z/OS Communications Server: SNA Network Implementation Guide
Previous topic | Next topic | Contents | Contact z/OS | Library | PDF


High availability using extended recovery facility

z/OS Communications Server: SNA Network Implementation Guide
SC27-3672-01

You can identify an application program as capable of operating in an extended recovery facility (XRF) environment. Two IBM® application programs, Information Management System (IMS™) and Customer Information Control System (CICS®), support an XRF environment.

XRF provides an alternate application program when certain host components fail. It reduces the impact of planned outages or outages caused by failures in VTAM®, MVS™, IMS, CICS, or the host processor on selected users of IMS, CICS, or other application programs that use XRF functions. The types of failures that cause the takeover of an XRF session are determined by how IMS, CICS, or the application program using XRF is defined. XRF might not reduce the impact of failures in the NCP or in telecommunication lines, depending on which NCP or line fails. However, the XRF function in CICS or IMS does not automatically perform a switch to the alternate application program under these conditions. Therefore, either the operator must switch from the active to the alternate system, or you can use NetView® automation to perform the switch.

To enable high availability, use the HAVAIL operand on the APPL definition statement.

VTAM, in conjunction with the application program, establishes a primary session from the currently active application program and a backup session from the alternate application program to the terminal. The NCP for a terminal in an XRF environment maintains two sessions through two VTAMs: one to a primary XRF application program and one backup session to the alternate XRF application program. The XRF application programs, in conjunction with VTAM and NCP, switch to the backup session when some component fails. The active and alternate application programs must be in the same network. However, a terminal participating in an XRF session can be in a different network than either of the application programs.

One application program acts as the active application program, and the other acts as the alternate in case the active application program or another component in the active host fails. Both application programs are identified by a single name that is associated with the currently active application program.

Although one VTAM in a single-processor environment can support both a primary and a backup session, this configuration does not provide the availability characteristics usually associated with an XRF environment.

XRF is also supported for application programs that use LU 6.2 sessions to communicate with other applications or logical units in peripheral nodes attached to an NCP. For further information, see LU 6.2 in an XRF Environment.

Go to the previous page Go to the next page




Copyright IBM Corporation 1990, 2014