112 Data Sharing in a Nutshell
6.3 Sysplex failure management (SFM) policy
If you have not implemented the sysplex failure management (SMF) policy, you
will realize its benefits when a z/OS image on which a DB2 member in the group
runs appears to hang. A member of the sysplex that has not updated its status
within the specified failure detection interval needs to be partitioned out promptly.
It is rarely possible to save a failed system through operator action after XCF has
detected failure of the system. Automatic removal of the failed system from the
sysplex is preferable to avoid “sympathy sickness” from the lack of access to
resources locked by the failing system. The objective is to minimize timeouts
from surviving DB2 members. When your ...