MVS failure and sysplex failure management

The TCP/IPs in a sysplex use MVS™ XCF messaging to exchange information about dynamic VIPAs. When a TCP/IP fails or is ended by operator command, but the underlying MVS remains active, the other TCP/IPs are immediately notified, and takeover of VIPADEFINE dynamic VIPAs is automated and very fast. In addition, each TCP/IP stack monitors for local problem conditions that might cause it to leave the TCP/IP sysplex group, at which point the other TCP/IPs will be immediately notified. For more information, see Sysplex problem detection and recovery.

However, when an MVS fails, there is normally an operator message on the console requiring a response (WTOR). Until this response is made by an operator or automation, the other MVS systems do not notify the remaining TCP/IPs in the sysplex of the failure of the TCP/IP on the failing MVS. This can delay automated backup of dynamic VIPAs defined with VIPADEFINE. Sysplex Failure Management (SFM) can be used to automate the required response to the console message of the failing MVS. See z/OS MVS Setting Up a Sysplex for information on how to set up SFM to avoid the requirement for a manual response and speed backup of dynamic VIPAs defined with VIPADEFINE.

For more information, see z/OS Communications Server: IP Diagnosis Guide.