App Connection Examples

These examples demonstrate connections from within the same Fly.io internal IPv6 private network . To adapt them for external connections, substitute the internal hostname for one that’s reachable over the public Internet.

psql

If you have an active WireGuard tunnel to your organization’s private network, you can connect from a local machine to your Postgres cluster the same way you would from a Fly app within the same private network. For example, the following command would start an interactive terminal session on the cluster leader with psql:

psql postgres://postgres:secret123@appname.internal:5432

You can, of course, run psql from a shell on a Fly app as well (as long as that app has psql installed).

Connecting with Ruby

docs

Ruby apps use the pg gem to connect to postgres.

require 'pg'

# Output a table of current connections to the DB
conn = PG.connect("postgres://postgres:secret123@postgresapp.internal:5432/yourdb")
conn.exec( "SELECT * FROM pg_stat_activity" ) do |result|
  puts "     PID | User             | Query"
  result.each do |row|
    puts " %7d | %-16s | %s " %
      row.values_at('pid', 'usename', 'query')
  end
end

Connecting with Rails

docs

Rails apps automatically connect to the database specified in the DATABASE_URL environment variable.

You can set this variable manually with flyctl secrets set

flyctl secrets set DATABASE_URL=postgres://postgres:secret123@postgresapp.internal:5432/yourdb

or by attaching the Postgres database to your Fly app.

Connecting with Python

psycopg is the recommended driver for connecting to postgres:

import psycopg

conninfo = "postgres://postgres:secret123@postgresapp.internal:5432/yourdb"
with psycopg.connect(conninfo) as connection:
    with connection.cursor() as cursor:
        cursor.execute(...)
        results = cursor.fetchall()

Connecting with Django

docs

psycopg is the recommended driver for connecting to postgres. You can use the django-environ package to translate the DATABASE_URL environment variable into Django database configuration. First, install both packages:

python -m pip install psycopg django-environ

Now we can define DATABASES in our settings.py file.

import environ  # ← Added

...

env = environ.Env()  # ← Added
environ.Env.read_env(BASE_DIR / ".env")  # ← Added

DATABASES = {
    # env.db() reads the DATABASE_URL environment variable.
    "default": env.db(),  # ← Updated
}

You can set the DATABASE_URL manually with flyctl secrets set:

flyctl secrets set DATABASE_URL=postgres://postgres:secret123@postgresapp.internal:5432/yourdb

or by adding it to the .env file:

# .env
SECRET_KEY=06\Rd75]S.vzZtrZoxc3!*~JR*/te[8BR%L-7~I7eFsGCqk,ze
DATABASE_URL=postgres://postgres:secret123@postgresapp.internal:5432/yourdb  # ← Added

Connecting with Go

docs

pgx is the recommended driver for connecting to postgres. It supports the standard database/sql interface as well as directly exposing low level / high performance APIs.

First, add github.com/jackc/pgx/v4 as a module dependency.

go get github.com/jackc/pgx/v4

The following program will connect to the database in DATABASE_URL and run a query.

package main

import (
  "database/sql"
  "fmt"
  "os"

  _ "github.com/jackc/pgx/v4/stdlib"
)

func main() {
  db, err := sql.Open("pgx", os.Getenv("DATABASE_URL"))
  if err != nil {
    fmt.Fprintf(os.Stderr, "Unable to connect to database: %v\n", err)
    os.Exit(1)
  }
  defer db.Close()

  var greeting string
  err = db.QueryRow("select 'Hello, world!'").Scan(&greeting)
  if err != nil {
    fmt.Fprintf(os.Stderr, "QueryRow failed: %v\n", err)
    os.Exit(1)
  }

  fmt.Println(greeting)
}

Connecting with Node.js

docs

You’ll use the pg npm module to connect to Postgres from a Node.js app.

const { Client } = require('pg')
const client = new Client({connectionString: process.env.DATABASE_URL})

await client.connect()
const res = await client.query('SELECT $1::text as message', ['Hello world!'])
console.log(res.rows[0].message) // Hello world!
await client.end()

Connecting with Prisma – Node.js

docs

Prisma is an open-source object-relational mapper (ORM) for Node.js and works with both JavaScript and TypeScript. It consists of 3 components:

  • Prisma Client - a type-safe query builder
  • Prisma Migrate - a data modeling and migration tool
  • Prisma Studio - a modern intuitive GUI for interacting with your database
Set up Prisma in your project

Install the Prisma CLI and Prisma Client dependencies in your project

npm i --save-dev prisma
npm i @prisma/client

Initialize Prisma in your project:

npx prisma init

This command does the following:

  • Creates a folder called prisma at the root of your project
  • Creates a .env file at the root of your project if it doesn’t exist
  • Creates a schema.prisma file inside the prisma folder. This is the file that you will use to model your data

Update the DATABASE_URL in the .env to your PostgreSQL database

DATABASE_URL="postgres://postgres:secret123@postgresapp.internal:5432/yourdb"

If you are working in a brownfield project, you can introspect your database to generate the models in your schema.prisma file:

npx prisma db pull

Assuming you have the following model in your schema.prisma file:

Add a model to your schema.prisma file:

datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")
}

generator client {
  provider  = "prisma-client-js"
}

model Post {
  id       Int     @id @default(autoincrement())
  title    String
  content  String?
}

You can query your database using Prisma as follows:

import { PrismaClient } from '@prisma/client'

const prisma = new PrismaClient()

async function main() {
  const posts = await prisma.post.findMany()

  const newPost = await prisma.post.create({
    data: {
      title: 'PostgreSQL on Fly',
      content: 'https://fly.io/docs/reference/postgres'
    }
  })
}

main()
  .catch((e) => {
    throw e
  })
  .finally(async () => {
    await prisma.$disconnect()
  })

Connecting with Sequelize – Node.js

docs

Sequelize is an open-source object-relational mapper (ORM) for Node.js and works with JavaScript and TypeScript.

Set up Sequelize inside your project

Install the Sequelize dependencies and CLI into your project

npm install --save sequelize
npm install --save-dev sequelize-cli

Initialize Sequelize into your project:

npx sequelize init

This will create the following folders:

  • config, contains config file, which tells CLI how to connect with database
  • models, contains all models for your project
  • migrations, contains all migration files
  • seeders, contains all seed files

Docs: Create your first model

After initializing you will need to show the CLI how to connect to the database inside of config/config.js

const fs = require('fs');

module.exports = {
  development: {
    username: 'database_dev',
    password: 'database_dev',
    database: 'database_dev',
    host: '127.0.0.1',
    port: 3306,
    dialect: 'mysql',
    dialectOptions: {
      bigNumberStrings: true
    }
  },
  test: {
    username: process.env.CI_DB_USERNAME,
    password: process.env.CI_DB_PASSWORD,
    database: process.env.CI_DB_NAME,
    host: '127.0.0.1',
    port: 3306,
    dialect: 'postgres',
    dialectOptions: {
      bigNumberStrings: true
    }
  },
  production: {
    database: process.env.DATABASE_URL,
    host: process.env.PROD_DB_HOSTNAME,
    dialect: 'postgres'
  }
};

Create a new file named .sequelizerc , this allows the CLI to be able to read your config.js file

const path = require('path');

module.exports = {
  'config': path.resolve('config', 'config.js')
}

Next pass the DATABASE_URL from the env into the Sequelize constructor in models/index.js to be able to connect to your fly postgres app and confirm the new seuqelize connection with the async function below

// Connect to fly postgres app
let sequelize = new Sequelize(process.env.DATABASE_URL);

// Confirms the connection to the database
(async () => {
  try {
    await sequelize.authenticate();
    console.log('Connected to the database!');
  } catch (err) {
    console.error('Error connecting to the database:', err);
  }
})();
Configure Sequelize to connect with your read replica databases

Fly docs: How to add Read Replica database with flyctl

Set the Sequelize constructor in model/index.js to conditionally connect to the primary/read-replica database depending on the primary region value set in fly.toml.

let sequelize;
if (config.use_env_variable) {
  const primary = process.env.PRIMARY_REGION;
  const current = process.env.FLY_REGION;
  let db_url = process.env.DATABASE_URL;

  if (!primary || !current || primary === current) {
    sequelize = new Sequelize(process.env.DATABASE_URL);

    // Check Fly logs to confirm that your primary database is connected
    console.log("DB connected in Region: ", current)
  } 
  else{
    const u = new URL(db_url);
    u.port = "5433";

    // Check Fly logs to confirm that your read replica database is connected
    console.log(`Read Rep DB connected in ${current}`)

    sequelize = new Sequelize(u.toString());
  }