# node_exporter

> Apache-2.0 official Prometheus host-metrics exporter

**Category:** Observability  
**Pricing:** open-source  
**URL:** https://prometheus.io/docs/guides/node-exporter/  
**GitHub:** https://github.com/prometheus/node_exporter  
**Added:** 2026-10-17

---

## The Hook

node_exporter is the Prometheus project's own agent for machine metrics: run it on a Linux or *NIX host and it serves CPU, memory, disk, filesystem and network metrics on port 9100 for Prometheus to scrape.

## What's Good

Apache-2.0 (prometheus/node_exporter), release v1.12.1, maintained since 2013. A single Go binary with pluggable collectors you can switch on or off with flags, packaged in EPEL, available as an Ansible role and as a container image. It is the data source behind most standard Linux host dashboards. ★~13.8k.

## The Catch

It only exposes metrics: you still need a Prometheus server (or a compatible scraper) to store them and something like Grafana to view and alert on them. Running it in a container needs host networking, the host PID namespace and a read-only bind mount of / with --path.rootfs, or it ends up measuring the container instead of the machine. Windows needs the separate windows_exporter, and NVIDIA GPUs need dcgm-exporter.

## Verdict

The first thing to install on any Linux box you want to watch with Prometheus.