We deploying ToIP in some of our remote offices with QoS (FR traffic-shaping+CBWFQ+LLQ, etc.), and in some of them, we are thinking to configure rtp header compression. However, I have some doubts. I hope I will be able to explain them properly
These are my questions:
1. "show policy-map" command shows offered rate for different traffic classes. If we configure rtp header compression for VoIP traffic class, show policy-map offered rate for VoIP will reflect offered rate with compression? I think the answer is not. What do you think?
2. If the answer is not for previous question, rtp header compression could be a problem. If I configure LLQ for VoIP with the sum of all compressed simultaneous voice channels and there is congestion in output interface, router could discard VoIP traffic with LLQ queue filled before to reach all simultaneous voice channels. Am I in the right way or I'm wrong?
3. Now one question about CBWFQ. If there is no congestion in output interface QoS policy will not work and traffic classes (VoIP and not VoIP) will use required bandwidth, even greater than configured with bandwith or priority commands. If there is congestion in output interface, I know traffic classes configured with bandwith commands will use unused bandwith from other classes but, from VoIP class configured with LLQ priority command too?