If you are currently using the Datadog agent to send metric and trace data to Datadog, by changing the configuration in the agent’s datadog.yml file, you can send your telemetry data to Lightstep Observability. You can choose to send it to both Lightstep and Datadog, or send it only to Lightstep. In this Learning Path, you’ll send the data to both so that before you completely cut over, you can verify that the data is synched correctly.

Lightstep Observability supports versions 6 and 7 of Datadog.

You’ll learn how to:

  • Configure the Datadog agent to send metric and trace data to Lightstep Observability and Datadog
  • View metric data and trace data together in a notebook

Before you begin

You’ll need the following to complete this Learning Path:

  • Your Datadog API key
  • The Lightstep Observability access token for the project you want to report the data to.
  • For trace data, you need to configure the Datadog tracing clients to set required and recommended attributes.

    Lightstep’s datadog trace ingest requires the lightstep.service_name attribute to be set on all incoming spans. We also recommend setting the lightstep.service_version attribute. Lightstep does not currently support setting these tags on the Datadog Agent. Instead they must be configured at the tracing client level (for example, dd-trace-py, etc). Each tracing client has various in-code configuration options, but the most consistent and straightforward way to set these is using the DD_TRACE_GLOBAL_TAGS environment variable.


    DD_TRACE_GLOBAL_TAGS is to be set on the Datadog tracing clients, not the agent.