Question

A Deployment named 'analytics' has one replica stuck in Pending. Running 'kubectl describe pod analytics-7d9f' shows the event: '0/4 nodes are available: 2 node(s) had untolerated taint {node-role.kubernetes.io/control-plane: }, 2 node(s) didn't match Pod's node affinity/selector.' The pod uses a requiredDuringSchedulingIgnoredDuringExecution nodeAffinity rule that selects nodes labeled 'zone=us-east-1a'. You confirm all 4 nodes exist and are Ready. What is the correct action to get the pod scheduled?

Accepted Answer

Label at least one schedulable worker node with 'zone=us-east-1a' so it satisfies the pod's required node affinity

Answer

Add a toleration for the control-plane taint to the pod spec so it can run on the control-plane nodes

Answer

Increase the CPU and memory requests on the pod so the scheduler prioritizes it

Answer

Restart the kube-scheduler static pod to clear its scheduling cache