An NCP can be shared by more than one SSCP. A VARY ACT command for an NCP causes the VTAM® SSCP to activate the NCP, establishing an SSCP-PU session with the NCP.

Owning resources

If multiple VTAMs activate an NCP, each of those VTAMs establishes an SSCP-PU session with the NCP, and each is considered an owner of that NCP. Each VTAM can also activate and share ownership of the NCP links and link stations. However, the NCP physical units and logical units can belong to only one VTAM at a time. The VTAM that activates a physical unit first becomes the exclusive owner of that physical unit; the other VTAM cannot activate that physical unit until the SSCP that currently owns the physical unit deactivates it.

Defining owners

The resources of an NCP that cannot be shared can be explicitly divided among the VTAMs that own the NCP. You can use the OWNER operand in NCP definition statements to explicitly identify the VTAM that is to own a particular resource. If a resource is allocated to a particular VTAM by the OWNER operand, that resource can be controlled only by operator commands entered from that VTAM. Therefore, VARY ACT and VARY INACT commands can be entered for these resources from the VTAM that owns them, but not from any of the other VTAMs that share the NCP. The OWNER operand can be specified on the PCCU, GROUP, LINE, and PU definition statements.

Backup owner

You can also specify that the NCP resources can be taken over by a backup VTAM by using the BACKUP operand in the NCP definition. The operator at a backup VTAM can enter a VARY ACQ command to acquire resources that belong to another VTAM. The VARY ACQ command can be used to acquire all of the NCP resources, regardless of how the OWNER operand is specified, or can be used selectively to acquire only resources assigned to a particular VTAM through the OWNER operand. You can return the resources by using the VARY REL command.

Dump or IPL of a shared NCP

You can automatically dump or IPL the same NCP from more than one channel-attached VTAM. (You can code YES for AUTODMP and AUTOIPL on multiple VTAMs for the same NCP.) If the NCP is already being IPLed or dumped by one VTAM and it receives a request for an IPL or dump from another VTAM, the second request is rejected and the operator is informed that the second VTAM is waiting for the communications controller to become available. When the contending operation completes, VTAM restarts the requested IPL or dump operation if the state of the communications controller indicates that it is required.

Note: More than one host might take a dump successfully.

This function requires SSP Version 3 Release 7 or a subsequent release.