microservices

Distributed Tracing (Sleuth and Zipkin).

Why needed?

In Microservices, a single request might traverse 10+ services. Without tracing, it's impossible to know where a delay or error occurred.

Key Concepts:

  • Trace ID: A unique ID for the entire request journey.
  • Span ID: A unique ID for a specific segment/hop (service call) within that journey.

Implementation:

  • Spring Cloud Sleuth / Micrometer: Automatically adds Trace IDs and Span IDs to logs and headers.
  • Zipkin: A visual UI to see the timing and flow of these traces across services.
Distributed Tracing (Sleuth and Zipkin). | DevExCode