Question / Help Self-Sufficient Streaming

[caruban]

Hi
I have a fairly powerful computer running Windows 7. I have OBS, Red5, and a static IP on which I have a website. I'm having trouble connecting all of these things. My goal is to have an embedded player that anyone can watch by visiting http://www.mywebsiteurl.com/stream. So far, I have Red5 running with the default configuration, and my OBS broadcast settings are set to custom, with the FMS URL set to rtmp://localhost:1935 (the default port). When I try to start streaming, I'm told my stream key is incorrect (which is currently blank, though I've tried some values). Can anyone guide me through setting this up?

Thanks

 

[dodgepong]

I've never used Red5, but I've used crtmpserver and nginx's RTMP module (I prefer the latter). With those systems, by default you just "invent" a stream key and enter it into the stream key box in OBS. So for me, if my FMS URL is rtmp://myserver/live and I enter a stream key of "test", then I feed the RTMP player (such as JWPlayer) a URL of rtmp://myserver/live/test and that works.

I'm guessing your problem is some setting on your Red5 server. I can't be certain, having never used it though. If all you're doing is what you're describing, then I recommend checking out nginx with the RTMP module for its lighter weight.

 

[caruban]

Thank you, though I'm having some trouble setting up the RTMP module. Could you possibly give me a quick walk through on how to do that?

 

[dodgepong]

Well, I'm running my RTMP server on a Raspberry Pi, so it's using the Raspbian distribution (which, for all intents and purposes, is Debian). There is a package in apt for installing nginx, but the instructions on the RTMP module site seemed to indicate that you had to add the module at compile time, and I don't believe the apt package came with the module precompiled. So I ended up downloading the nginx source and compiling it manually, adding the module on the ./configure line. I basically just ran the 3 commands listed on the nginx-rtmp-module readme: https://github.com/arut/nginx-rtmp-modu ... /README.md

./configure --add-module=<path-to-nginx-rtmp-module>
make
make install

I basically just followed that README for setup instructions, tbh. After installing, I ran nginx and then added this to /usr/local/nginx/conf/nginx.conf:

rtmp {
        server {
                listen 1935;
                chunk_size 4096;

                application live {
                        live on;
                        record off;
                }
        }
}

What that does is tells nginx to load the rtmp module, listen on port 1935 for that module (1935 = RTMP port), and then make an "application" (what you append to the server URL, i.e. rtmp://servername/<application name>, so "live" in my case). "live on" means I'm telling it to enable streaming, and "record off" tells the server not to store a recording of the stream. You, of course, can read the README and edit the conf options to your liking.

If you've never managed a Linux server or compiled something from source on Linux...then you probably didn't understand any of this and I probably have more explaining to do :|

 

[caruban]

I've done enough work in linux to know what that means. I gathered most of that from the readme, though your discription was much more clear.

Problem is, I'm in Windows. XD Everything after compiling the source is transferable, I'd imagine. I guess that means I need to learn to compile source code in Windows and add the module to it...

Thank you so much for your help.