Skip to main content

Netdata

Netdata is a distributed, real-time performance and health monitoring system. It provides unparalleled insights, in real-time, of everything happening on the system it runs on (including applications such as web and database servers).

Core Features​

  • Real-time, high-resolution: Collects metrics per second.
  • Low footprint: Designed to be lightweight and run indefinitely on all systems.
  • Zero configuration: Auto-detects thousands of metrics out of the box.
  • Interactive web dashboards: Stunning, highly responsive visualization.
  • Customizable alerting: Powerful alarm engine.
  • Extensible: Supports hundreds of plugins for various applications and services.

Installation​

Netdata can be installed on almost any Linux distribution with a single command. The most common and recommended way is using their kickstart script:

bash <(curl -Ss https://my-netdata.io/kickstart.sh)

This script will determine the OS, install necessary dependencies, and install Netdata.

Architecture​

Netdata operates primarily as a standalone agent on each node.

  1. Data Collection: Internal and external plugins gather metrics from the OS, hardware, and applications.
  2. Time-Series Database (TSDB): Metrics are stored in a custom, highly efficient TSDB in memory, optionally spilling to disk.
  3. Alarm Engine: Evaluates metrics against pre-configured or custom rules.
  4. Web Server: Serves the API and the interactive dashboard.

Using the Dashboard​

Once installed, the dashboard is accessible at http://[your-server-ip]:19999.

  • Navigation: The right-hand menu provides quick navigation to different sections (CPU, Memory, Disks, Network, Applications).
  • Interactivity: You can pan, zoom, and highlight specific timeframes on any chart. All charts sync automatically to the selected timeframe.
  • Alerts: Active alerts are visible directly on the dashboard.

Configuration​

The main configuration file is usually located at /etc/netdata/netdata.conf. However, Netdata provides a helper script to edit configuration files safely:

cd /etc/netdata
sudo ./edit-config netdata.conf

Configuring Alarms​

Alarms are configured in .conf files under /etc/netdata/health.d/. You can edit existing ones or create new ones using the edit-config tool:

sudo ./edit-config health.d/cpu.conf

Integrating with Centralized Monitoring​

While Netdata excels at per-node real-time monitoring, it can also stream metrics to centralized Time-Series Databases (like Prometheus, Graphite, or InfluxDB) for long-term storage and global analysis. This allows you to combine Netdata's high-resolution, zero-config collection with the centralized capabilities of systems like Prometheus and Grafana.