webrtc stats

[sambit swain]

in a webrtc connection, Is it possible that between 2 peers one side is on stun and the other side is on relay?

I am getting this from Google stats. Please see the attached image for the stats of the 2 peers.

[Benjamin Schwartz]

Yes, that is possible.

On Mon, Apr 7, 2014 at 12:50 AM, sambit swain <99.s...@gmail.com> wrote:

in a webrtc connection, Is it possible that between 2 peers one side is on stun and the other side is on relay?

I am getting this from Google stats. Please see the attached image for the stats of the 2 peers.

--

---
You received this message because you are subscribed to the Google Groups "discuss-webrtc" group.
To unsubscribe from this group and stop receiving emails from it, send an email to discuss-webrt...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Fred Clark]

In a situation like this where the connection is relay from A to B and Stun from B to A the nomenclature of the Google Stats is confusing to me.  It just doesn't seem to be consistent with what I would expect.  Use the example that is posted and also the example of a connection using either the Local method or the Stun method. In this later example the local IP address is the local private IP of the local side and the remote IP address is the public IP of the remote side.  This is all very understandable to me and seems consistent with what I would expect.

Now take this posted example.  On the relay side the local IP is the private IP of the local side and the remote IP is the relay server.  This seems very reasonable.  On the receiving side (B in this case and a STUN connection), the local IP is the relay server and the remote IP is the public IP of the remote side.  I don't understand the logic behind the local IP of side B,.  It seems inconsistent.  What am I missing?