Troubleshooting Flashcards

How do you view logs for a specific container named 'nginx' in a multi-container pod named 'mypod'?

Answer: kubectl logs mypod -c nginx

The -c flag specifies the container name when a pod runs multiple containers.

A pod is stuck in the 'Pending' state. Which command provides the most useful diagnostic information?

Answer: kubectl describe pod

kubectl describe shows the Events section and conditions that reveal why a pod is pending, such as insufficient resources or unschedulable nodes.

What does the 'CrashLoopBackOff' status indicate for a pod?

Answer: The pod's container repeatedly crashes and Kubernetes is backing off restart attempts

CrashLoopBackOff means the container starts, crashes, and Kubernetes keeps restarting it with increasing exponential delays.

Which command lists events in a specific namespace sorted chronologically?

Answer: kubectl get events -n --sort-by='.lastTimestamp'

kubectl get events with --sort-by='.lastTimestamp' lists events in chronological order, making it easy to trace the sequence of recent issues.

A node shows 'NotReady' in kubectl get nodes. Which command gives the most detail about the node's condition?

Answer: kubectl describe node

kubectl describe node shows Conditions, allocatable resources, taints, and events that explain why a node is NotReady.

You need a temporary interactive debug pod to troubleshoot network connectivity in the cluster. Which command is most appropriate?

Answer: kubectl run debug --image=busybox --restart=Never --rm -it -- /bin/sh

kubectl run with --restart=Never --rm -it creates a temporary interactive pod that is automatically deleted when you exit the shell.

Where are kubelet logs found on a systemd-based Linux node?

Answer: journalctl -u kubelet

On systemd-based systems, kubelet is managed as a systemd service, so its logs are accessible via journalctl -u kubelet.