Redis Enterprise pods are in terminating state due to preemption

Question

Why are the Redis Enterprise pods in a Kubernetes deployment in a terminating state? Pods' events show "Preempted"

Answer

This means the worker nodes don't have sufficient resources to schedule the pods, and other pods with higher priorities are preempting the RS pods. The Pod's Events will show a message similar to the following:

Events:  
Type     Reason     Age                  From               Message  
----     ------     ----                 ----               -------  
Normal   Preempted  42m                  default-scheduler  Preempted by a pod on node XYZ

To fix this, check the resources on the worker nodes and the current workload consuming them, add more resources, or terminate other pods to free up the needed resources for the RS pods. The RS pods can be configured with a higher priorityClass to prevent them from being preempted by other workloads.

References

Using priority to protect from preemption