Question

A worker node named worker-2 has been showing status NotReady for the last 10 minutes. Pods scheduled to it are stuck and new pods will not schedule there. You SSH into worker-2 and run 'systemctl status kubelet', which reports the service as 'active (running)', but 'kubectl get nodes' from the control plane still shows worker-2 as NotReady. The kubelet logs (journalctl -u kubelet) repeatedly show: 'Container runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:cni plugin not initialized'. What is the most likely root cause you should investigate first?

Accepted Answer

The CNI network plugin (e.g., the pod-network DaemonSet) is not running or not installed on the node, so the kubelet cannot report the node as Ready

Answer

The kubelet client certificate has expired and must be renewed with kubeadm before the node can rejoin the cluster

Answer

The kube-apiserver is unreachable from worker-2, so the node heartbeat is not being received

Answer

The node has insufficient memory and the kubelet has triggered a MemoryPressure eviction that marks the node NotReady