Our on-host integrations send data to the infrastructure agent, which in turn sends that data to New Relic. How integrations interact with the agent is controlled by each integration's config.
For more information about what integrations this applies to, see On-host integration installation.
Overview of how configuration works
Some of our on-host integrations are external programs executed by the infrastructure agent. Each integration monitors a specific service. An integration has, at minimum, these files:
- An executable that exports various types of data in a JSON format expected by the agent
- One or more YAML-format config files (for example, the Apache integration configuration). (We recommend linting YAML config files before use to avoid formatting issues.)
Note that in addition to the specific on-host integration's configuration, you can also edit the infrastructure agent's configuration.
Configuration file location
With standard on-host integration installations, the configuration is located in the infrastructure agent's directory. The agent determines this config location by a setting in its own configuration file.
For some implementations, the integration's configuration will be located elsewhere. For example:
- Services running on Kubernetes: The configuration is located in the Kubernetes integration config file.
- Services running on Amazon ECS: The configuration is placed in the AWS console.
Configuration formats
On-host integrations use two configuration formats:
- Standard: Starting December 2019, infrastructure agent version 1.8.0 began supporting a new format used by some integrations. This format uses a single configuration file and provides other improvements. For more details, see Standard configuration.
- Legacy: This is the format used by most on-host integrations. This configuration uses two files: a definition file and a configuration file. For more details, see Legacy configuration.