Open Source

Glouton Monitoring Agent

A lightweight, efficient monitoring agent written in Go. Collect system and application metrics with minimal resource usage and send them to Bleemeo or any compatible platform.

View on GitHubDocumentation

Features

Why Glouton?

Lightweight & Fast

Written in Go for maximum performance with minimal resource usage. Typically uses less than 50MB of RAM.

Easy to Deploy

One-line installation on Linux, Docker, Kubernetes, and Windows. Configuration through simple YAML files.

Secure by Design

TLS encryption for all communications. No incoming connections required, works behind firewalls.

Comprehensive Monitoring

Monitors system resources, Docker containers, services, and custom metrics out of the box.

Architecture

How Glouton Works

One binary auto-discovers your services, collects and stores metrics in an embedded TSDB, surfaces them in a live local panel and a Prometheus-compatible API, and optionally forwards everything to Bleemeo Cloud or your own MQTT broker.

Glouton architecture: sources flow into Glouton, which serves a local panel, a Prometheus endpoint and an on-disk TSDB, and optionally pushes to Bleemeo Cloud or an MQTT broker.

Local panel

A Live Panel, Out of the Box

Every agent serves a status-first web panel at localhost:8015 — KPI cards, a discovered-services row and range-aware system, network and I/O charts, powered by the built-in TSDB. No account required.

Glouton local web panel showing KPI cards, a discovered-services row and a grid of system and network charts.

Install

Quick Installation

Linux (Debian/Ubuntu)

curl -fsSL https://get.bleemeo.com | sh

Or using apt:

wget -O- https://packages.bleemeo.com/bleemeo.gpg | sudo apt-key add -
echo "deb https://packages.bleemeo.com/debian stable main" | sudo tee /etc/apt/sources.list.d/bleemeo.list
sudo apt update && sudo apt install glouton

Docker

docker run -d --name glouton \
  --net host --pid host \
  --privileged \
  -v /var/lib/glouton:/var/lib/glouton \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -e BLEEMEO_ACCOUNT_ID=your-account-id \
  -e BLEEMEO_REGISTRATION_KEY=your-registration-key \
  bleemeo/glouton

Kubernetes

kubectl create namespace bleemeo
kubectl create secret generic bleemeo-secret \
  --from-literal=account-id=your-account-id \
  --from-literal=registration-key=your-key \
  -n bleemeo
kubectl apply -f https://packages.bleemeo.com/kubernetes/glouton-daemonset.yaml

Footprint

Light on Resources

~100 MB
RAM on a typical host
3–5%
of a single CPU core
~700
metric series @ 10s

Capabilities

What Glouton Monitors

System Metrics

  • CPU usage and load average
  • Memory and swap usage
  • Disk I/O and space
  • Network traffic and connections
  • System processes

Docker Containers

  • Container resource usage
  • Container status and health
  • Image and registry information
  • Network and volume metrics

Services & Applications

  • Apache, Nginx, HAProxy
  • PostgreSQL, MySQL, MongoDB
  • Redis, Memcached, Elasticsearch
  • RabbitMQ, Kafka
  • Custom application metrics

Integrations

  • Prometheus exporter
  • StatsD receiver
  • SNMP monitoring
  • Custom checks and scripts
  • REST API endpoints

Join the Community

Glouton is open source and actively maintained. We welcome contributions, bug reports, and feature requests from the community.

Apache 2.0
License
Go
Language
Active
Development
Star on GitHubReport an IssueRead the Docs