Cette page est affichée en anglais. Une traduction française est en cours.
ScriptsMay 9, 2026·3 min de lecture

Glances — Cross-Platform System Monitoring Tool

A curses-based system monitoring tool written in Python that displays real-time CPU, memory, disk, network, and process information in a single terminal view.

Introduction

Glances is a cross-platform system monitoring tool written in Python. It uses the psutil library to gather system metrics and presents them in a compact, real-time terminal dashboard. It can also export data to external services or run as a web server for remote monitoring.

What Glances Does

  • Displays CPU, memory, swap, disk I/O, network, and process stats in one view
  • Supports client-server mode for remote system monitoring
  • Runs a built-in web UI accessible from any browser
  • Exports metrics to InfluxDB, Prometheus, Elasticsearch, and other backends
  • Monitors Docker containers and their resource usage natively

Architecture Overview

Glances is built on top of psutil for cross-platform system metric collection. It uses a plugin-based architecture where each metric type (CPU, memory, disk, etc.) is an independent module. The curses library handles terminal rendering, while Bottle provides the optional web server. An XML-RPC or RESTful API enables client-server communication and third-party integrations.

Self-Hosting & Configuration

  • Install via pip, Homebrew, or your distribution's package manager
  • Run glances -w to start the web interface on port 61208
  • Configure thresholds and alerts in ~/.config/glances/glances.conf
  • Deploy as a Docker container with docker run --rm -e TZ=UTC -v /var/run/docker.sock:/var/run/docker.sock:ro --pid host --network host -it nicolargo/glances
  • Enable export plugins by adding backend credentials to the config file

Key Features

  • Single-binary-like experience with pip install and zero config needed
  • Real-time curses UI with color-coded alerts based on configurable thresholds
  • Built-in REST API for JSON metric access at /api/3
  • Plugin system supporting custom metric collectors
  • Wide export support including InfluxDB, Prometheus, MQTT, and CSV

Comparison with Similar Tools

  • htop — Interactive process viewer but limited to CPU/memory; Glances shows disk, network, sensors, and containers in one screen
  • btop — Polished TUI monitor with GPU support; Glances adds web mode, export backends, and client-server architecture
  • Netdata — Full observability platform with dashboards; Glances is lighter and runs without a persistent agent or database
  • nmon — AIX/Linux performance tool; Glances is cross-platform and has a modern Python API

FAQ

Q: Does Glances work on macOS and Windows? A: Yes. It runs on Linux, macOS, FreeBSD, and Windows via psutil's cross-platform support.

Q: Can I monitor remote machines? A: Yes. Run glances -s on the remote host and connect with glances -c @ip. The web mode also allows browser-based remote access.

Q: How does the Docker monitoring work? A: Mount the Docker socket into the Glances container or install the docker Python package. Glances auto-detects running containers and shows per-container CPU, memory, and network stats.

Q: Can I send alerts to Slack or email? A: Glances supports export to many backends. For alerting, pipe metrics to InfluxDB or Prometheus and use Grafana or Alertmanager for notifications.

Sources

Fil de discussion

Connectez-vous pour rejoindre la discussion.
Aucun commentaire pour l'instant. Soyez le premier à partager votre avis.

Actifs similaires