Skip to main content

Running Your First Chaos Engineering Experiment with Litmus

How to install Litmus on Kubernetes and run a controlled failure experiment from a written hypothesis to a verdict you can act on, without breaking production by accident.

14items
Back to all checklists
Chaos EngineeringBeginner
chaos-engineeringlitmuskubernetesresilience
Progress0 / 14 completed
0%

Write down a hypothesis and a steady-state metric before touching anything

Critical

Run the first experiment in staging on a single stateless workload

Critical

Install Litmus in its own namespace with Helm

Critical

Confirm the chaos CRDs and operator are installed and healthy

Critical

Create a ServiceAccount with only the permissions the experiment needs

Critical

Install the pod-delete ChaosExperiment from ChaosHub

Add probes so the experiment knows what 'healthy' means

Critical

Write the ChaosEngine with exact label selectors and a short duration

Critical

Apply the ChaosEngine and tail the runner pod and ChaosResult

Keep your real dashboards and logs open while chaos is running

Critical

Read the ChaosResult, then delete the ChaosEngine

Increase blast radius only after a clean run

Write up the run and file tickets for whatever broke

Critical

Schedule a recurring gameday so the system stays tested

Sponsored
Carbon Ads

More checklists

Also worth your time on this topic