SNAX Connection Manager (SNAX/CM) Manual

Running SNAX/CM
Introduction to SNAX Connection Manager (SNAX/CM)
1–8 097842 Tandem Computers Incorporated
Running SNAX/CM Interactively
In Figure 1-2, SNAX/CM is run on demand as follows:
1. A TACL process sends a message to a SNAX/CM process, which calls internal
SNAX/CM procedures to request connection manager operations.
2. When a SNAX/CM operation results in a SNAX/XF switched line being
connected, SNAX/CM sends a message to a user-written server process. In this
illustration, the server is a customer-provided process that accepts a standard
SNAX/CM CONNECTED message.
3. The server calls a NEWPROCESS procedure to start the data application.
4. The data application opens an LU on the connected line.
Running SNAX/CM Programmatically
Figure 1-2 also shows how a user-written application program can run SNAX/CM
programmatically:
1. The application opens SNAX/CM through the server and issues a WRITEREAD
call with text message commands in the buffer. (Text should be formatted using
the syntax defined in Section 3, "SNAX/CM Commands.")
2. The application issues a WRITEREAD request to send text commands to
SNAX/CM.
3. SNAX/CM receives the WRITEREAD text command from its $RECEIVE file and
issues the appropriate connection management command to SNAX/XF.
4. When a SNAX/XF switched line is connected, SNAX/CM sends a structured
message to the server.
5. The server process passes the LU aliases to the data application, which passes it to
SNAX/HLS.
6. SNAX/HLS opens an LU on the connected line so that the user application can use
the established connection.
Tracing SNAX/CM Tracing is implemented through the SNAX/CM TRACE command, described in the
next section. To format a trace, use the PTrace utility. Typically, you will also need to
trace SNAX/XF lines using the SCF TRACE command.