Artifacts: Reports, Screenshots, Logs and Traces

📌 Definition:

📖 Detailed Explanation:

Since CI failures can't be reproduced by re-running interactively, artifacts are how you debug them. On a failing E2E run you want the screenshot at failure, the Playwright trace, the console/server logs and the HTML report — captured automatically and uploaded so any reviewer can open them. The crucial detail is conditional upload: you must upload artifacts even when (especially when) the tests fail, using if: always() or if: !cancelled(), because the default is to skip subsequent steps after a failure. Artifacts also transfer outputs between jobs (build → deploy). Manage retention deliberately: artifacts consume storage, so set sensible retention-days and avoid uploading gigabytes of video on every green run — capture heavy artifacts only on failure or first retry.

🔑 Key Points:

• Artifacts persist files after a job for download or hand-off to later jobs
• They are the primary way to debug non-reproducible CI failures
• Upload on failure too: guard with if: always() / if: !cancelled()
• Capture heavy artifacts (video/traces) only on failure or first retry
• Set retention-days to control storage cost

🌍 Real-World Example:

📎 Code Example:

- name: Run E2E
  run: npx playwright test

- name: Upload report + traces (even on failure)
  if: ${{ !cancelled() }}
  uses: actions/upload-artifact@v4
  with:
    name: playwright-report
    path: |
      playwright-report/
      test-results/
    retention-days: 7

# In playwright.config: capture heavy artifacts only when needed
#   use: { trace: 'on-first-retry', screenshot: 'only-on-failure', video: 'retain-on-failure' }