Distributed Tracing
Distributed tracing follows requests across service call chains with trace IDs and spans, visualizing latency bottlenecks and error paths.
Purpose
Understand the full request path across microservices, pinpointing cross-service latency and errors.
Mechanism
The request entry point generates a trace ID; each operation creates a span (with start/end/tags). Spans form parent-child relationships. Context propagation (W3C Trace Context). Jaeger/Zipkin store and visualize traces.
Use cases
- Microservices troubleshooting
- Latency analysis
- Request tracing