Deploy Vendure to Railway

If you'd like to quickly get started with a ready-made Vendure project which includes sample data, you can clone our Vendure one-click-deploy repo.

Configuration

Port

Railway defines the port via the PORT environment variable, so make sure your Vendure Config uses this variable:

Database connection

Make sure your DB connection options uses the following environment variables:

Asset storage

In this guide we will use the AssetServerPlugin's default local disk storage strategy. Make sure you use the ASSET_UPLOAD_DIR environment variable to set the path to the directory where the uploaded assets will be stored.

Create a new Railway project

From the Railway dashboard, click "New Project" and select "Empty Project". You'll be taken to a screen where you can add the first service to your project.

Create the database

Click the "Add a Service" button and select "database". Choose a database that matches the one you are using in your Vendure project. If you are following along using the one-click-deploy repo, then choose "Postgres".

Create the Vendure server

Click the "new" button to create a new service, and select "GitHub repo". Select the repository which contains your Vendure project. You may need to configure access to this repo if you haven't already done so.

Configure the server service

You should then see a card representing this service in the main area of the dashboard. Click the card and go to the "settings" tab.

• Scroll to the "Service" section and rename the service to "vendure-server".
• Check the "Build" section and make sure the build settings make sense for your repo. If you are using the one-click-deploy repo, then it should detect the Dockerfile.
• In the "Deploy" section, set the "Custom start command" to node ./dist/index.js.
• Finally, scroll up to the "Networking" section and click "Generate domain" to set up a temporary domain for your Vendure server.

Create a Volume

In order to persist the uploaded product images, we need to create a volume. Click the "new" button and select "Volume". Attach it to the "vendure-server" service and set the mount path to /vendure-assets.

Configure server env vars

Click on the "vendure-server" service and go to the "Variables" tab. This is where we will set up the environment variables which are used in our Vendure Config. You can use the raw editor to add the following variables, making sure to replace the highlighted values with your own:

Create the Vendure worker

Finally, we need to define the worker process which will run the background tasks. Click the "new" button and select "GitHub repo". Select again the repository which contains your Vendure project.

Configure the worker service

You should then see a card representing this service in the main area of the dashboard. Click the card and go to the "settings" tab.

• Scroll to the "Service" section and rename the service to "vendure-worker".
• Check the "Build" section and make sure the build settings make sense for your repo. If you are using the one-click-deploy repo, then it should detect the Dockerfile.
• In the "Deploy" section, set the "Custom start command" to node ./dist/index-worker.js.

Configure worker env vars

The worker will need to know how to connect to the database, so add the following variables to the "Variables" tab:

Test your Vendure server

To test that everything is working, click the "vendure-server" card and then the link to the temporary domain.

Next Steps

This setup gives you a basic Vendure server to get started with. When moving to a more production-ready setup, you'll want to consider the following:

• Use MinIO for asset storage. This is a more robust and scalable solution than the local disk storage used here.
  • MinIO template for Railway,
  • Configuring the AssetServerPlugin for MinIO
• Use Redis to power the job queue and session cache. This is not only more performant, but will enable horizontal scaling of your server and worker instances.
  • Railway Redis docs
  • Vendure horizontal scaling docs