| opendevreview | Michal Nasiadka proposed openstack/kolla master: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992483 | 06:56 |
|---|---|---|
| opendevreview | Michal Nasiadka proposed openstack/kolla master: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992483 | 07:02 |
| mikal | mnasiadka: I think this was the week you were in meetings? In an effort to make it easy -- https://review.opendev.org/c/openstack/ansible-collections-openstack/+/992526 (the openstacksdk version thing) now passes CI. It was a bit of an adventure to get there. | 07:08 |
| mikal | mnasiadka: also, I tested those other COPR packages and they don't work for us because they're for Fedora not Rocky. So https://review.opendev.org/c/openstack/kolla/+/986283 is as good as it gets in terms of Rocky support for SPICE in Kolla right now. I think its failing CI because of the other patch though, so no rush. | 07:08 |
| mnasiadka | mikal: yeah, I asked on #openstack-ansible-sig to prioritise merging and releasing a-c-o | 07:30 |
| mikal | Oh cool, I didn't even realise there was a channel for it! | 07:32 |
| opendevreview | Pierre Riteau proposed openstack/kolla master: Install openstacksdk using upper constraints https://review.opendev.org/c/openstack/kolla/+/992591 | 07:32 |
| opendevreview | Bartosz Bezak proposed openstack/kayobe master: CI: add Rocky 10 aarch64 jobs https://review.opendev.org/c/openstack/kayobe/+/992288 | 09:07 |
| *** jhorstmann is now known as Guest11104 | 09:31 | |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2025.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992602 | 10:03 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2025.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992602 | 10:03 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2025.2: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992603 | 10:03 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2026.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992604 | 10:04 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2026.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992604 | 10:04 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2026.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992604 | 10:04 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992483 | 10:05 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2025.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992602 | 10:07 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2025.2: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992603 | 10:07 |
| frickler | mnasiadka: was https://review.opendev.org/c/openstack/kolla/+/992483 a misdirected cherry-pick or did you just forget to adapt the commit message? | 10:12 |
| mnasiadka | I didn’t adapt commit msg, let me do that | 10:13 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992483 | 10:14 |
| mnasiadka | We still need to merge 2025.1 first because of upgrade jobs | 10:14 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2025.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992602 | 10:15 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2025.2: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992603 | 10:15 |
| opendevreview | Michal Nasiadka proposed openstack/kolla stable/2026.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992604 | 10:16 |
| frickler | I wouldn't mind force-merging these to avoid another cycle of waiting for each branch in turn (and likely dealing with umpteen rechecks for unstable jobs in addition) | 10:33 |
| mnasiadka | frickler: seems the debian builds have some package discrepancy problems - so if you have time to do that - go on - I’m stuck in meetings unfortunately :( | 10:36 |
| frickler | yes, I saw these, I already checked the mirror update logs, but they seem sane to me, so maybe some issue in the upstream mirrors. going to invoke the force now, please step back ;) | 11:39 |
| opendevreview | Merged openstack/kolla master: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992483 | 11:46 |
| opendevreview | Merged openstack/kolla stable/2026.1: kolla-toolbox: Pin openstacksdk to 4.14.0 https://review.opendev.org/c/openstack/kolla/+/992604 | 11:47 |
| frickler | it looks like the fix is wrong for 2025.x since we need to have older openstacksdk there anyway? see e.g. https://aaeb0a9024d6da4fc3f8-f9c86595ee722ed5765b04c30efd2d71.ssl.cf5.rackcdn.com/openstack/c1a073319cc84bf7b73d171e8a0d7ad3/kolla/build/000_FAILED_kolla-toolbox.log | 11:51 |
| frickler | do we have any failure for those branches due to sdk and not trixie repo messup? | 11:52 |
| kevko | frickler, mnasiadka: what about patch openstack.sdk code inside ? | 12:37 |
| kevko | with kolla-patch ? | 12:37 |
| jwitko | Hello all, Is the openstack-kolla meeting taking place? From what I read it takes place in this IRC channel? | 13:05 |
| Vii | „Due to my and some other people unavailability - I’m cancelling tomorrow’s weekly meeting. | 13:07 |
| Vii | See you next week!” | 13:07 |
| Vii | [kolla] Weekly meeting on 10th June 2026 cancelled | 13:08 |
| jwitko | I see, thanks Vii | 13:08 |
| Vii | np | 13:09 |
| opendevreview | Piotr Milewski proposed openstack/kolla master: Add YAML support to kolla_set_configs engine https://review.opendev.org/c/openstack/kolla/+/982275 | 13:10 |
| *** jwitko1 is now known as jwitko | 13:14 | |
| opendevreview | Michal Arbet proposed openstack/kolla master: Temporarily fix kolla-toolbox because of openstacksdk version https://review.opendev.org/c/openstack/kolla/+/992640 | 13:26 |
| mnasiadka | frickler: I abandoned 2025.1 and .2, surely there’s older openstacksdk | 13:50 |
| dcapone2004 | mnasiadka: So I manually applied that patch to my kolla installation and it did not correct the issue. I am unsure if that means it is a bigger "bug" or if my issue is back to my original thoughts that I have an issue with the bridge configuration and I do not know how that may have been lost | 14:59 |
| dcapone2004 | mnasiadka: I should correct my statement slightly. My deploy got past where it was most recently hung up AFTER I stopped the cluster to redeploy after my initial reconfigure failed. Now I am back to where mariadb does not respond on the VIP, but I can connect to the Mariadb on all controller nodes using their direct IPs | 15:03 |
| mnasiadka | dcapone2004: for this you would need https://review.opendev.org/c/openstack/kolla-ansible/+/992397 (and the Kolla depends-on) or just disable the track script by setting keepalived_track_script_enabled: "no" | 15:04 |
| dcapone2004 | how I got here. Did a deploy with KA using stables/2026.1 branch. At the time the version number ended with .dev3. Deploy went fine except there was an issue with Horizon and glance and uplaoding images. worked in skyline and via cli, so figured possible container configuration issue. Updated KA with pip install -U from git. KA version updated to .dev6. Ran reconfigure | 15:05 |
| dcapone2004 | got stuck on this VIP issue with Mariadb | 15:05 |
| dcapone2004 | yeah i applied that fix. The script is updated in the keepalived container as expected. check_alive.sh still fails at proxysql | 15:08 |
| dcapone2004 | I think the issue is that when I enter the container via bash and run the command line command proposed in the fix, it returns a mariadb command not found | 15:08 |
| mnasiadka | dcapone2004: because the depends-on in Kolla installs mariadb in keepalived container | 15:11 |
| dcapone2004 | mnasiadka: Does that require a separate configuration change somewhere in a different patch? For clarity, I applied the "patch" by editing the ansible/roles/loadbalancer/templates/keepalived/check_alive_proxysql.sh.j2 file directly, and then running deploy. I did not make the change directly in the container. I only entered the container to verify that the container script updated as expected. | 15:14 |
| mnasiadka | Well, either you rebuild the container on your own (probably easier) - or install mariadb-client inside the container (but the repo definition for mariadb is surely not there) | 15:14 |
| dcapone2004 | got it, basically your saying because the patch is not merged, the current container image on quay.io does not have that dependancy in it yet. Once the patch merges, I would assume the quay.io image will update? | 15:16 |
| mnasiadka | Yes, we run publish jobs nightly | 15:16 |
| mnasiadka | (Well, nightly Europe time) | 15:16 |
| dcapone2004 | great thanks. also since, I ave your attention, one last question. new KA at prechecks seems to throw that warning about using the quay.io images. Is that because of essentially what you just said that they are updates nightly and might not be suitable for production so better to keep your own repo or is that because 2026.1 is still RC? | 15:18 |
| Vii | TLDR: quay images are intended for CI testing, for your own environment you should build images from scratch | 15:20 |
| mnasiadka | Basically there was some communication, that we’re doing this - because we don’t care for these images as in watch for security bugs in them and so on - we use them for CI runs and they are rather suitable for development purposes only - you should build your own locally for production purposes with proper security scanning and remediation measures | 15:20 |
| dcapone2004 | got it. good information | 15:22 |
| opendevreview | Grzegorz Koper proposed openstack/kolla-ansible master: neutron: Split neutron role into neutron and neutron-agents https://review.opendev.org/c/openstack/kolla-ansible/+/975614 | 16:35 |
| opendevreview | Grzegorz Koper proposed openstack/kolla-ansible master: neutron: Split neutron role into neutron and neutron-agents https://review.opendev.org/c/openstack/kolla-ansible/+/975614 | 16:43 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Centralize service log directory management https://review.opendev.org/c/openstack/kolla-ansible/+/989850 | 21:39 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Deploy MariaDB before configuring loadbalancer https://review.opendev.org/c/openstack/kolla-ansible/+/992478 | 21:49 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Centralize service log directory management https://review.opendev.org/c/openstack/kolla-ansible/+/989850 | 21:49 |
| opendevreview | Pierre Riteau proposed openstack/kayobe master: Fix paths to EPEL repositories https://review.opendev.org/c/openstack/kayobe/+/992770 | 21:50 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: kolla-build: Use SandboxedEnvironment to prevent SSTI https://review.opendev.org/c/openstack/kolla/+/992777 | 23:20 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: Revert "kolla-toolbox: Pin openstacksdk to 4.14.0" https://review.opendev.org/c/openstack/kolla/+/992780 | 23:42 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: kolla-build: Use SandboxedEnvironment to prevent SSTI https://review.opendev.org/c/openstack/kolla/+/992777 | 23:45 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: kolla-build: Use SandboxedEnvironment to prevent SSTI https://review.opendev.org/c/openstack/kolla/+/992777 | 23:45 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!