Announcing the Beta Release of OpenTelemetry Go Auto-Instrumentation using eBPF
The OpenTelemetry community is excited to announce the beta release of the OpenTelemetry Go Auto-Instrumentation project! This milestone brings us closer to our mission of making observability simple, accessible, and effective for Go applications.
What is Go Auto-Instrumentation?
OpenTelemetry Go Auto-Instrumentation allows developers to collect traces from their Go applications without requiring manual code modifications or rebuilding binaries. By dynamically instrumenting applications at runtime using eBPF, this project lowers the barrier to adopting observability best practices and provides deep insights into your application’s behavior.
Key features of the beta release
The beta release offers foundational support for automatic instrumentation with these key features:
- HTTP server instrumentation: Automatically trace incoming and outgoing
HTTP requests with trace context propagation when using the
net/httppackage. - Database instrumentation: Instrument database queries and connections that
use the
database/sqlpackage. - gRPC instrumentation: Easily collect telemetry data from gRPC clients and servers.
- Kafka-go instrumentation: Monitor and trace Kafka messaging using the
kafka-gopackage. - Extensible with OpenTelemetry’s Trace API: Seamlessly extend auto-instrumentation with custom spans using the OpenTelemetry Go Trace API.
- Configuration using environment variables: Simplify configuration with environment-based settings, reducing the need for code changes.
- Semantic convention compliance: Produced telemetry complies with the latest OpenTelemetry semantic conventions ensuring compatibility with the OTel ecosystem.
Getting started
Getting started with OpenTelemetry Go Auto-Instrumentation is straightforward! For detailed instructions on installation, configuration, and running your application with auto-instrumentation, check out the Getting Started guide.
To see a complete example, check out the
rolldice application.
The road to stabilization
With the beta release now available, the project team is focused on preparing for a stable release. Over the coming year, we’ll concentrate on achieving the following goals:
Optimize runtime instrumentation with eBPF
We will continue to leverage and improve eBPF for dynamic runtime instrumentation, ensuring that Go applications have reliable, low-overhead observability. This includes staying up to date with the latest developments in both the Go and eBPF ecosystems.
Expand ecosystem support
While the beta release supports a limited number of Go packages and only a single telemetry processing pipeline, we plan to broaden this support. We’ll introduce instrumentation for additional Go packages and allow users to create custom telemetry processing pipelines. This expansion will make it easier to integrate with popular Go packages and provide flexibility for custom instrumentation.
Integrate with Beyla’s donation to OpenTelemetry
The proposed donation of Beyla to OpenTelemetry presents an opportunity to enhance OpenTelemetry’s eBPF-based auto-instrumentation capabilities. If accepted, this donation will help expand support for additional signals, protocols, and languages, creating a more comprehensive eBPF-based observability solution. OpenTelemetry Go Auto-Instrumentation will evolve alongside these developments, ensuring seamless collaboration while continuing to provide robust tracing for Go applications.
Keep track of our progress in the 2025 Goals tracking issue.
Join the journey
As with all OpenTelemetry projects, the success of Go Auto-Instrumentation depends on the community. Whether you’re a developer interested in contributing, a company looking to adopt the project, or simply curious about observability, we’d love for you to join us.
Here’s how you can get involved:
- Try the beta: Integrate the project into your applications and provide feedback.
- Contribute: Check out open issues and contribute to the project on GitHub.
- Join the discussion: Participate in our SIG meetings and discussions on Slack.
Acknowledgments
This beta release is the result of countless hours of work by contributors from around the world. Thank you to everyone who has contributed code, documentation, feedback, and enthusiasm to make this milestone possible.