New Relic Agent Control

Our Agent Control combines the power of existing New Relic capabilities with open standards technologies. Agent Control is designed to be lightweight and efficient, and it can be configured to collect a wide range of data, including metrics, traces, logs, and events. This means that you no longer need to deploy multiple agents to monitor your application. With our Agent Control, you can collect all the data you need from a single package, making it easier to troubleshoot performance problems and identify issues.

Agent Control has a modular architecture, with a main supervisor that is responsible for orchestrating different agents based on the provided configuration. This allows the agent to be customized to meet the specific needs of different applications.

The agent integrates with New Relic Fleet Control providing a single point of control for all of your telemetry data and making it easier to monitor and configure your system at scale.

Supported environments

Supported agents

Below is a detailed list of the supported agents that can be managed by Agent Control:

Configuration flexibility: Local or remote

Agent Control and the managed agents support two configuration modes:

• Local configuration:
  • Settings reside directly within the environment where the agents operate.
  • Represents the default approach for straightforward management.
• Remote configuration:
  • Centrally managed through New Relic Fleet Control, offering:
    • Streamlined control for large-scale deployments.
    • Consistent settings across multiple environments.
  • Requires an active Deployment with Fleet Control.
  • Overrides local settings when enabled. Changes to local configuration are not considered when remote configuration is enabled for each agent.
  • When remote configuration is deleted, agents fallback to local configuration (if defined).

In all cases, the agent will report its own and managed agents configuration as effective configuration and it's accessible on the Fleet Control > Fleet > Agents page.

Security considerations

Users who want to leverage the Agent Control experience will need to have the All product admin role assigned to them in New Relic.

Customers should understand that unlike other features of New Relic, Agent Control does send information to your entities and has the ability to restart instrumentation, change instrumentation configuration, and potentially break configuration if invalid or malicious configurations are deployed from your fleet. We recommend that customers always test their configurations in non-production environments prior to deploying in production. While Agent Control is in preview customers should try out the functionality in non-critical, non-production environments that contain no sensitive information or data. Customers that have concerns should consider testing the feature in a separate New Relic organization or account.

Cost considerations

Agent Control can be installed on any supported enviroment and doesn't require any paid entitlement. The agent itself doesn't push any new telemetry to New Relic, ingest cost will depend on the instrumentation enabled in the managed agents (Infrastructure, APM, Logs). Refer to Fleet Control documentation for its pricing requirements.