Skip to main content
Skip to main content

Deploy Your Medusa Backend on Heroku

In this document, you'll learn how to deploy your Medusa backend on Heroku. Heroku is a PaaS (Platform as a Service) that allows you to easily deploy your applications in the cloud.

Alternatively, you can use this button to deploy the Medusa backend to Heroku directly:

Deploy to Heroku


Medusa Backend

It is assumed that you already have a Medusa backend installed locally. If you don’t, please follow the quickstart guide.

Furthermore, your Medusa backend should be configured to work with PostgreSQL and Redis. You can follow the Configure your Backend documentation to learn how to do that.

Needed Accounts

Required Tools

Deploy to Heroku

1. Login to Heroku from your CLI

Before you can create an app with Heroku, you must login with the CLI tool:

heroku login

Depending on your operating system, you must follow either the instructions in your terminal or a page in your browser will open.

2. Create an App with Heroku

In the root directory of your Medusa backend, run the following commands to create an app on Heroku and add it as a remote origin:

heroku create <APP_NAME>
heroku git:remote -a <APP_NAME>

Where <APP_NAME> is the name of the app you'll create. You can use any name you want.

3. Install Postgresql and Redis on Heroku

Medusa requires a Postgres database and a Redis instance to work. You can add those to your Heroku app using Add-ons.


If you don't have a payment method set up in your Heroku account, you'll be asked to enter your payment details when you try to install these addons.


Add a Postgres add-on to your Heroku app with the following command:

heroku addons:create heroku-postgresql:mini

This uses Heroku Postgres's smallest plan. You can check out the available plans and pricing of Heroku Postgres on Heroku's website.


Add a Redis instance to your Heroku app with the following command:

heroku addons:create stackhero-redis:ist-m4euc0

This uses the lowest plan in Stackhero Redis. You can check out the plans and pricing of Stackhero Redis on Heroku's website.

4. Configure Environment Variables on Heroku

Medusa requires a set of environment variables to be configured. You can learn more about Medusa's configurations in the Configure your Medusa backend document.

Run the following commands in the root directory of your Medusa backend to set some environment variables:

heroku config:set NODE_ENV=production
heroku config:set JWT_SECRET=your-super-secret
heroku config:set COOKIE_SECRET=your-super-secret-pt2
heroku config:set NPM_CONFIG_PRODUCTION=false

Make sure to replace your-super-secret and your-super-secret-pt2 with actual secrets in a production environment.

Set Buildpack

Additionally, you need to set the buildpack to Node.js using the following command:

heroku buildpacks:set heroku/nodejs

Configure the Redis URL

Stackhero Redis adds the Redis URL under the environment variable STACKHERO_REDIS_URL_TLS. However, Medusa looks for the REDIS_URL environment variable when initializing the connection with Redis.

Retrieve the value of STACKHERO_REDIS_URL_TLS with the following command:

heroku config:get STACKHERO_REDIS_URL_TLS

This prints the value of the environment variable which is a Redis connection string.


If you see nothing, you have to wait until the creation process of the stackhero redis instance is done. Try waiting a few minutes then trying again.

Copy that value and use it to set the environment variable REDIS_URL with the following command:

heroku config:set REDIS_URL=<YOUR_REDIS_URL>

Where <YOUR_REDIS_URL> is the value you received from the previous command.

Configure the PostgreSQL Database URL

If you're using the Heroku PostgreSQL Add-on, it should configure the environment variable DATABASE_URL. In that case, you don't need to perform any additional actions.

However, if you use another add-on, make sure to set the environment variable DATABASE_URL to the PostgreSQL Database URL.

(Optional) Configure Modules Environment Variables

If you use modules in your Medusa backend that require setting environment variables, then you should set them at this point.

For example, if you use the Redis Event Bus module:


Make sure to change EVENTS_REDIS_URL to the environment variable name you use for your module's configurations.

If your module requires setting up other services, do that then add the environment variables.

(Optional) Configure CORS Variables

Optionally, if you've deployed the admin dashboard and you want to ensure it can use the backend's REST APIs, you must set the following environment variable:

heroku config:set ADMIN_CORS=<YOUR_ADMIN_URL>

Where <YOUR_ADMIN_URL> is the URL of your admin dashboard.

Similarly, if you've deployed the storefront and you want to ensure it can use the backend's REST APIs, you must set the following environment variable:


Where <YOUR_STOREFRONT_URL> is the URL of your storefront.

5. Configure Medusa Backend

Before jumping into the deployment, you must configure your Medusa backend to use the previous environment variables and the recommended production configurations.


Update module.exports to include the following configurations:

module.exports = {
projectConfig: {
redis_url: REDIS_URL,
database_url: DATABASE_URL,
database_type: "postgres",
store_cors: STORE_CORS,
admin_cors: ADMIN_CORS,
process.env.NODE_ENV !== "development"
? { ssl: { rejectUnauthorized: false } }
: {},


Update scripts to include the following scripts:

"scripts": {
"serve": "medusa start",
"start": "medusa develop",
"heroku-postbuild": "medusa migrations run",
"prepare": "npm run build",
"build": "babel src -d dist --extensions \".ts,.js\""

(Optional) 6. Configure the Admin

If you're using the Medusa Admin plugin, you have two options to deploy it: either with the backend or separately.

Deploying with the Backend

To deploy the admin with the backend:

  1. Your chosen plan must offer at least 2GB of RAM.
  2. Enable the autoRebuild option of the admin plugin:
const plugins = [
// ...
resolve: "@medusajs/admin",
/** @type {import('@medusajs/admin').PluginOptions} */
options: {
autoRebuild: true,
// other options...

Alternatively, you can use a GitHub action to build the admin as explained here.

Deploying Separately

If you choose to deploy the admin separately, disable the admin plugin's serve option:

const plugins = [
// ...
resolve: "@medusajs/admin",
/** @type {import('@medusajs/admin').PluginOptions} */
options: {
// only enable `serve` in development
// you may need to add the NODE_ENV variable
// manually
serve: process.env.NODE_ENV === "development",
// other options...

This ensures that the admin isn't built or served in production. You can also change @medusajs/admin dependency to be a dev dependency in package.json.

You can alternatively remove the admin plugin for the plugins array.

7. Launch your Medusa Backend

Finally, commit and push all changes to Heroku:

git add .
git commit -m "Deploy Medusa Backend on Heroku"
git push heroku HEAD:master

This triggers a redeploy of the Medusa backend with all the new configurations.

Test your Backend

To test your backend, run the following command to retrieve the backend's URL:

heroku apps:info -a <APP_NAME>

Where <APP_NAME> is the name of the app. You should see as the output a bunch of info of the app.

The backend's URL is available under "Web URL". You can copy it and perform requests to it to test it out.

For example, you can send a request to <YOUR_BACKEND_URL>/store/products and you should receive a JSON response with the products in your store.

Health Route

You can access /health to get health status of your deployed backend.

Testing the Admin

If you deployed the admin dashboard with the backend, you can test it by going to <YOUR_APP_URL>/app. If you changed the admin path, make sure to change /app to the path you've set.


Inspect Build Logs

If an error occurs during the deployment, you can explore your Heroku app build logs using the following command in the root directory of your Medusa backend:

heroku logs -n 500000 --remote heroku --tail -a <APP_NAME>

Where <APP_NAME> is the name of the app.

Error: Babel not found

If you get the following error in the logs of your application:

/bin/sh: 1: /app/node_modules/.bin/babel: not found

You can resolve it by creating a file in the root of your Medusa backend directory named Procfile (without an extension) with the following content:

web: npm run serve

Then, push the changes with Git:

git add .
git commit -m "Added Procfile"
git push heroku HEAD:master

Once the application is re-published, you can access your store as expected.

Run Commands on Your Backend

To run commands on your backend, you can use the following command:

heroku run -a <APP_NAME> -- <COMMAND>

Where <APP_NAME> is the name of the app and <COMMAND> is the command you want to run.

For example, to create an admin user you can run the following command:

heroku run -a <APP_NAME> -- npx medusa user -e "<EMAIL>" -p "<PASSWORD>"

Where <APP_NAME> is the name of your Heroku app, and <EMAIL> and <PASSWORD> are the credentials you want to use to log in to the Medusa Admin dashboard.

Add Environment Variables

You’ll likely need to add environment variables later such as Admin Cross-Origin Resource Sharing (CORS) and Store CORS variables.

To set or change an environment variable's value, you can use the following command:

heroku config:set <ENV_NAME>=<ENV_VALUE> -a <APP_NAME>

Where <APP_NAME> is the name of your Heroku app, <ENV_NAME> is the name of the environment variable, and <ENV_VALUE> is the value.

See Also

Was this section helpful?