Skip to main content

15. Lab: repair a failing workload

Hands-on — diagnose and fix a Deployment stuck in a crash/pull loop.

⏱ ≈ 25 min this module · 📚 ≈ 6.9 h full track

🔒 Unlock this course to read the material and run the labs.

See options to unlock →
🔒 The written material unlocks with the course.
🧪 Hands-on lab Medium · 170 XP · 25min

Goal

A Deployment broken in your lab namespace is not coming up. Diagnose it (kubectl describe, kubectl logs, kubectl get events) and fix it so all replicas are Ready.

There are two faults: a bad image tag, and a failing readiness probe pointing at the wrong port.

Done when

  • Deployment broken has all replicas Ready on a valid image

🔒 Unlock the course to run this lab.