Machine Config Operator Force Node Drain after grace period exceeded in RHOCP4
Issue
- Machine Config Operator (MCO) is unable to drain a node after a grace period exceeded.
- Pod eviction fails due to disruption budget violation.
- Drain process retries after failure, but continues to fail after a certain duration.
- Desired change: Configure a global grace time / timeout for MCO to wait for a node to drain before forcing it.
Environment
- Red Hat OpenShift Container Platform (RHOCP)
- 4.12
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
