Description of problem: The test "[bz-Machine Config Operator] Nodes should reach OSUpdateStaged in a timely fashion" is failing due to missing OSUpdateStarted events that should have been recorded by the openshift-tests watcher, but weren't. We believe this is a defect somewhere in the disruption monitoring framework but are unsure where. The events exist when queried at the end of CI in gather-extra/must-gather. How reproducible: Rare but happening daily several times. https://search.ci.openshift.org/?search=but+no+OSUpdateStarted+event+was+recorded&maxAge=48h&context=1&type=bug%2Bjunit&name=&excludeName=&maxMatches=5&maxBytes=20971520&groupBy=job To fix we will break up the test, stop failing the existing one when we see these events missing, and add a new test which will flake when this happens so we can track it clearly and separately.
For the purposes of this fix, the new test is live and the old is no longer failing on this problem. Next steps will be to actually fix the missing events but for purposes of this bug we're good.