Set up Grafana Agent in static mode on Grafana Labs

Install Grafana Agent in static mode

Fri, 13 Mar 2026 14:43:40 -0400

Install Grafana Agent in static mode

You can install Grafana Agent in static mode on Docker, Kubernetes, Linux, macOS, or Windows.

The following architectures are supported:

Linux: AMD64, ARM64
Windows: AMD64
macOS: AMD64 (Intel), ARM64 (Apple Silicon)
FreeBSD: AMD64

Note
ppc64le builds are considered secondary release targets and do not have the same level of support and testing as other platforms.

Note
Installing Grafana Agent on other operating systems is possible, but is not recommended or supported.

Grafana Cloud

Use the Grafana Agent Kubernetes configuration or follow instructions for installing the Grafana Agent in the Walkthrough.

Data collection

By default, Grafana Agent sends anonymous usage information to Grafana Labs. Refer to data collection for more information about what data is collected and how you can opt-out.

Start, restart, and stop Grafana Agent in static mode

Fri, 13 Mar 2026 14:43:40 -0400

Start, restart, and stop Grafana Agent in static mode

You can start, restart, and stop Grafana Agent after it is installed.

Linux

Grafana Agent is installed as a systemd service on Linux.

Start Grafana Agent

To start Grafana Agent, run the following command in a terminal window:

sudo systemctl start grafana-agent

(Optional) Verify that the service is running:

sudo systemctl status grafana-agent

Configure Grafana Agent to start at boot

To automatically run Grafana Agent when the system starts, run the following command in a terminal window:

sudo systemctl enable grafana-agent.service

Restart Grafana Agent

To restart Grafana Agent, run the following command in a terminal window:

sudo systemctl restart grafana-agent

Stop Grafana Agent

To stop Grafana Agent, run the following command in a terminal window:

sudo systemctl stop grafana-agent

View Grafana Agent logs on Linux

To view the Grafana Agent log files, run the following command in a terminal window:

sudo journalctl -u grafana-agent

macOS

Grafana Agent is installed as a launchd service on macOS.

Start Grafana Agent

To start Grafana Agent, run the following command in a terminal window:

brew services start grafana-agent

Grafana Agent automatically runs when the system starts.

Optional: Verify that the service is running:

brew services info grafana-agent

Restart Grafana Agent

To restart Grafana Agent, run the following command in a terminal window:

brew services restart grafana-agent

Stop Grafana Agent

To stop Grafana Agent, run the following command in a terminal window:

brew services stop grafana-agent

View Grafana Agent logs on macOS

By default, logs are written to $(brew --prefix)/var/log/grafana-agent.log and $(brew --prefix)/var/log/grafana-agent.err.log.

If you followed Configure steps in the macOS install instructions and changed the path where logs are written, refer to your current copy of the Grafana Agent formula to locate your log files.

Windows

Grafana Agent is installed as a Windows Service. The service is configured to automatically run on startup.

To verify that Grafana Agent is running as a Windows Service:

Open the Windows Services manager (services.msc):
1. Right click on the Start Menu and select Run.
2. Type: services.msc and click OK.
Scroll down to find the Grafana Agent service and verify that the Status is Running.

View Grafana Agent logs

When running on Windows, Grafana Agent writes its logs to Windows Event Logs with an event source name of Grafana Agent.

To view the logs, perform the following steps:

Open the Event Viewer:
1. Right click on the Start Menu and select Run.
2. Type eventvwr and click OK.
In the Event Viewer, click on Windows Logs > Application.
Search for events with the source Grafana Agent.

Standalone binary

If you downloaded the standalone binary, you must run the agent from a terminal or command window.

Start Grafana Agent on Linux, macOS, FreeBSD, or Windows

To start Grafana Agent on Linux, macOS, FreeBSD, or Windows run the following command in a terminal window or command prompt:

BINARY_PATH -config.file CONFIG_FILE

Replace the following:

BINARY_PATH: The path to the Grafana Agent binary file
CONFIG_FILE: The path to the Grafana Agent configuration file.

Deploy Grafana Agent in static mode

Fri, 13 Mar 2026 14:43:40 -0400

Deploy Grafana Agent

Grafana Agent is a flexible, vendor-neutral telemetry collector. This flexibility means that Grafana Agent doesn’t enforce a specific deployment topology but can work in multiple scenarios.

This page lists common topologies used for deployments of Grafana Agent, when to consider using each topology, issues you may run into, and scaling considerations.

As a centralized collection service

Deploying Grafana Agent as a centralized service is recommended for collecting application telemetry. This topology allows you to use a smaller number of agents to coordinate service discovery, collection, and remote writing.

Using this topology requires deploying the Agent on separate infrastructure, and making sure that agents can discover and reach these applications over the network. The main predictor for the size of the agent is the number of active metrics series it is scraping; a rule of thumb is approximately 10 KB of memory for each series. We recommend you start looking towards horizontal scaling around the 1 million active series mark.

Using Kubernetes StatefulSets

Deploying Grafana Agent as a StatefulSet is the recommended option for metrics collection. The persistent pod identifiers make it possible to consistently match volumes with pods so that you can use them for the WAL directory.

You can also use a Kubernetes deployment in cases where persistent storage is not required, such as a traces-only pipeline.

Pros

Straightforward scaling using clustering or hashmod sharding
Minimizes the “noisy neighbor” effect
Easy to meta-monitor

Cons

Requires running on separate infrastructure

Use for

Scalable telemetry collection

Don’t use for

Host-level metrics and logs

As a host daemon

Deploying one Grafana Agent per machine is required for collecting machine-level metrics and logs, such as node_exporter hardware and network metrics or journald system logs.

Each Grafana Agent requires you to open an outgoing connection for each remote endpoint it’s shipping data to. This can lead to NAT port exhaustion on the egress infrastructure. Each egress IP can support up to (65535 - 1024 = 64511) outgoing connections on different ports. So, if all agents are shipping metrics and log data, an egress IP can support up to 32,255 agents.

Using Kubernetes DaemonSets

The simplest use case of the host daemon topology is a Kubernetes DaemonSet, and it is required for node-level observability (for example cAdvisor metrics) and collecting pod logs.

Pros

Doesn’t require running on separate infrastructure
Typically leads to smaller-sized agents
Lower network latency to instrumented applications

Cons

Requires planning a process for provisioning Grafana Agent on new machines, as well as keeping configuration up to date to avoid configuration drift
Not possible to scale agents independently when using Kubernetes DaemonSets
Scaling the topology can strain external APIs (like service discovery) and network infrastructure (like firewalls, proxy servers, and egress points)

Use for

Collecting machine-level metrics and logs (for example, node_exporter hardware metrics, Kubernetes pod logs)

Don’t use for

Scenarios where Grafana Agent grows so large it can become a noisy neighbor
Collecting an unpredictable amount of telemetry

As a container sidecar

Deploying Grafana Agent as a container sidecar is only recommended for short-lived applications or specialized agent deployments.

Using Kubernetes pod sidecars

In a Kubernetes environment, the sidecar model consists of deploying Grafana Agent as an extra container on the pod. The pod’s controller, network configuration, enabled capabilities, and available resources are shared between the actual application and the sidecar agent.

Pros

Doesn’t require running on separate infrastructure
Straightforward networking with partner applications

Cons

Doesn’t scale separately
Makes resource consumption harder to monitor and predict
Agents do not have a life cycle of their own, making it harder to reason about things like recovering from network outages

Use for

Serverless services
Job/batch applications that work with a push model
Air-gapped applications that can’t be otherwise reached over the network

Don’t use for

Long-lived applications
Scenarios where the agent size grows so large it can become a noisy neighbor

Grafana Agent quick starts

Fri, 13 Mar 2026 14:43:40 -0400

Grafana Agent quick starts

The following quick starts help you get up and running with Grafana Agent. You’ll learn how to send your metrics, logs, and traces to the Grafana Stack or Grafana Cloud.

Grafana Stack quick starts

Send metrics to Mimir using Grafana Agent.
Send traces to Tempo using Grafana Agent.
Send logs to Loki using Grafana Agent.

Grafana Cloud quick starts

Grafana Agent for Grafana Cloud.
Monitoring a Linux host using the Linux Node integration.
Grafana Agent Kubernetes configuration.