Question

why does a client need the same endpoint as the service?
Is it that the client looks to the address which is in its endpoint, and the service will listen with the address that it has put in his endpoint???

Answer 1

The simple answer is that the service and client must agree on the mechanics of communication – that’s what the endpoint is for

• They must agree on the address – the service will listen for requests there and the client will send its requests there.
• They must agree on the binding – this is the “how” of the communication – what transport protocol will be used, what will messages look like on the wire and how are things like security configured
• They must agree on the contract as this is a definition of what functionality is available and what data is expected and potentially returned

In other words they must agree on the endpoint definition

However, things are a little more complex than that in reality

• the address is really a “logical address” the physical address the service listens on is defined by the listenUri which the client can send messages to by defining a clientVia endpoint behavior
• the binding must be compatible, the client can send additional data in its binding which isn’t intended for the service but rather, say, some routing intermediary
• the contracts must be compatible in that the WCF serialization engine must be able to render or deserialise the structure of the message based on its contract definition