uncloud is a lightweight clustering and container orchestration tool created by Pasha Sviderski that lets you deploy and manage web apps across cloud VMs and bare metal with minimal cluster management overhead. It bridges the gap between Docker simplicity and Kubernetes power, providing Docker simplicity with multi-machine power.
🎯 The Sweet Spot: Modern cloud platforms like Heroku offer amazing developer experiences but at premium prices. Kubernetes provides power but requires significant expertise. uncloud finds the pragmatic middle ground - own your infrastructure, stay simple as you grow, build with proven primitives.
⚠️ Development Status: uncloud is currently in active development and is not ready for production use. Features may change significantly between releases. Perfect for experimentation and learning!
🚀 Why Latest Versions Matter: uncloud is rapidly evolving with frequent releases containing new orchestration features, performance improvements, and enhanced multi-machine capabilities. The latest versions include better WireGuard mesh networking, improved service discovery, and enhanced Docker Compose compatibility.
⚡ Key Features of uncloud
🌐 Multi-Machine Deployment
Deploy across cloud VMs, dedicated servers, and bare metal. Combine different providers into unified computing environment.
🐳 Docker Compose Compatible
Use familiar Docker Compose format for defining services. No need to learn new configuration languages or complex YAML.
🔒 Zero-Config Private Network
Automatic WireGuard mesh with peer discovery and NAT traversal. Containers get unique IPs for direct cross-machine communication.
🎯 No Control Plane
Fully decentralized design eliminates single points of failure. Each machine maintains synchronized cluster state through peer-to-peer communication.
🔍 Service Discovery
Built-in DNS server resolves service names to container IPs. Automatic load balancing and ingress with HTTPS via Caddy.
📦 Persistent Storage
Run stateful services with Docker volumes managed across machines. Data persistence without complex storage orchestration.
# Initialize cluster on first machine
uc machine init root@your-server-ip
# Add more machines to cluster
uc machine add --name server2 root@second-server-ip
# List cluster machines
uc machine ls
Deploy applications:
# Deploy from Docker image with HTTPS
uc run -p app.example.com:8000/https image/my-app
# Deploy from Docker Compose file
uc compose up -f compose.yaml
# Scale service across machines
uc scale my-app 3
# List running services
uc ls
Manage deployments:
# Update service with new image
uc update my-app image/my-app:v2
# View service logs
uc logs my-app
# Remove service
uc rm my-app
# Clean up cluster
uc machine rm server2
🚀 为什么选择 deb-free.griffo.io？
📊 Installation Comparison:
Official Debian: Not available in official repositories
Install Script: Manual updates and dependency management
Manual Binary: No package management or automatic updates
deb.griffo.io (main): Latest version with automatic updates
✅ Latest Features: Get newest orchestration and networking improvements
✅ No Manual Setup: Pre-built binaries with all dependencies
✅ Automatic Updates: Packages updated within hours of upstream releases
✅ Complete Package: Includes all required components and scripts
✅ Multi-Distribution: Works on Bookworm, Trixie, and Sid
✅ Easy Maintenance: Standard apt commands for updates
🎯 Perfect for: Developers who want multi-machine deployment without Kubernetes complexity, teams needing simple container orchestration, homelab enthusiasts, and anyone seeking the sweet spot between Docker and Kubernetes.