Agent Skills: Fly.io Production Checklist

Fly.io Production Checklist

Pre-Deployment

Infrastructure

• [ ] min_machines_running = 1 (avoid cold starts)
• [ ] Machines in 2+ regions for redundancy
• [ ] VM sized appropriately (fly scale show)
• [ ] Volumes backed up (if using persistent storage)
• [ ] Postgres has standby replica

Configuration

• [ ] All secrets set via fly secrets (not [env])
• [ ] force_https = true
• [ ] Health check configured with appropriate grace period
• [ ] Custom domain with TLS certificate active
• [ ] Concurrency limits tuned for your app

Code Quality

• [ ] Dockerfile builds successfully locally
• [ ] App responds on health check endpoint
• [ ] Graceful shutdown handles SIGTERM
• [ ] No hardcoded secrets in codebase

Production fly.toml

app = "my-app"
primary_region = "iad"

[http_service]
  internal_port = 3000
  force_https = true
  auto_stop_machines = "stop"
  auto_start_machines = true
  min_machines_running = 1

[http_service.concurrency]
  type = "requests"
  hard_limit = 250
  soft_limit = 200

[http_service.checks]
  grace_period = "15s"
  interval = "10s"
  timeout = "3s"
  path = "/health"

[[vm]]
  cpu_kind = "shared"
  cpus = 2
  memory = "1gb"

Rollback Procedure

# List recent releases
fly releases -a my-app

# Rollback to previous release
fly deploy --image registry.fly.io/my-app:previous-version

# Or rollback to specific release
fly releases rollback 5 -a my-app

Monitoring

# Live logs
fly logs -a my-app

# Machine metrics
fly machine status <machine-id> -a my-app

# Platform status
curl -s https://status.flyio.net/api/v2/status.json | jq '.status.description'

Resources

• Fly.io Production Checklist
• Auto Stop/Start

Next Steps

For version upgrades, see flyio-upgrade-migration.