Skip to content

Lightrun for Apache Airflow

To run Lightrun for Python with Apache Airflow you must select one or more tasks in the DAG that you want to debug.


  1. With the same Python interpreter used by Airflow, install the Python agent, by running python -m pip install lightrun.
  2. Within your Airflow task definition, add the @lightrun_airflow_task decorator; for example:
    def foo():
        # Do something


To import the decorator, add following line:

from lightrun.decorators import lightrun_airflow_task

Decorator parameters

The following table lists the available parameters for the lightrun_airflow_task decorator.

Parameter Explanation
company The Lightrun registered company to which the user belongs. Must be supplied with either this parameter or the LIGHTRUN_COMPANY environment variable.
company_key The company's secret API key. Must be supplied with either this parameter or the LIGHTRUN_KEY environment variable.
config_path [Optional] The filesystem path of a configuration for the lightrun agent
initial_sleep_ms The time, in milliseconds, to sleep after initializing the agent. If this parameter is set to a low value, the agent might have insufficient time to initialize before the task begins execution, which will result in missing breakpoints inserted in lines that are executed before the agent finishes initialization.

You may need more agents

A separate agent is required for each Airflow task that you want to apply simultaneously with other tasks. Contact us if you need more agents.

Last update: April 17, 2024