Summary of passing the socket process

The process of passing the socket is accomplished in the following way:

After creating a subtask, the server main task issues the GIVESOCKET call to pass the socket to the subtask. If the subtask’s address space name and subtask ID are specified in the GIVESOCKET call, (as with CICS®) only a subtask with a matching address space and subtask ID can take the socket. If this field is set to blanks, (as with IMS™) any MVS™ address space requesting a socket can take this socket.
The server main task then passes the socket descriptor and concurrent server’s ID to the subtask using some form of commonly addressable technique such as the IMS Message Queue.
The concurrent server issues the SELECT call to determine when the GIVESOCKET has successfully completed.
The subtask calls TAKESOCKET with the concurrent server’s ID and socket descriptor and uses the resulting socket descriptor for communication with the client.
When the GIVESOCKET has successfully completed, the concurrent server issues the CLOSE call to complete the handoff.

An example of a concurrent server is the IMS Listener. It is described in How to customize and operate the IMS Listener. Figure 2 shows a concurrent server.