deployment

Deployment — Infrastructure and Service Management

Triggers: /deploy, /deployment, /setup-server, /port, /reserve-ports, /docker, deploy to, start service, stop service, port conflict, port range, reserve ports, docker-compose, container, infrastructure, provision, configure server

---

WORKFLOW

1. Port Registry Management

Maintain a port registry at .sisyphus/ports.json to track allocated ports and prevent conflicts.

Registry Schema:

{
  "version": 1,
  "ranges": {
    "omo-hub": { "start": 3000, "end": 3010, "allocated": "2026-03-19", "contact": "ezotoff" },
    "kraken": { "start": 3100, "end": 3120, "allocated": "2026-03-15", "contact": "ezotoff" }
  },
  "ports": {
    "3000": { "service": "web-app", "project": "omo-hub", "allocated": "2026-03-19" },
    "3001": { "service": "api-server", "project": "omo-hub", "allocated": "2026-03-19" },
    "5432": { "service": "postgres", "project": "omo-hub", "allocated": "2026-03-19" },
    "6379": { "service": "redis", "project": "omo-hub", "allocated": "2026-03-19" }
  }
}

Reserve a port range for a new project:

When a project doesn't have a reserved port range:

1. Ask for requirements:

   • Project name (slug, e.g., my-project)
   • Number of ports needed (recommend: 10 for small, 20 for medium, 50 for large projects)
   • Preferred base port (optional, default: find next available)

2. Find available range:

   # Check what ranges are already reserved
   cat ~/.sisyphus/ports.json | jq '.ranges'
   
   # Find gap between existing ranges
   # Example: if ranges are 3000-3010 and 3100-3120, suggest 3011-3099 or 3121+
   

3. Validate range is free:

   # Check if any port in range is in use
   RANGE_START=3200
   RANGE_END=3220
   for port in $(seq $RANGE_START $RANGE_END); do
     netstat -tuln 2>/dev/null | grep -q ":${port} " && echo "CONFLICT: $port"
   done