(Feel free to update this bug's summary to be more specific.)
Component Readiness has found a potential regression in the following test:

[sig-arch] events should not repeat pathologically

Extreme regression detected.
Fishers Exact probability of a regression: 100.00%.
Test pass rate dropped from 100.00% to 46.43%.

Sample (being evaluated) Release: 4.21
Start Time: 2025-09-03T00:00:00Z
End Time: 2025-10-01T12:00:00Z
Success Rate: 46.43%
Successes: 13
Failures: 15
Flakes: 0
Base (historical) Release: 4.19
Start Time: 2025-05-18T00:00:00Z
End Time: 2025-06-17T00:00:00Z
Success Rate: 100.00%
Successes: 79
Failures: 0
Flakes: 0

View the test details report for additional context.

You can see the operator blipping degraded at a regular cadence in this chart

The full error is:

{  1 events happened too frequently

event happened 24 times, something is wrong: clusteroperator/cluster-api hmsg/dbc7a25cc8 - reason/Status degraded error generating kubeconfig: token can't be empty (15:39:29Z) result=reject }

This is failing routinely in this multiarch job, but I can't seem to find any hits on amd64.

Note that 4.19 looks fully passing, but 4.20 looks equally bad, meaning we need to know ASAP how serious this is and if we can still GA 4.20 with it.

Filed by: dgoodwin@redhat.com