Interval Server is the central node used to run applications developed with the Interval SDK.

🚧 Note: Previously Interval Server was only available as a closed-source cloud service. Now, it is possible to run an instance yourself. This is a new direction for the Interval project and the fully open-source/self-hosted Interval Server application is still in early phases of development. If you encounter an issues setting up an Interval Server instance, please let us know by opening an issue!

Interval Server requires a Postgres database to work. In the future, especially for local development, we may ease this requirement.

We have tested Interval Server with Postgres versions 11.x and 12.x. Newer versions should work, but we do not plan to support anything older than 11.x.

Interval Server is a pure Node.js application. Node.js version 16 or higher is required to run Interval Server.

• Postmark is used for sending application emails. In the future we may introduce a vendor-agnostic path for sending emails. If a POSTMARK_API_KEY environment variable is not provided when running Interval server, emails will not be sent. Use the EMAIL_FROM environment variable to customize the email from field. The default value is Interval <[email protected]>.
• WorkOS is used for SSO, directory sync, and Sign in with Google. If WORKOS_API_KEY,WORKOS_CLIENT_ID, and WORKOS_WEBHOOK_SECRET environment variables are not provided when running Interval Server, these functions will not be available.
• Slack can be used to send notifications via Interval's notify methods. If SLACK_CLIENT_ID and SLACK_CLIENT_SECRET environment variables are not provided when running Interval Server, notifications cannot be sent via Slack.
• S3 or S3-compatible storage providers (like MinIO or Cloudflare R2) can be used to support file uploads via Interval's file input methods. If S3_KEY_ID,S3_KEY_SECRET,S3_BUCKET,S3_REGION, and S3_ENDPOINT environment variables are not provided when running Interval Server, file uploads will not function properly. For S3-compatible providers, make sure to set the appropriate endpoint URL in S3_ENDPOINT.

To support file uploads, you will need to configure your S3 bucket for Cross-origin Resource Sharing. Here's an example policy:

[
    {
        "AllowedHeaders": [
            "*"
        ],
        "AllowedMethods": [
            "PUT",
            "POST",
            "DELETE"
        ],
        "AllowedOrigins": [
            "https://your-interval-server.com"
        ],
        "ExposeHeaders": [
            "x-amz-server-side-encryption",
            "x-amz-request-id",
            "x-amz-id-2"
        ],
        "MaxAgeSeconds": 3000
    }
]

• APP_URL is the URL where your Interval Server instance is running. For example: http://localhost:3000 or https://example.com.
• DATABASE_URL is the Postgres connection string. It should follow the format postgresql://username:password@host:port/dbname.
• SECRET is a secret that you must provide for use in encrypting passwords. Any string is valid for this value, but you should use something secure!
• WSS_API_SECRET is a secret that you must provide. It is used internally by Interval Server for communication between Interval services. Any string is valid for this value, but you should use something secure!
• AUTH_COOKIE_SECRET is a secret that you must provide for use in encrypting session cookies. Any string at least 32 characters in length is valid for this value, but you should use something secure!

Interval Server runs services on port 3000.

For development, you may wish to run an instance of Interval Server locally.

1. npm i -g @interval/server
2. From the directory where you would like to run Interval Server, create a .env file like this:

DATABASE_URL=<YOUR DATABASE URL>
SECRET=<YOUR SECRET VALUE>
APP_URL=<YOUR APP URL>
AUTH_COOKIE_SECRET=<YOUR AUTH COOKIE SECRET>
WSS_API_SECRET=<YOUR WSS API SECRET>

Note: you don't need to use a .env file. As long as the required variables are set, you should be good to go.

3. Run interval-server start to run interval-server.
4. 🎉 Visit http://localhost:3000 to access your Interval Server!

Running Interval Server in production is largely the same as running in development. For convenience, we've created a Docker image to make this even easier.

The Interval Server Docker image is: docker.io/alexarena/interval-server:latest.

Many services like Render make it trivial to deploy Docker images with just a few clicks.

Important things to know:

• You'll still need to provide all required environment variables when running the Interval Server Docker image.
• Hosting providers like Render will automatically discover the Interval Server service running on port 3000 and will expose this port for you, but if your hosting provider doesn't do this, you'll have to handle exposing this yourself.

Once your Interval Server instance is up and running, it's trivial to connect to it from your Interval apps. Just add an endpoint property pointing to your Interval Server instance to the Interval SDK's constructor. Interval server also prints the URL on startup in case you need it. For example:

const interval = new Interval({
  apiKey: process.env.INTERVAL_KEY,
  endpoint: 'wss://<YOUR INTERVAL SERVER URL>/websocket', // Don't forget the /websocket path!
})

Note: if you're running Interval Server locally, this URL will use the insecure ws:// protocol, not the secure wss:// version used in production deployments.

Once you run npm i -g @interval/server, the following commands are available:

Starts Interval Server. See above for information on running Interval Server locally or in production.

For our initial release, we're focused on making it easy to setup and run your own Interval Server instance. We'll make it easier to contribute (and document how you can) in the future, but for now we aren't actively soliciting new contributions.