Agent skill

fly-io

Deploys applications globally on Fly.io edge infrastructure with Firecracker VMs, multi-region deployment, and usage-based pricing. Use when deploying to multiple regions, running containers at the edge, or needing fast cold starts.

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/devops/fly-io-mgd34msu-goodvibes-plugin

SKILL.md

Fly.io

Global edge deployment platform running applications in Firecracker micro-VMs close to users.

Quick Start

bash
# Install flyctl
curl -L https://fly.io/install.sh | sh

# Or on macOS
brew install flyctl

# Login
fly auth login

# Launch app (creates fly.toml)
fly launch

# Deploy
fly deploy

Project Setup

fly.toml

toml
app = "my-app"
primary_region = "ord"

[build]
  dockerfile = "Dockerfile"

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

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

[[vm]]
  cpu_kind = "shared"
  cpus = 1
  memory_mb = 256

Basic Dockerfile

dockerfile
FROM node:20-alpine

WORKDIR /app

COPY package*.json ./
RUN npm ci --only=production

COPY . .
RUN npm run build

EXPOSE 3000

CMD ["node", "dist/index.js"]

Deployment

Deploy from Git

bash
# First deploy creates app
fly launch

# Subsequent deploys
fly deploy

# Deploy specific Dockerfile
fly deploy --dockerfile Dockerfile.prod

# Deploy with build args
fly deploy --build-arg API_KEY=$API_KEY

Deploy Strategies

toml
[deploy]
  strategy = "rolling"  # default
  # strategy = "immediate"
  # strategy = "canary"
  # strategy = "bluegreen"

Regions

Multi-Region Deployment

bash
# Add regions
fly regions add ams lhr syd

# List regions
fly regions list

# Remove region
fly regions remove syd

Region Configuration

toml
app = "my-app"
primary_region = "ord"

# Scale to multiple regions
[[vm]]
  memory = "256mb"
  cpu_kind = "shared"
  cpus = 1

  [vm.processes]
    web = 2  # 2 machines per region

Available Regions

Code Location
ord Chicago
iad Virginia
lax Los Angeles
ams Amsterdam
lhr London
fra Frankfurt
syd Sydney
nrt Tokyo
sin Singapore

Secrets & Environment

Secrets (Encrypted)

bash
# Set secrets
fly secrets set DATABASE_URL=postgres://... API_KEY=secret

# List secrets
fly secrets list

# Unset secret
fly secrets unset API_KEY

Environment Variables

toml
# fly.toml
[env]
  NODE_ENV = "production"
  LOG_LEVEL = "info"

Per-Region Environment

toml
[[env]]
  NODE_ENV = "production"

  [[env.region]]
    region = "ord"
    CDN_URL = "https://us-cdn.example.com"

  [[env.region]]
    region = "ams"
    CDN_URL = "https://eu-cdn.example.com"

Scaling

Machine Scaling

bash
# Scale up
fly scale count 3

# Scale per region
fly scale count 2 --region ord
fly scale count 1 --region ams

# Scale VM size
fly scale vm shared-cpu-1x
fly scale vm dedicated-cpu-1x
fly scale memory 512

Auto-scaling

toml
[http_service]
  auto_stop_machines = true
  auto_start_machines = true
  min_machines_running = 1
  max_machines_running = 10

  [http_service.concurrency]
    type = "requests"
    soft_limit = 100
    hard_limit = 250

Databases

Postgres

bash
# Create Postgres cluster
fly postgres create

# Attach to app
fly postgres attach my-postgres-db

# Connect
fly postgres connect -a my-postgres-db

# Proxy to local
fly proxy 5432 -a my-postgres-db

Redis

bash
# Create Redis
fly redis create

# Get connection string
fly redis status my-redis

LiteFS (SQLite)

bash
# Create LiteFS cluster
fly apps create my-app-litefs

# Configure in fly.toml
[mounts]
  source = "litefs"
  destination = "/var/lib/litefs"

Volumes (Persistent Storage)

bash
# Create volume
fly volumes create mydata --size 1 --region ord

# List volumes
fly volumes list

# Destroy volume
fly volumes destroy vol_xxx
toml
# fly.toml
[mounts]
  source = "mydata"
  destination = "/data"

Networking

Custom Domains

bash
# Add domain
fly certs add example.com

# Check certificate status
fly certs show example.com

Private Networking

toml
# Internal services communicate via .internal DNS
# api.internal:3000

Static IPs

bash
# Allocate IPv4 (paid)
fly ips allocate-v4

# List IPs
fly ips list

Health Checks

toml
[http_service]
  internal_port = 3000

  [[http_service.checks]]
    grace_period = "5s"
    interval = "10s"
    timeout = "2s"
    path = "/health"
    method = "GET"
    protocol = "http"

Processes

Multiple Processes

toml
[processes]
  web = "node dist/server.js"
  worker = "node dist/worker.js"

[[http_service]]
  processes = ["web"]
  internal_port = 3000

[[services]]
  processes = ["worker"]
  internal_port = 0

Background Workers

toml
[processes]
  worker = "node dist/worker.js"

[[services]]
  processes = ["worker"]
  internal_port = 0
  auto_stop_machines = false

Node.js Configuration

Optimized Dockerfile

dockerfile
FROM node:20-alpine AS builder

WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build
RUN npm prune --production

FROM node:20-alpine

WORKDIR /app
COPY --from=builder /app/dist ./dist
COPY --from=builder /app/node_modules ./node_modules
COPY --from=builder /app/package.json ./

ENV NODE_ENV=production
USER node

CMD ["node", "dist/index.js"]

Next.js

toml
app = "my-nextjs-app"
primary_region = "ord"

[build]
  dockerfile = "Dockerfile"

[http_service]
  internal_port = 3000
  force_https = true
  auto_stop_machines = true
  auto_start_machines = true

[[vm]]
  memory = "512mb"
  cpu_kind = "shared"
  cpus = 1

CLI Commands

bash
# App management
fly apps create my-app
fly apps list
fly apps destroy my-app

# Deployment
fly deploy
fly deploy --strategy immediate
fly status

# Logs
fly logs
fly logs -a my-app

# SSH
fly ssh console
fly ssh console -C "node -v"

# Monitoring
fly status
fly dashboard

# Scaling
fly scale show
fly scale count 3
fly scale memory 512

# Regions
fly regions list
fly regions add ord ams

# Secrets
fly secrets set KEY=value
fly secrets list

See references/configuration.md for complete fly.toml options.

Didn't find tool you were looking for?

Be as detailed as possible for better results