Node Unavailable
Error message
Node is NotReady
Quick summary
A node in the cluster is down or not ready, reducing capacity and potentially causing Pod failures.
Why this happens
The node's kubelet is not running or has lost connection to the API server
The node has run out of disk space (evicted)
Hardware failure or network issues on the node
The node was cordoned or drained for maintenance
Minimal example
$ kubectl get nodes
NAME STATUS ROLES AGE VERSION
node-1 NotReady worker 30d v1.28.0# SSH into node and restart kubelet
ssh node-1
sudo systemctl restart kubelet
# Verify
kubectl get nodes # node-1 should be ReadyThe node is NotReady, likely because kubelet is not running. The fix restarts kubelet on the node.
How to diagnose
Check node status:
kubectl get nodes— look for NotReadyDescribe the node:
kubectl describe node <node>for events and conditionsSSH into the node and check kubelet:
systemctl status kubeletCheck node disk space:
df -hon the node
How to fix
Restart the kubelet on the node:
systemctl restart kubeletFree up disk space on the node if it is full
Reboot the node if it is in a bad state
Uncordon the node if it was cordoned:
kubectl uncordon <node>
How to prevent
Monitor node health and alert on NotReady status
Set up cluster autoscaling to replace unhealthy nodes
Regularly check disk space and clean up old logs/images