When a pivot error happens (MCO cannot apply a new osImage), the MCP should be degraded and a PivotError alert should be triggered.

We can see the alert being triggered, but the MCP is not correctly degraded, it remains in "working" status instead.

It looks like that eventually, after a good amount of time (15 minutes in our cluster), it is fixed.

{noformat}
E1013 11:57:13.938404    2596 writer.go:231] Marking Degraded due to: "Failed to update OS to quay.io/mcoqe/layering@sha256:5177a092968e50b2be8d98c15a68bc65016de18dacfc693f99187d2a1457ac85 after retries: timed out waiting for the condition"
I1013 11:57:13.950238    2596 daemon.go:784] Transitioned from state: Done -> Working


I1013 12:12:16.822035    2596 daemon.go:784] Transitioned from state: Working -> Degraded
{noformat}


    

4.20.0-0.nightly-2025-10-13-053645
    

Frequently
    

    1. Break the "rpm-ostree upgrade" using the steps defined in https://polarion.engineering.redhat.com/polarion/#/project/OSE/workitem?id=OCP-63866

    2. Apply a new osImage to the worker pool
    
    

The worker pools should be degraded and a pivot alert should be raised. The alert is properly raised but MCO is not able to properly degrade the worker MCP.


    

The worker MCP should be properly degraded with the right message.