Skip to main content
OpenTelemetry offers a single set of APIs and libraries that standardize how you collect and transfer telemetry data. This guide focuses on setting up OpenTelemetry in a Go app to send traces to Axiom.

Prerequisites

Installing Dependencies

First, run the following in your terminal to install the necessary Go packages:
This installs the OpenTelemetry Go SDK, the OTLP (OpenTelemetry Protocol) trace exporter, and other necessary packages for instrumentation and resource definition.

Initializing a Go module and managing dependencies

Before installing the OpenTelemetry dependencies, ensure your Go project is properly initialized as a module and all dependencies are correctly managed. This step is important for resolving import issues and managing your project’s dependencies effectively.

Initialize a Go module

If your project is not already initialized as a Go module, run the following command in your project’s root directory. This step creates a go.mod file which tracks your project’s dependencies.
Replace <module-name> with your project’s name or the GitHub repository path if you plan to push the code to GitHub. For example, go mod init github.com/yourusername/yourprojectname.

Manage dependencies

After initializing your Go module, tidy up your project’s dependencies. This ensures that your go.mod file accurately reflects the packages your project depends on, including the correct versions of the OpenTelemetry libraries you’ll be using. Run the following command in your project’s root directory:
This command will download the necessary dependencies and update your go.mod and go.sum files accordingly. It’s a good practice to run go mod tidy after adding new imports to your project or periodically to keep dependencies up to date.

HTTP server configuration (main.go)

main.go is the entry point of the app. It invokes InstallExportPipeline from exporter.go to set up the tracing exporter. It also sets up a basic HTTP server with OpenTelemetry instrumentation to demonstrate how telemetry data can be collected and exported in a simple web app context. It also demonstrates the usage of span links to establish relationships between spans across different traces.

Exporter configuration (exporter.go)

exporter.go is responsible for setting up the OpenTelemetry tracing exporter. It defines the resource attributes, initializes the tracer, and configures the OTLP (OpenTelemetry Protocol) exporter with appropriate endpoints and headers, allowing your app to send telemetry data to Axiom.

Run the app

To run the app, execute both exporter.go and main.go. Use the command go run main.go exporter.go to start the app. Once your app is running, traces collected by your app are exported to Axiom. The server starts on the specified port, and you can interact with it by sending requests to the /rolldice endpoint. For example, if you are using port 8080, your app will be accessible locally at http://localhost:8080/rolldice. This URL will direct your requests to the /rolldice endpoint of your server running on your local machine.

Observe the telemetry data in Axiom

After deploying your app, you can log into your Axiom account to view and analyze the telemetry data. As you interact with your app, traces will be collected and exported to Axiom, where you can monitor and analyze your app’s performance and behavior.
Observing the Telemetry Data in Axiom image

Observing the Telemetry Data in Axiom image

Dynamic OpenTelemetry traces dashboard

This data can then be further viewed and analyzed in Axiom’s dashboard, providing insights into the performance and behavior of your app.
Dynamic OpenTelemetry Traces Dashboard Go

Dynamic OpenTelemetry Traces Dashboard Go

Send data from an existing Golang project

Manual Instrumentation

Manual instrumentation in Go involves managing spans within your code to track operations and events. This method offers precise control over what is instrumented and how spans are configured.
  1. Initialize the tracer:
Use the OpenTelemetry API to obtain a tracer instance. This tracer will be used to start and manage spans.
  1. Create and manage spans:
Manually start spans before the operations you want to trace and ensure they are ended after the operations complete.
  1. Annotate spans:
Enhance spans with additional information using attributes or events to provide more context about the traced operation.

Automatic Instrumentation

Automatic instrumentation in Go uses libraries and integrations that automatically create spans for operations, simplifying the addition of observability to your app.
  1. Instrumentation libraries:
Use OpenTelemetry-contrib libraries designed for automatic instrumentation of standard Go frameworks and libraries, such as net/http.
  1. Wrap handlers and clients:
Automatically instrument HTTP servers and clients by wrapping them with OpenTelemetry’s instrumentation. For HTTP servers, wrap your handlers with otelhttp.NewHandler.
  1. Minimal code changes:
After setting up automatic instrumentation, no further changes are required for tracing standard operations. The instrumentation takes care of starting, managing, and ending spans.

Reference

List of OpenTelemetry trace fields

List of imported libraries

OpenTelemetry Go SDK

go.opentelemetry.io/otel This is the core SDK for OpenTelemetry in Go. It provides the necessary tools to create and manage telemetry data (traces, metrics, and logs).

OTLP Trace Exporter

go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp This package allows your app to export telemetry data over HTTP using the OpenTelemetry Protocol (OTLP). It’s important for sending data to Axiom or any other backend that supports OTLP.

Resource and Trace Packages

go.opentelemetry.io/otel/sdk/resource and go.opentelemetry.io/otel/sdk/trace These packages help define the properties of your telemetry data, such as service name and version, and manage trace data within your app.

Semantic Conventions

go.opentelemetry.io/otel/semconv/v1.24.0 This package provides standardized schema URLs and attributes, ensuring consistency across different OpenTelemetry implementations.

Tracing API

go.opentelemetry.io/otel/trace This package offers the API for tracing. It enables you to create spans, record events, and manage context propagation in your app.

HTTP Instrumentation

go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp Used for instrumenting HTTP clients and servers. It automatically records data about HTTP requests and responses, which is essential for web apps.

Propagators

go.opentelemetry.io/otel/propagation This package provides the ability to propagate context and trace information across service boundaries.