In this session, we'll delve into the thrilling tale of a passionate new engineer who unwittingly throws a wrench into a Kubernetes-based production environment. Through the session, you'll discover how chaos engineering can be a game-changer in identifying and tackling vulnerabilities within intricate cloud-native ecosystems.

In the presentation, we use ChaosToolkit to test system redundancy by deleting some pods and by introducing network failures. Logs, metrics, and traces through OpenTelemetry allow for analysis. We delve into tools like Prometheus and Perse but also tackle the question: what is the minimum amount of telemetry data that we need to understand our system quickly and recover from failure?

Failures are unavoidable and chaos engineering prepares teams to handle them more effectively. Join a session of planned chaos and see how it fosters a better understanding of Kubernetes system components and observability tooling while promoting collaboration and communication.