Tuesday, 2026-10-13

*** mrunge_ is now known as mrunge00:06
opendevreviewTakashi Kajinami proposed openstack/aodh master: Rename helper to get notification transport  https://review.opendev.org/c/openstack/aodh/+/101010303:19
opendevreviewTakashi Kajinami proposed openstack/aodh master: Rename helper to get notification transport  https://review.opendev.org/c/openstack/aodh/+/101010303:20
opendevreviewTakashi Kajinami proposed openstack/ceilometer master: Rename helper to get notification transport  https://review.opendev.org/c/openstack/ceilometer/+/101010403:24
tobias-urdintkajinam: thanks!06:54
tkajinamFatal glibc error: CPU does not support x86-64-v310:12
tkajinamhmm it seems sg-core container is not runnable in some of our CI Nodes.10:12
jwysoglayep, I'm actually looking at this right now10:13
tkajinamI hit that problem a few days ago and recheck "fixed" the problem likely due to rescheduling to a different node, but we might need to find a way to avoid that problem more effectively10:13
jwysoglathe base image for the sg-core image was recently bumped from ubi9 to ubi10. We tested the new image on centos, but I guess us running it in ubuntu may be part of the issue10:13
tkajinamjwysogla, I'd rather suspect old vcpu model in some cloud provides but I might be wrong10:14
jwysoglaoh, so it's more random then10:14
tkajinamyeah that's my concern10:14
tkajinamI think c10s removed x86-64-v2 support (I remember I saw that in some articles about cern moving to debian for old cpu compatibility)10:15
jwysoglayea I actually read something like that as well at some point10:15
tkajinamProvider: opendev.org%2Fopendev%2Fzuul-providers/rax-ord-main10:16
tkajinamthis is where it failed initially, for record10:16
jwysoglatkajinam, we're thinking of modifying the build and push pipelines of sg-core to automatically build another image based on top of ubi9 and tag it :ubi9-latest. I'll modify the sg-core devstack plugin as well to use that image, so we should get a working image automatically once this all is done.10:37
tkajinamjwysogla, thanks. I saw jlarriba also mentioned the same in gerrit10:39
mnasiadkatkajinam: Rackspace classic has x86-64-v2 cpus (but still that’s a big chunk of available capacity of OpenDev)10:43
tkajinammnasiadka, ah, thanks for that info. noted10:44
jwysoglatkajinam, we now have a :ubi9-latest tagged image available on quay, it's now the default in sg-core's devstack plugin, so it should be picked up automatically on recheck. I hope it'll all work now.14:34
tkajinamjwysogla, ah thanks !14:34
tkajinamI was wondering where that devstack plugin resides but noticed it's in sg-core repo14:35
jwysoglaIMO Ideally we'd get rid of sg-core at some point now that there is the prometheus exporter directly in ceilometer. But first we'll need to figure out how to run ceilometer in tests for all namespaces. There was an attempt to do that here https://review.opendev.org/c/openstack/ceilometer/+/975807 , but we didn't like it that much and it's currently not much of a priority. We'll see when we can 14:41
jwysoglaget back to thinking about it.14:41
tkajinamah, ok14:52
opendevreviewTakashi Kajinami proposed openstack/ceilometer master: DNM: Test SQLAlchemy 2.1  https://review.opendev.org/c/openstack/ceilometer/+/101021715:36
-opendevstatus- NOTICE: The Etherpad service at etherpad.opendev.org (also used by meetpad.opendev.org) is being restarted now to apply a security update; any open pads will display an error message until refreshed, and author colors may get reset.19:53
opendevreviewCallum Dickinson proposed openstack/ceilometer master: Fix polling load balancers with multiple meters  https://review.opendev.org/c/openstack/ceilometer/+/101030623:22

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!