I am using kurento to record the stream to the server-disk. I followed the change of the kurento tutorial here where the tutorial hello-world in node.js was changed to record the stream to the disk. Now what I want to do is change the tutorial4 one-to-one call to record the caller and the callee stream to 2 files.
As I understanded kurento the recorder is linked to the pipeline. The pipeline is the representation of the connection between 2 peers. How can I record the stream of a single peer in the pipeline, respectively both streams the one of the caller and the one of the callee?
I tried a lot but could not find a sollution.
Perhaps you should be reading the basic Kurento documentation introducing what's a media element and what's a pipeline to have a more clear image on how Kurento's API work.
For you specific question, a RecorerEndpoint is just media element capable of recording to a file the stream offered as its input. This means that you can connect a RecorderEndpoint to any other source element. For example, if you have a B2B call among two peers, every peer will have associated a WebRtcEndpoint so that the resulting pipeline will be something like this:
Peer A <--------->webRtcEndpointA<==>webRtcEndpointB<---------->Peer B
For recording the streams coming from both Peer A and Peer B, you just need to create two RecorderEndpoints and connect them appropriately so that the final pipeline is something like:
------>recorderEndpointA
|
Peer A <--------->webRtcEndpointA<==>webRtcEndpointB<---------->Peer B
|
--->recorderEndpointB
The source code for this should include (I ommit the boilerplate code that you already have in the tutorials you cite):
pipeline.create('RecorderEndpoint', {uri: "file:///tmp/fileA.webm"}, function(error, recorderEndpointA ...
...
webRtcEndpointA.connect(recorderEndpointA);
recorderEndpointA.record();
pipeline.create('RecorderEndpoint', {uri: "file:///tmp/fileB.webm"}, function(error, recoderEndpointB ...
...
webRtcEndpointB.connect(recorderEndpointB);
recorderEndpointB.record();