Skip to content

Building in the Pipeline

Your pipeline runs, but so far it only prints a message. Time for the first real station on the belt: building your Docker image automatically on every push. 🏗️

What we will do (in very simple steps)

  1. Learn how a job gets your code
  2. Replace the "hello" step with a real build
  3. Watch GitHub build your image for you

First, a new idea: getting your code

Remember that every job starts on a fresh, empty machine. It has Docker installed, but it does not have your code yet. Before you can build, you have to copy your repository onto the runner.

GitHub provides a ready-made action for this. That introduces the second kind of step:

  • run: runs a shell command you write
  • uses: runs a prebuilt action someone else wrote

The action actions/checkout fetches your repository onto the runner. You will use it at the start of almost every job.

Step 1: Update your workflow

Open .github/workflows/ci.yml and replace its contents with this:

name: CI

on: push

jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - name: Check out the code
        uses: actions/checkout@v4

      - name: Build the image
        run: docker build -t snapshot-app .

What changed:

  • The job is now called build.
  • The first step uses actions/checkout@v4 to copy your repo onto the runner. The @v4 pins the version, so it does not change unexpectedly.
  • The second step runs the exact same docker build command you know from the Docker course.

Step 2: Push and watch

git add .
git commit -m "Build the image in the pipeline"
git push

Open the Actions tab on GitHub and click into the new run. Expand the Build the image step, and you will see Docker working through your Dockerfile, step by step, on GitHub's machine. When it finishes with a green tick, your image built successfully in the cloud, with no effort from you. 🎉

From now on, every push automatically confirms that your image still builds. If someone breaks the Dockerfile, the pipeline goes red immediately.

✅ Checkpoint

You are ready for the next lesson if:

  • Your workflow has a checkout step followed by a docker build step
  • The Actions run completes with a green tick
  • The build log shows Docker working through your Dockerfile

🩹 Common hiccups

  • "failed to read dockerfile" or "no such file": you are missing the checkout step, so the runner has no code. Make sure actions/checkout@v4 comes first.
  • The build fails at a specific step: the log points at the exact instruction. It fails here for the same reasons it would fail on your machine, so fix the Dockerfile and push again.
  • YAML error before anything runs: check indentation. Every level uses two spaces, and never a tab.

Next up: Testing in the Pipeline, where you add automated tests so the pipeline checks your app actually works, not just that it builds.