Kubernetes Troubleshooting Commands Every DevOps Engineer Must Know

Kubernetes Troubleshooting Commands Every DevOps Engineer Must Know

Day in day out from debugging crashing pods to scaling deployments, we the Kubernetes practitioners are tied to running commands.

But here's the catch: Kubernetes commands and logs are inseparable.

While commands help you interact with the cluster, logs are the guiding light in the right direction.

However, many practitioners jump straight into commands without touching on the basics of Kubernetes logs - where they’re stored, how they’re structured, and how to locate the right log for effective troubleshooting.

Here, I’ve made the image below that illustrates the Kubernetes log directory structure - this is your starting point for any troubleshooting journey.

Top Kubernetes Troubleshooting Commands Every DevOps Engineer Must Know

kubectl get pods --all-namespaces: Check pod statuses across all namespaces.

kubectl describe pod <pod_name>: Gather detailed information about a specific pod.

kubectl logs <pod_name> -c <container_name>: View logs of a specific container.

kubectl get events --sort-by='.metadata.creationTimestamp': Review recent events for errors.

kubectl get nodes: Verify the health and status of cluster nodes.

kubectl drain <node_name> --ignore-daemonsets: Safely evacuate pods from a node.

kubectl cordon <node_name>: Mark a node as unschedulable.

kubectl uncordon <node_name>: Mark a node as schedulable again.

kubectl delete pod <pod_name> --grace-period=0 --force: Forcefully delete a crashed pod.

kubectl rollout undo deployment <deployment_name>: Roll back a problematic deployment.

kubectl exec -it <pod_name> -- /bin/sh: Access a container for debugging.

kubectl get componentstatuses: Check the health of core cluster components.

kubectl top nodes: Monitor node resource usage for bottlenecks.

kubectl top pods --all-namespaces: Identify resource-hungry pods.

kubectl delete node <node_name>: Remove a failed node from the cluster.

kubectl get ingress: Verify ingress resources and their statuses.

kubectl describe <resource_type> <resource_name>: Detailed information for a resource.

kubectl port-forward <pod_name> <local_port>:<remote_port>: Forward a local port to a pod.

kubectl get endpoints <service_name>: Verify service endpoints.

kubectl apply -f <backup.yaml>: Restore configurations from a backup manifest.

kubectl taint nodes <node_name> key=value:NoSchedule: Prevent scheduling on a problematic node.

kubectl debug <pod_name>: Launch an ephemeral container for troubleshooting.

kubectl edit <resource_type> <resource_name>: Manually modify resource configurations.

kubectl proxy: Start a proxy to the Kubernetes API for debugging.