-
Introduction As far as I understand I've been using TCPRelay then since that was in the default config. I only changed some minor things. Here's the most relevant parts of my configuration (I removed some of the last parts in here as I don't think that part is relevant): Variables:
Describe the bug Recently I moved to a new place where I don't have access to an internet provider that can provide me with a public IP address at all. So I'm basically behind double NAT. A friend of mine bought multiple IP Addresses though so we decided to setup a Tunnel and route all traffic through that IP Address to me through that Tunnel. It was a bit of a hassle to setup but also after a bit I managed to make OSP working again. I then started to try to get OME working again as well but I keep running into issues. Since I still don't quite understand what ports I do need to forward and which ones are internal I decided to forward the following ports:
I keep running into multiple issues with this setup: 1. Could not parse STUN packet
When I check my docker logs I see this message all over the place. I'm sure this wasn't the case before. From what I understand a STUN server helps determine the public IP Address of OME. Since I'm behind a double NAT and we should actually be using my friends IP Address I changed the following config sections:
And I did not change:
I have been playing around with TcpForce 2. wss connection in browser doesn't seem to connect When I start streaming the logs seem to show correcty that OME has accepted the stream and started the WebRTC publisher. If I go to the streaming page then, I see the following logs in my Chrome Dev Tools: And this is what the logs show on Docker: To Reproduce Expected behavior Logs
Server (please complete the following information):
Player (please complete the following information):
Additional context |
Beta Was this translation helpful? Give feedback.
Replies: 6 comments 8 replies
-
I did some more digging to see if all ports are open. So with this command I ran a server on the docker host:
Then with this command I tried to connect to it and send some data:
I can't really explain why but all ports work consistently except port 10001, this is the output of the server after running the client command:
For now I've changed the OME config to use ports 10002/udp to 10005/udp since those are confirmed working. Well either way, I've confirmed that all ports are open from the internet to the docker host in question. |
Beta Was this translation helpful? Give feedback.
-
When Is the connection to FriendIP:3334/tcp and FriendIP:3478/tcp good from outside and inside? |
Beta Was this translation helpful? Give feedback.
-
Hi @getroot , Yes it is. In my reply jusylt above this comment I've noted down all ports that are forwarded and how I verified if they actually are. Anything else that could be going wrong? Can you maybe explain a bit around what part is not working for me? Because it seems at least some messages are flowing through. Maybe it would help to compare the docker logs of a working server with mine? Devedse |
Beta Was this translation helpful? Give feedback.
-
Here's my full Server.xml:
|
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
Is there a way to figure out if the client is using UDP or TCP btw? |
Beta Was this translation helpful? Give feedback.
Ok another update, I managed to resolve the issue. Apparently by changing Keyframes to 1-sec instead of Automatic it now works again.
My resolution is 3440x1440 and here's my other settings:
I don't really understand why but this seems to be the fix consistently.