AIvoices/server-deno

Elato AI WebSocket Server (Deno)

This directory contains the WebSocket server implementation for Elato AI, built with Deno.

Prerequisites

• Deno installed on your system
• Network access (for both local development and deployment)

Running the Server

Local Development

1. Start the server:

   # Navigate to the server-deno directory
   cd server-deno
   
   # Run the server (replace server.ts with your main file if different)
   deno run -A --env-file=.env main.ts
   

2. Verify the server:

   • The server should be running on port 8000 (default)
   • You should see a message confirming the server is listening

Production Deployment

1. Deploy to Deno Deploy:

   • Create an account on Deno Deploy
   • Create a new project
   • Connect your GitHub repository or upload your server code
   • Deploy to a custom domain (e.g., your-app.deno.dev)

2. Update firmware configuration:

   • In the firmware project, make sure DEV_MODE is NOT defined in Config.cpp
   • The WebSocket server will be set to your Deno Deploy URL

3. Get the Root CA Certificate using terminal:

   • To get the full certificate chain with the root CA:

   openssl s_client -showcerts -connect your-app.deno.dev:443 </dev/null
   

   • Copy the last certificate in the chain (the root CA certificate) including the BEGIN and END lines
   • Paste this into the CA_cert variable in Config.cpp

Troubleshooting

Connection Issues

• Ensure your ESP32 device and development computer are on the same network
• Check if any firewalls are blocking port 8000
• Verify your local IP address hasn't changed (dynamic IPs can change after reconnecting)

Server Errors

• Check the server logs for error messages
• Ensure Deno has proper permissions to access network resources

Additional Resources

• Deno Documentation
• Deno Deploy Documentation
• WebSocket API Reference