Current Situation

I'm using two-way webrtc streaming service on UV4L.

As a default, UV4L server is deployed on raspberry pi.

We can connect to UV4L's websocket server through raspberry pi's local ip such as wss://

enter image description here

Expectation & Purpose

But I need to make UV4L server deploy on outside server with public ip such as ec2. I want to use that ec2 server as a broker.

So wherever a user is, he can connect to RPI's WebRTC streaming server.
I think this can be an alternative of port forwarding.

enter image description here


Is this idea possible? And how can I implement it?

I set an ec2's ip on server option '--bind-host-address=13.125.xxx.xxx', but it doesn't work. I think additional handling on server side is needed

1 Answer 1


Since UV4L is not a WebRTC SFU itself, I do not think that makes sense. To do what you want you will have to deploy the Janus Gateway SFU on your public EC2 instance. UV4L supports Janus videorooms. In other words the pipeline will become: Rpi (UV4L) <--> Janus Gateway SFU on EC2 (Videoroom) <--> WebBrowsers.

  • Does Janus Gateway SFU video room supports two way audio/video streaming?
    – Josh
    Commented Jan 11, 2018 at 23:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.