Status recovery mode for end-user significant status

The status recovery mode defines the scope and location of recovery for resources with End-User Significant Status.

You can set the default mode for each IMS, which is used for all resources unless overridden during a logon or signon. You can then set what End-User Significant Status to recover.

The following list describes how to set the status recovery modes for TM resources:
  • Specify IMS defaults in DFSDCxxx.
  • Modify user descriptors.
  • Set the DFSLGNX0 override for static terminal logon and dynamic STSN terminal logon.
  • Set DFSSGNX0 override for dynamic non-STSN user signon.

The following list describes the three recovery modes for End-User Significant Status.

GLOBAL Status Recovery Mode

All recoverable status is saved locally in the IMS record logs, but uses RM to recover the status instead of the logs. Status is restored at the next logon or signon and is available to any IMS in the IMSplex. When the resource becomes active, status is copied to the local system. When the resource becomes inactive, status is deleted from the local system.

RM, a coupling facility resource structure, and shared queues are required. If you are sharing queues and have a resource structure, GLOBAL is the default. The default is overridden by DFSDCxxx, user descriptors, or logon and sign-on exit routines.

LOCAL Status Recovery Mode

All recoverable status is saved locally in the local control blocks and log records. Status is restored at the next logon or signon and is only available to the same IMS that the user or node was accessing. Additionally, the user or node can only access the IMS where the local status is. This affinity is called RM affinity, which is enforced when End-User Significant Status exists. With RM affinity, IMS does not allow a terminal or user to log on to or sign on to an IMS if RM indicates that the user or terminal has RM affinity to another IMS. This affinity occurs because end-user significant status (conversation, STSN, or Fast Path) is being recovered on another IMS.

When the IMS with the RM affinity fails, the RM affinity still exists. The user or node can access another IMS immediately if the Logon exit routine (DFSLGNX0) or the Sign-on exit routine (DFSSGNX0) allows it, but resource status is not recovered and local status is deleted on the failed IMS at restart.

LOCAL is the default if a resource structure or shared queues are not used.

RM is not required. If RM is not active, RM affinity is not enforced.

NONE Status Recovery Mode

No status is saved in RM or local log records. At logon or signon, significant status is cold. STSN, conversation, and fast path status is automatically nonrecoverable.

RM and a resource structure are not required.