163 Tracing and Context Propagation

163 Tracing and Context Propagation

Tracing makes distributed latency visible by preserving causal context across boundaries.

Propagation Path

gateway -> service A -> service B -> datastore
   context + trace headers propagated at each hop

Common Failure Mode

A single missing propagation point breaks end-to-end trace visibility. This often happens in background goroutines or manual HTTP client construction.

Design Guidance

• Accept context.Context at API boundaries.
• Pass context through outbound calls without replacement.
• Record operation-level spans around DB and external dependencies.

Outcome

Correct tracing turns unknown latency into attributable latency.