Primary pod is in pending state for a long duration
Primary pod and Operator pod have pod anti-affinity set.
To resolve the issue of long duration pending state of primary pod
- Verify if Operator pod and Primary pod are scheduled to same node using the following commands:
kubectl get all -n <primary pod namespace> -o wide
kubectl get all -n <operator pod namespace> -o wide
- If it is allocated to same node then create new node with same node selector given in CR for primary server.
- Delete the Primary pod which is in pending state.
The newly created Primary pod must not be in pending state now.