Nodes and Clusters

• Node: machine (physical/VM) where Kubernetes runs; worker that hosts Pods (formerly called minion)
• Cluster: group of nodes; failed node → workloads continue on others (HA + load distribution)
• Control plane (master): monitors cluster, stores state, reschedules failed workloads
• Worker node: runs Pods via kubelet + container runtime

kubectl

• kubectl get nodes: list nodes
• kubectl get nodes -o wide: includes CONTAINER-RUNTIME column

Exam tips

• Pod IP is not stable across restarts: use Services for stable networking
• Distinguish node (infrastructure) from Pod (workload unit)