In the world of microservices, failures are inevitable—and figuring out why can feel like chasing shadows. Traditional debugging fails when requests span dozens of services, and you can't pause production to inspect what's happening.

In this demo-driven session, we'll walk through real-world failures and show how to make sense of the chaos using modern observability tools. We'll show basic logging, leverage Linkerd for problematic service interactions, utilize OpenTelemetry for end-to-end tracing and pinpointing failure points, and employ Prometheus to understand system health via metrics.

While we focus on these specific tools, the troubleshooting strategies shared are universally applicable to any cloud-native stack. Whether you're a platform engineer building reliable infrastructure or a developer fighting production fires, you'll gain concrete skills to make your microservices more observable and your on-call rotations much more manageable.