Write Dockerfiles and Build Custom Images

A Dockerfile is a plain-text recipe that tells Docker how to assemble a container image, instruction by instruction. Each line (FROM, COPY, RUN, and CMD) becomes a stacked, content-addressed layer, and Docker's build engine BuildKit reuses any layer whose inputs have not changed since the last build. This is what makes container builds reproducible across machines and dramatically faster on repeat runs.

Writing Dockerfiles by hand is the foundational skill that everything else in the container ecosystem (compose stacks, Kubernetes manifests, CI/CD pipelines, registry workflows) sits on top of. Sloppy ordering produces 800 MB images that take three minutes to push; careful ordering produces 25 MB images that rebuild in seconds. This lab uses a small static website as the canvas to give you a working mental model for how BuildKit assembles, caches, and invalidates layers - so that the Dockerfiles you write next, for your own applications, are fast, small, and predictable from the start.

Objectives

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

• Author a Dockerfile and build a tagged custom image using FROM, WORKDIR, COPY, EXPOSE, and CMD.
• Run the built image as a container and verify it serves the expected content over HTTP.
• Inspect image layers with docker image history and observe how edits invalidate the BuildKit cache.
• Use a .dockerignore file to control what enters the build context.

Who is this lab for?

This lab is designed for:

• Engineers and students taking their first hands-on steps with Docker
• DevOps practitioners who want to write smaller, faster, more cache-friendly images for their CI/CD pipelines