Restarting a task after an error in a system task of VM2000

After an error in the VM2000 monitor, the system task involved (VM2M, VM2G, VM2C, VM2J, VM2P, VM2E, VM2T) is continued at a consistency point.

A task restart after an error in the VM2000 monitor may have the following consequences:

• Error in the main task (VM2M)
  Inconsistent table entries may exist. A VM2000 command may not have been executed completely. Information on the current VM state is supplied by the VM2000 information commands.

• Error in the general task (VM2G, SU /390)
  Requests to the VM2000 monitor may not have been processed completely.

• Error in the $VMCONS task (VM2C)
  All connections to $VMCONS (ADMIN dialog, VC dialog) are cleared. Once $VMCONS has restarted automatically, these connections must be established again.

• Error in the job variable task (VM2J)
  The current update of a MONJV for monitoring a VM, main memory reconfiguration or VM migration is not executed. When the monitoring is finished, e.g. with /DELETE-VM, problems may occur in resetting the MONJV. Such a MONJV can only be used again when the MONJV protection has been reset manually with /MODIFY-JV-ATTRIBUTES JV-NAME=...,PROTECTION=(MONJV-PROTECTION=*NO) or in the next system run of the monitor system.
  The updating of further MONJVs is delayed until the VM2J task is restarted automatically.

• Error in the pubset task (VM2P)
  The last action for a pubset is rejected with the message VMS4252 (“timeout”).

• Error in the task for the VM2000 Agent (VM2E, SU x86)
  Requests of the VM2000 Agent may have not been executed.

• Error in the task for the VM Migration on the target SU (VM2T, SU /390) or in the dynamically created task for the VM Migration on the source SU (SU /390)
  The VM Migration is aborted.