This is a clone of issue OCPBUGS-49606. The following is the description of the original issue:
—
This is a clone of issue OCPBUGS-48250. The following is the description of the original issue:
—
Trying to solve the root issue from this bug: https://issues.redhat.com/browse/OCPBUGS-39199?focusedId=26104570&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-26104570 

To fix this, we need each of the sync functions to be able to individually clear a CO degrade that they have set earlier. Our current flow only clears a CO degrade when the all of sync functions are successful and that tends to be problematic if they happen to get stuck in one of the sync functions. We typically see this for syncRequiredMachineConfigPools, which waits until the master nodes have finished updating during an upgrade.