When: Weekly on Wed, 9:45 – 10:15am
Notes: KubeVirt CI SIG meeting notes
Attendees: nirdothan, dywhite, fossedihelm
Reminders:
we will create GitHub issues for tracking
GitHub issues and PRs
should be marked with /sig ci and /kind flake if applicable
should be marked with the target sig
Topics:
[urgent]
review and merge the quarantine PRs in 2 working days without SIG lgtm.
Quarantine flaky ClusterProfiler pprof subresource access test
[dhiller] has some failures in the lanes probably due to cache issues, need to investigate those
[Nir] please unhold https://github.com/kubevirt/kubevirt/pull/17536
[dhiller] will do
previous action items
state of existing issues: https://github.com/kubevirt/kubevirt/issues?q=is%3Aissue+is%3Aopen+label%3Akind%2Fflake+sort%3Aupdated-asc+label%3Asig%2Fci
n/a
[non-urgent]
[Nir] Please LGTM removal of quarantine https://github.com/kubevirt/kubevirt/pull/17840
suggestion by fossedihelm - set test to Pending (PEntry in this case)
[dhiller] will add that to the proposal of test lifecycle
[fossedihelm] Where are we about `Mandatory pre-submit lanes`?
[dhiller] currently at always_run: true, optional: false for 1.36 lanes
Look at flakes
sig-compute
[rfe_id:273][crit:high][vendor:cnv...@redhat.com][level:component][sig-compute]VMIlifecycle [rfe_id:273][crit:high][vendor:cnv...@redhat.com][level:component]Creating a VirtualMachineInstance [test_id:1622]should log libvirtd logs
being investigated, might be an intermittent failure
[sig-compute] Infrastructure cluster profiler for pprof data aggregation when ClusterProfiler configuration is enabled it should allow subresource access
quarantine pending, see above
[sig-compute] vitiofs config volumes With a ServiceAccount defined Should be the namespace and token the same for a pod and vmi with virtiofs
failing on 1.36 only, issue exists, see above
sig-monitoring
[sig-monitoring]Component Monitoring Up metrics VirtControllerDown should be triggered when virt-controller is down
failures might be fallout from working on fixing the flakes - Joao is working on that IIRC
sig-storage
Look at held tests:
dequarantine tests:
look at list of quarantined tests
Count: xx tests in quarantine currently
(insert screenshot from ci-health)
check status, i.e. who is working on those
look at PRs that want to fix flakes
see whether we can dequarantine tests
misc topics
dhiller will be out next Monday, public holiday in Germany (Pentecost)
Dylan will step in for the meeting
[Nir] failures in sig-network with network injecting mutating webhook
not on a specific test
flakes randomly
Action items
update/create issues with latest flakes spotted
communication
send meeting notes to kubevirt-dev, bcc sig people for spotted flakes (include meeting changes for upcoming instances)
Kind regards,
Daniel Hiller
He / Him / His
Principal Software Engineer, KubeVirt CI, OpenShift Virtualization
![]() |
Red Hat GmbH, Registered seat: Werner von Siemens Ring 12, D-85630 Grasbrunn, Germany Commercial register: Amtsgericht Muenchen/Munich, HRB 153243, Managing Directors: Ryan Barnhart, Charles Cachera, Avril Crosse O'Flaherty