StatNet is an ultra-lightweight real-time monitoring app with a simple web interface. It’s designed to be fast, minimal, and efficient.

Built as a fully static site using Next.js and powered by a Go backend.

Click to view

Light Mode	Dark Mode
Light Mode	Dark Mode
Light Mode	Dark Mode
Light Mode	Dark Mode
Light Mode	Dark Mode

• Host System Stats: Displays CPU, memory, network, and disk usage of the host system.
• Container Stats: Monitors all containers on the host, showing stats like CPU, memory, network, disk, and uptime (including offline containers).
• Custom Container Name Mappings: Allows you to customize container names in the web interface for easier identification.
• Coolify Stats: Tracks CPU, memory, network, and disk usage for Coolify containers.
• Responsive Design: Optimized for both mobile and desktop views.
• Extremly Lightweight: The app’s final Docker image is under 18MB, using less than 15MB of memory and under 1% CPU usage (single-core) during runtime.

StatNet uses Next.js for the static frontend and a Go backend server to serve the site and gather system stats.

The Go backend pulls data via the Docker API and reads the /proc directory on the host system for performance metrics.

It can also read Coolify directories and volumes to fetch disk usage stats for Coolify containers.

I couldn't find a lightweight, real-time monitoring app with a clean web interface that met my needs, so I decided to build my own.

I made it open source so anyone can customize it and use it for their own setups.

I’ve deployed StatNet on my home server via Coolify, on a custom internal domain that's only accessible on my local network.

I use the web interface to quickly check the performance of my server and containers, including their resource usage.

You can deploy StatNet using the provided Docker Compose fi A73D le or through Coolify.

1. Create a compose.yaml file on your server and paste the contents of the docker-compose.yaml from this repo.
2. Create a container-names.json file based on the container-names-sample.json in this repo, and update the file path in the Compose file.
3. Run docker compose up to launch the containers.

1. Create a container-names.json file based on the container-names-sample.json in this repo, and update the file path in the Compose file.
2. Add a new resource in Coolify → "Docker Compose Empty."
3. Paste the contents of the coolify.yaml from the repo into the input field.
4. Click "Deploy!"

• No Database

  • All data is served in real-time to the web interface—there is no historical data stored, and no database is used.

• Single Server

  • StatNet only monitors a single server at a time. To monitor multiple servers, you’ll need to deploy StatNet on each one individually.

Error message:

Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: open /proc/sys/net/ipv4/ip_unprivileged_port_start: read-only file system: unknown

Explanation and Fix:

This error typically occurs on amd64 processors. The solution is to modify the mount configuration in your Docker Compose file:

• Original mount:

  - '/proc:/proc:ro'
  

• Updated mount (remove read-only flag):

  - '/proc:/proc'
  

Reason: On amd64 processors, Docker sometimes needs write access to /proc to adjust kernel networking parameters and security settings during container initialization. Mounting /proc as read-only blocks this necessary functionality, leading to the error. Removing the read-only flag (- '/proc:/proc') allows Docker to make the required changes and resolve the issue.

To get started with development, you'll need to have Node.js and bun installed.

# Clone the repo
git clone https://github.com/airoflare/statnet.git
cd statnet

# Build the docker imageand run docker image locally (I fully test using docker locally)
docker build -t metrics .

# Run the docker image on port 80
docker run -p 80:80 -v /var/run/docker.sock:/var/run/docker.sock metrics
# → Opens at http://localhost:80

Optional Configuration:

# If you want to allow other origins like 10.X.X.X or any Ipv4 with localhost you can add it using docker's built in enviroment argument and use the "ALLOWED_CORS_ORIGINS" enviroment variable

#Example:
docker run -p 80:80 -e ALLOWED_CORS_ORIGINS="10.0.0.8" -v /var/run/docker.sock:/var/run/docker.sock metrics

# This will allow you to access the monitor using http://10.0.0.8:80 (10.0.0.8 is an example IP, make sure to use an IP related to the machine that you are running this on)

I built this with v0.dev, Gemini CLI and ChatGPT.

I don’t really know much about Next.js or React or Go, so the codebase is kind of a mess — but it works!

I only tested this app on a home server running Linux Debian 12