do a stripped down version that might be acceptable (mainly only the backup fix) | 20:33 |
frickler | mnasiadka: 944194 is failing again ... I guess I should just force-merge it to unblock CI? | 20:34 |
mikal | frickler: on https://review.opendev.org/c/openstack/kolla-ansible/+/944184, I removed all examples of dashes for lists because https://sublime-and-sphinx-guide.readthedocs.io/en/latest/lists.html does not list it as supported formatting. | 20:34 |
frickler | mikal: hmm. I'm not completely opposed to changing it, I was just thinking it is not needed if it renders the same. maybe wait what other reviewers say about it, then | 20:39 |
mnasiadka | frickler: I see in each iteration at least one upgrade job is timing out - I guess we need to investigate that - I’m fine with force merge if you want to do that | 23:33 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!