ℹ️  SignalFx was acquired by Splunk in October 2019. See Splunk SignalFx for more information.

SignalFx Distributed Tracing Instrumentation

Instrumentation Agnostic

SignalFx Distributed Tracing intends to be instrumentation agnostic; supporting & ingesting variety of popular open instrumentation libraries including OpenTracing, Zipkin & OpenCensus. So long as the tracer configured to send spans to SignalFx ingest endpoint uses the Zipkin v1/2 JSON wire format or Jaeger Thrift format, we will accept spans irrespective of how they were instrumented - via one of the above mentioned open libraries or a homegrown one. Our goal is to build on the shoulders of giant communities rather than re-invent the wheel with proprietary libraries/agents and give customers choice to decide what works for them without worrying about vendor lock-in.

For customers who have not yet instrumented their code, our default recommendation is to use OpenTracing to instrument since it has a growing ecosystem of library owners & frameworks instrumenting their code with it & use Jaeger tracer libraries to export spans to us - both of these are CNCF projects with a rapidly growing community behind them.

How should I go about instrumenting my application for distributed tracing?

RPC layer(s) and service/web framework(s) are the best places to start when thinking about how to go about instrumentating your application - both of these will likely have a large coverage area and touch a significant number of transaction paths to give you baseline tracing coverage and visualize an end-to-end trace with a service-map.

Next you should identify services critical to your business and look for areas not covered by rpc or service/web frameworks. Within these services, identify high value transactions and critical paths - instrument enough of these.

• Guide to help strategize instrumentation for your distributed application
• RPC and Web Frameworks pre-instrumented with OpenTracing
• Service-Mesh is another increasingly popular way to quickly get rpc/inter-service visibility

Examples

This repository contains a set of simple example applications that demonstrate using OpenTelemetry or SignalFx with various open source tracers. They are broken down by language/platform.

.NET

• SignalFx.Tracing: An example using the SignalFx.Tracing.OpenTracing NuGet Package to instrument .NET libraries and applications.

Java

• OpenTelemetry Java: our recommended tracer for Java.

Python

• OpenTelemetry Python: our recommended tracer for Python.

Go

• SignalFx Go: SignalFx tracer for Golang.

Service Meshes

• Istio
  • Istio Mixer Adapter: We have an out of process adapter available for Istio. This is an example configuration for that adapter.
  • Istio E-Commerce Application: An example application with Istio tracing using the SignalFx adapter and Envoy tracing.
• Envoy: the Envoy proxy can be configured to report trace spans to SignalFx.
• AWS App Mesh: An example E-Commerce application with deployment and configuration files for App Mesh on ECS.

Auto-Instrumentation

For customers who have not instrumented their applications, or have done so in an OpenTracing-compatible fashion, we offer several SignalFx Tracing libraries. Their detailed documentation is available in their respective source locations:

• .NET
• Go
• PHP

Examples of those auto-instrumentation techniques are available in this repository:

• SignalFx Tracing Examples