Docker for Reproducible Test Environments

📌 Definition:

📖 Detailed Explanation:

Test flakiness and CI-only failures often trace to environment drift: a different OS, a missing system library, a different browser or font-rendering behavior. A Docker image pins all of that. Playwright, for instance, publishes an official image with the three browsers and all their Linux system dependencies pre-installed, so a job that runs container: mcr.microsoft.com/playwright:... skips dependency installation and, crucially, matches the environment used to generate visual baselines. Beyond consistency, containers give isolation (each run is clean) and portability (the same image runs locally and in CI, so a developer can reproduce a CI failure exactly). The trade-offs are image size and build time, mitigated by layer caching and using slim base images. The interview point is that Docker converts 'the environment' from an implicit variable into an explicit, version-controlled dependency.

🔑 Key Points:

• Images pin OS libs, runtime, browsers and tools — identical everywhere
• Eliminates environment-drift flakiness and CI-only failures
• Official Playwright image ships browsers + deps and matches visual-baseline OS
• Same image runs locally and in CI, so failures reproduce exactly
• Trade-offs: image size/build time — mitigate with layer caching and slim bases

🌍 Real-World Example:

📎 Code Example:

# Run the whole job inside the official Playwright image
jobs:
  e2e:
    runs-on: ubuntu-latest
    container: mcr.microsoft.com/playwright:v1.50.0-jammy   # browsers + deps preinstalled
    steps:
      - uses: actions/checkout@v4
      - run: npm ci                 # no 'playwright install --with-deps' needed
      - run: npx playwright test

# Or a custom Dockerfile pinning the environment:
# FROM node:20-bookworm-slim
# RUN apt-get update && apt-get install -y --no-install-recommends <libs>
# WORKDIR /app
# COPY package*.json ./ && RUN npm ci
# COPY . .