Problems with server-side in production

Hello! As mentioned in the title, I am having some problems with the server-side of the production build. This is to underline that there are not problems in the logic, which in development works correctly.

I have a server running a websocket registered in the server.config.js file; so the standard way to connect to it would be to hit that endpoint at localhost:8911/ws; however, in production the server is secured, so that localhost:8911 is closed and so that endpoint is also closed and my Context on the front end cannot do new WebSocket('ws://localhost:8911/ws'). How can I get around this problem? I can see that GraphQL itself to work in production actually turns its requests to localhost:8910/.redwood/functions/graphql, so maybe I should setup something similar? Also, I put in the redwood.toml file in the api field the path to the server config file, as mentioned in the documentation. It did not work

Solved it. In production, by default, the Fastify routes (like /ws) are actually exposed to localhost:8910. This may be worth mentioning in the docs, since it was not really clear to me