Ubuntu Code Server

Visual Studio Code Server running on Ubuntu with Docker support and CUDA capability.

🚀 Overview

This project provides a Docker-based Visual Studio Code Server. It includes support for multiple architectures (AMD64, ARM64), CUDA for GPU acceleration, and comes pre-configured with essential development tools.

✨ Features

• Visual Studio Code Server - Access your development environment via web browser
• Multi-architecture Support - Works on AMD64 and ARM64 platforms
• CUDA Support - GPU acceleration for machine learning and data science workloads
• Pre-installed Tools:
  • Python, Java Node.js
  • MySQL, SQLite clients
  • Git, OpenSSH, LFTP
  • OpenResty (Nginx + Lua)
  • Supervisor for process management
• Language Support:
  • English (en_US)
  • Russian (ru_RU)
• Web-based Access - Code from anywhere via browser

📦 Prerequisites

• Docker installed on your system
• For CUDA support: NVIDIA GPU with appropriate drivers
• Docker Buildx for multi-architecture builds

🚀 Quick Start

Pull from Docker Hub

# AMD64/ARM64 version
docker pull bayrell/ubuntu_code_server:latest-default

# CUDA version (for NVIDIA GPU)
docker pull bayrell/ubuntu_code_server:latest-cuda

Run the container

# Basic run
docker run -d \
  --name code-server \
  -p 8080:8000 \
  -e CODE_SERVER_ENABLE_ADMIN=1 \
  -v ~/code-server-data:/data \
  bayrell/ubuntu_code_server:latest

# With GPU support
docker run -d \
  --name code-server \
  --gpus all \
  -p 8080:8000 \
  -e CODE_SERVER_ENABLE_ADMIN=1 \
  -v ~/code-server-data:/data \
  bayrell/ubuntu_code_server:latest-cuda

🔧 Building Images

1. Download Code Server binaries

./build.sh download

2. Build for different architectures

# AMD64 build
./build.sh amd64

# ARM64 build
./build.sh arm64

# CUDA build (AMD64 only)
./build.sh cuda-amd64

🐳 Running the Container

Basic Docker Run

docker run -d \
  --name code-server \
  -p 8080:8000 \
  -v /path/to/data:/data \
  -e WWW_UID=1000 \
  -e WWW_GID=1000 \
  -e CODE_SERVER_ENABLE_ADMIN=1 \
  bayrell/ubuntu_code_server:latest-default

Docker Compose

version: '3.8'

services:
  code-server:
    image: bayrell/ubuntu_code_server:latest-default
    container_name: code-server
    ports:
      - "8080:8000"
    volumes:
      - ./data:/data
      - /var/run/docker.sock:/var/run/docker.sock:ro
    environment:
      - WWW_UID=1000
      - WWW_GID=1000
      - CODE_SERVER_ENABLE_ADMIN=1
    restart: unless-stopped

With GPU Support

services:
  code-server:
    image: bayrell/ubuntu_code_server:latest-cuda
    container_name: code-server
    ports:
      - "8080:8000"
    volumes:
      - ./data:/data
      - /var/run/docker.sock:/var/run/docker.sock:ro
    environment:
      - WWW_UID=1000
      - WWW_GID=1000
      - CODE_SERVER_ENABLE_ADMIN=1
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

⚙️ Configuration

Environment Variables

Variable	Default	Description
WWW_UID	1000	User ID for the www-data user
WWW_GID	1000	Group ID for the www-data user
CODE_SERVER_ENABLE_ADMIN	0	Enable admin mode (1 for admin)
CODE_SERVER_CUDA	0	Set to 1 for CUDA-enabled image

Ports

• 8000 - Code Server web interface

Volumes

Volume	Description
/data	User data, extensions, settings
/var/run/docker.sock	Docker socket (read-only) for container management

🚀 CUDA Support

Requirements

• NVIDIA GPU with CUDA support
• NVIDIA Container Toolkit installed
• CUDA-capable drivers

Running with CUDA

# Check GPU availability
nvidia-smi

# Run with GPU
docker run --gpus all bayrell/ubuntu_code_server:latest-cuda

CUDA-Enabled Tools

• Python with GPU acceleration
• Jupyter notebooks with CUDA
• Machine learning frameworks (TensorFlow, PyTorch)

📄 License

This project is licensed under the terms found in the LICENSE file.

🔗 Links

• Docker Hub
• GitHub Repository
• Code Server Official
• Visual Studio Code

📞 Support

For issues and questions:

• Create an issue on GitHub
• Check the troubleshooting section
• Review the Docker Hub documentation

---

Built with ❤️ for developers