Inspect, Create, and Debug Your First Pod with kubectl

A Pod is the smallest deployable unit in Kubernetes and the wrapper that every higher-level workload (Deployment, Job, StatefulSet, DaemonSet) eventually creates on your behalf. Each Pod is a co-scheduled group of one or more containers that share a network namespace (one IP, shared localhost) and can share storage through volumes, so understanding Pods is the prerequisite to understanding everything else the platform schedules. When something is wrong in a cluster, four kubectl verbs do almost all of the diagnostic work: get reveals state at a glance, describe exposes the events the scheduler and kubelet emit while reconciling the Pod, logs shows what the container itself printed, and exec puts you inside the running container so you can poke at the filesystem and network.

In this lab, you will author a single-container Pod manifest, apply it, and watch the lifecycle transitions roll through in real time. You will deliberately break the Pod by typoing the image tag, read ImagePullBackOff events out of describe, repair the manifest, and then deploy a multi-container Pod that proves two containers in the same Pod can curl each other on localhost and share files through an emptyDir volume.

Objectives

Upon completion of this beginner level lab, you will be able to:

• Author and apply a single-container Pod manifest and observe its lifecycle transitions with kubectl get -w
• Diagnose an ImagePullBackOff failure using kubectl describe events and repair the manifest
• Inspect a running container with kubectl logs and enter it with kubectl exec -it to verify behavior
• Demonstrate that two containers in the same Pod share a network namespace and an emptyDir volume

Who is this lab for?

This lab is designed for:

• Engineers and developers brand new to Kubernetes
• SREs learning the core Pod debugging workflow
• CKAD and CKA candidates building Pod fundamentals