Cause three probe failures and identify each from its symptom

applied · 65 min · Objective 1.5

Task

Misconfigure health probes deliberately, one fault at a time, and record the distinct signature each produces. The three faults here are the ones that cause real outages, and each looks different in a way you will recognise afterwards.

Steps

  1. Deploy the application with correct liveness and readiness probes as a baseline. Record a normal rollout in lab/probes.md.
  2. Fault one: set the liveness probe's initial delay far shorter than the application's start time. Deploy, and record what you observe -- the restart count, whether it ever becomes ready, and what the application log does or does not show.
  3. Fix it with a startup probe, and record that the restart loop stops.
  4. Fault two: point the LIVENESS probe at a dependency, then make the dependency fail. Record how many replicas restart and what that turns a dependency outage into.
  5. Fault three: remove the readiness probe entirely and perform a rolling update. Record whether requests fail during the rollout, and write the one-line rule each of the three faults teaches.

Verify

grep -Eci 'restart' lab/probes.md
grep -Eci 'startup probe' lab/probes.md
grep -Eci 'dependency' lab/probes.md
grep -Ec 'rule|lesson' lab/probes.md

All four non-zero, and the file must record three distinct signatures. If your notes describe the same symptom three times, the faults were not isolated from each other.

This is an independent study companion for CompTIA Cloud+ CV0-004 and is not produced by or endorsed by CompTIA.