Distributed Tracing

Distributed tracing in ThousandEyes provides end-to-end visibility across your application stack. It works by automatically injecting standard trace headers into HTTP Server and API tests. This lets you correlate network performance issues with application traces in a supported observability backend.

Create a ThousandEyes Test with Distributed Tracing

Before setting up any distributed tracing integration, create a ThousandEyes test with distributed tracing enabled:

1. Create an HTTP Server Test or API Test in ThousandEyes. For more information on creating a test, see Getting Started with API Test and HTTP Server Tests.

2. Under Advanced Settings, enable Distributed Tracing.

3. ThousandEyes automatically injects the following trace headers into requests:

   • b3

   • traceparent

   • tracestate

   

   Request Headers

   This lets ThousandEyes capture the trace ID of the request (for example, 0a0c3e94ca224a898f43038e8e99519a).

Monitored Endpoint Requirements

The monitored endpoint must meet the following requirements:

• Accept HTTP headers.

• Be instrumented with OpenTelemetry.

• Propagate trace context.

• Send traces to your observability backend.

Do not use a web page URL with distributed tracing. Browsers do not accept custom trace headers.