Is CFF7LINK required/used by any other task/address space besides TSSMFA ?
So if there are no batch jobs submitted etc that call the programs in the CFF7LINK library, only the TSSMFA address space, then the library won’t require linklisting if it’s in the APF list and TSSMFA is steplibbing to it ???
Also , have seen software that will dynamically define an SSI if none exists already … should there be a definition for the subsystem in SYS1.PARMLIB(IEFSSIxx)?
With no STEPLIB card, TSSMFA only fails on the first startup after IPL, but comes up successfully when restarted anytime thereafter. Please explain why? Is this by design ?
If CFF7LINK is in the linklist, and the linklist is APF’d, why does the first startup at IPL time fail, but then TSSMFA can be started up successfully any number of times afterwards
Release : 16.0
Component : CA Top Secret for z/OS
No.
2. So if there are no batch jobs submitted etc that call the programs in the CFF7LINK library, only the TSSMFA address space, then the library won’t require link listing if it’s in the APF list and TSSMFA is step-lib to it?
Correct.
3. Have seen software that will dynamically define an SSI if none exists already … should be a definition for the subsystem in SYS1.PARMLIB(IEFSSIxx)?
AAM installs its SSI dynamically, and no need for PARMLIB update.
4. With no STEPLIB card, TSSMFA only fails on the first startup after IPL, but comes up successfully when restarted anytime thereafter. Please explain why? Is this by design? If CFF7LINK is in the linklist, and the linklist is APF’d, why does the first startup at IPL time fail, but then TSSMFA can be started up successfully any number of times afterwards.
AAM has an end-of-memory routine called MAARTTRM. This needs to be in the STEPLIB. We don't try to find it in the linklist. It gets planted during the first AAM Subsystem Initialization. Once it's planted, it's there for the life of the IPL, we don't try to plant it again.