When allocations occur during NIP, particularly within ASID=1, they are not normal allocations. They allocate the volume. This is referred to in the code as "NIP MOUNT" (and later there is "NIP OPEN"). If I remember correctly, in some cases that persists for the life of the IPL (the device is marked allocated and remains so), and there is nothing that can be done about that for functional and serialization reasons. This might apply only to PDSEs. Did the OP's case have any LNKLST PDSEs on the volume in question (I hope the volume is not the SYSRES volume; I would not expect to be able to take that offline). Once dynamic allocation is available (around the time of master scheduler initialization), this is no longer relevant. If a volume is marked allocated, then it cannot be taken offline. I thought this was documented somewhere. Maybe it is, but I can't find it now.
LNKLST allocations, for example, go through that processing, whether from LNKLSTxx or PROGxx, for the IPL-time LNKLST. Subsequently the LNKLST data sets are allocated within XCFAS and, generally, LLA. It is those allocations that can be gotten rid of. Those are the allocations that would prevent you from getting the data set allocated DISP=OLD if you needed to work with an uncataloged data set of the same name. And those are the ones that it is OK to (temporarily) get rid of. Peter Relson z/OS Core Technology Design ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
