I'm listening using my own RTP socket implementation and receiving and parsing the packets including the version, padding, extension, csrc count, marker, payload type, sequence number, time-stamp, ssrc and the data itself.
The codec are H.264 and G.722.
Basically I use similar code to publish the stream:
Wowza does not use RTP packets directly as you are trying to do. It first passes them through a depacketizer that reassembles them into AMFPackets.
RTP video packets are normally fragments of complete video packets and the timecodes are based on video clock cycles (90K clock). RTP audio packets have their timecodes based on the sample rate. AMFPacket timecodes are based on wall clock values and are in milliseconds.
If you push your RTP stream into a pair of ports that wowza is listening to (use an rtp mediaCaster on wowza to connect to the source stream), then Wowza will be able to create an inbound stream and depackatize your rtp stream into AMFPackets. You can then monitor the packet list of the stream and get the AMFPackets. Use stream.getPlayPackets() to get the currently available packets that you can insert into your publisher stream. You may need to provide an sdp file for Wowza to know how to read your source stream
As each packet is inserted into the inbound stream, it is allocated a sequence number. You need to track this sequence number to make sure you are not reinserting the same packets into the publisher stream.
We are not really sure what it is that you are wanting to do and I think the best approach will be to open a ticket with
email@example.com. Please explain in detail what you are wanting to do. Also, please link to this thread.