DeployCrane

A minimal PaaS-inspired deployment system that builds and runs Git-based applications using Docker with realtime status tracking.

---

System Architecture

DeployCrane is organized into four core layers:

• API Layer — REST request handling
• Service Layer — deployment orchestration
• Persistence Layer — SQLite-based state storage
• Infrastructure Layer — Git + Docker integration

---

Deployment Workflow

A deployment in DeployCrane follows this lifecycle:

Request → Clone Repo → Store Metadata → Build Image → Start Container → Monitor → Stream Status

---

Built With

DeployCrane leverages:

• Go — backend API and orchestration
• Docker — container runtime and deployment isolation
• SQLite — lightweight persistent state and app details tracking
• Server-Sent Events (SSE) — real-time deployment updates
• REST API — automation and integrations

---

Setup

# Pull the latest image from DockerHub
docker pull parsasafavi/deploycrane:latest

# Run it with Docker socket
docker run -d \
  --name deploycrane \
  -p 8080:8080 \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v ./data:/app/data \
  -v ./clones:/app/clones \
  --restart unless-stopped \
  parsasafavi/deploycrane:latest

For a more detailed guide on how to setup and run your own deploycrane instance refer to this link.

---

Planned Improvements:

Core enhancements

• Multi-user authentication
• Deployment rollback support
• Application logs

Scaling & infrastructure

• Multi-host deployments
• Docker image-based deployment mode

Observability

• Prometheus metrics integration
• Enhanced runtime monitoring