Amazon EKS — Troubleshooting
EKS issues span cluster access, pod scheduling, networking, and node health. Systematic debugging — starting from the right layer — saves hours.
Topics Covered
| # | Topic | Description |
|---|---|---|
| 1 | Cluster Access Issues | Authentication and authorization errors |
| 2 | Pod & Workload Failures | ImagePullBackOff, CrashLoopBackOff, Pending |
| 3 | Networking Issues | DNS, connectivity, load balancer problems |
| 4 | Node Issues | NotReady, join failures, disk pressure |
| 5 | Troubleshooting Decision Tree | Systematic debugging flowchart |
| 6 | Hands-On Lab | Debug common EKS failures |
| 7 | Practice Questions | Test your debugging skills |
First Responder Commands
# Cluster health
kubectl cluster-info
kubectl get nodes
kubectl get pods --all-namespaces --field-selector status.phase!=Running,status.phase!=Succeeded
# Recent events
kubectl get events --sort-by=.lastTimestamp -A | tail -20
# System pods
kubectl get pods -n kube-system
