For Agent v7.21+ / v6.21+:

1. Run the following command to install the Agent integration:

   datadog-agent integration install -t datadog-<INTEGRATION_NAME>==<INTEGRATION_VERSION>
   

   The version for the integration can be found in the respective changelog on the integration’s Github repository

2. Configure your integration similar to core integrations.

3. Restart the Agent.

Note: If necessary, prepend sudo -u dd-agent to the install command.

To use a community or Marketplace integration in a containerized environment, you must build a custom image that includes your desired community integration.

Use the following Dockerfile to build a custom version of the Agent that includes the <INTEGRATION_NAME> from integrations-extras. If you are installing a Marketplace integration, the <INTEGRATION_NAME> is available in the configuration instructions.

FROM gcr.io/datadoghq/agent:latest
RUN agent integration install -r -t datadog-<INTEGRATION_NAME>==<INTEGRATION_VERSION>

The datadog-agent integration install command (run inside Docker) issues the following harmless warning: Error loading config: Config File "datadog" Not Found in "[/etc/datadog-agent]": warn. You can ignore this warning.

If you are using Kubernetes, update your Helm chart or Datadog Operator configuration to pull your custom image.

Use Autodiscovery to enable and configure the integration.

For Agent < v7.21 / v6.21:

1. Download the files in the <INTEGRATION_NAME>/datadog_checks/<INTEGRATION_NAME>/ folder from the integrations-extra repository.
2. Place <INTEGRATION_NAME>.py and any other Python files in the Agent’s checks.d directory.
3. Create a new <INTEGRATION_NAME>.d/ folder in your Agent configuration directory.
4. Place the conf.yaml.example file from the <INTEGRATION_NAME>/datadog_checks/<INTEGRATION_NAME>/data/ folder in the created directory.
5. Rename this file to conf.yaml.
6. Configure your integration similar to core integrations.
7. Restart the Agent.