Skip to main content
Tracelit’s backend SDKs speak OpenTelemetry natively. Drop in the SDK and every inbound request, outbound call, and database query becomes a span — stitched into a single distributed trace across all your services.

Automatic instrumentation

Call one function and the SDK instruments your entire runtime. No per-library config.

Manual spans

Add custom spans anywhere in your code to trace business-level operations.

Span kinds

Use the correct span kind so Tracelit renders the right context in the trace view.

Context propagation

Trace context is propagated automatically across service boundaries using standard W3C TraceContext headers. Every SDK injects and extracts these headers on all instrumented HTTP and gRPC calls with no extra code. For manual propagation (e.g. message queues or async jobs):

Error guarantee

Tracelit’s SDKs guarantee that error spans are always exported — even when the parent trace falls outside your configured sample rate. This ensures no error goes undetected regardless of sampling settings.
Error spans bypass the sampler entirely. Setting sampleRate: 0.1 keeps 10% of normal traces but 100% of error spans always reach Tracelit.

Viewing traces

All traces are available in the Traces tab of your service in the Tracelit dashboard. You can:
  • Filter by service, environment, status, and time range
  • Drill into any trace to see the full span tree with timings and attributes
  • Jump directly from a trace to the linked incident or error

Learn more