On Thursday, 11/20/2008 at 03:16 EST, RPN01 <[EMAIL PROTECTED]> wrote: > That doesn't help, though, because the system names remain POLAR and > GRIZZLY, so there isn't a way to get the new TCPIP machines to read the > POLAR2 and GRIZZLY2 profiles, other than naming the virtual machines that > way. What I really wanted to do was to run the same virtual machine (TCPIP2) > on both systems, just as we run the primary TCPIP virtual machine on both.
I think I see the point. You want to share TCPMAINT 198 among all 4 stacks (POLAR.TCPIP, GRIZZLY.TCPIP, POLAR.TCPIP2, GRIZZLY.TCPIP2) and have each stack read a unique PROFILE, but the naming conventions as they exist don't let that work. Use the :Exit. tag to define an exec that will run a use the output of IDENTIFY to COPYFILE (OLDDATE an appropriate profile from the 198 to 191, with the name <userid> TCPIP. Use TCPRUNXT SAMPEXEC (?) as a skeleton, but don't call it "TCPRUNXT". That is the name of the global profile exit. See Ch 4 or 5 of the TCP/IP Planning. Do the COPYFILE only during the SETUP phase. So you would have 4 profiles: POLAR TCPIP, POLAR TCPIP2, GRIZZLY TCPIP, and GRIZZLY TCPIP2. The output of IDENTIFY would let you determine the correct file. Copying it to the A-disk as <userid> TCPIP ensures that it is the one TCPIP will read. Alan Altmark z/VM Development IBM Endicott
