MVS and JES2 failure

If the failed member holds the JES2 checkpoint data set, the operator must:

If the checkpoint resides on DASD, reset the hardware lock if held (as indicated by message $HASP263).
If the checkpoint resides on DASD, reset the software checkpoint lock by using the $E CKPTLOCK,HELDBY= command.
Requeue the jobs for restart by entering the $E MEMBER command, which makes jobs eligible for restart on another JES2 member. Note that jobs with member affinity to the failed system will not be restarted until the failed member is available or member affinity for those jobs is modified.

Figure 1 shows a sequence of events in a 2-member MAS configuration where the software checkpoint lock is held by the failed JES2 member (SYSA). The operator must respond to the $HASP264 message by entering the $E CKPTLOCK,HELDBY= command, even though the MVS™ system (SYSTEMA) continues to process jobs.

Figure 1. Two-member MAS with JES2 failure (checkpoint on DASD)

On SYSTEMA:                        On SYSTEMB:

JES2 Fails
                                      $HASP264 WAITING FOR ACCESS TO
                                           JES2 CHECKPOINT VOLUME
vvvvvv

                                      $E CKPTLOCK,HELBY=SYSA
Restart JES2

In a 2-member MAS configuration where the checkpoint resides on a Coupling facility structure and the Coupling facility lock that JES2 uses to control access to the checkpoint was held by the failed JES2 member (SYSA), the operator does not have to release it; the Coupling facility lock is released without operator intervention while the MVS system (SYSTEMA) continues to process jobs.

Note that the $HASP263 message contains a modified text that is informational.

Figure 2. Two-member MAS with JES2 failure (checkpoint on Coupling facilities)

On SYSTEMA:                          On SYSTEMB:

JES2 Fails
                                       $HASP263 WAITING FOR ACCESS
                                        TO JES2 CHECKPOINT.  LOCK
                                        HELD BY MEMBER member_name

Restart JES2

Figure 3 shows a sequence of events in a 2-member MAS configuration where the checkpoint data set resides on DASD and both the hardware lock and the software checkpoint lock are held by the failed JES2 member (SYSA). In this instance, both the JES2 member and the MVS system (SYSTEMA) have failed.

Figure 3. Two-member MAS with JES2 and MVS failure (DASD)

On SYSTEMA:           On SYSTEMB:

  SYSTEMA Fails
  (For example,
  the system failed
  and entered
  a nonrestartable
  wait state.)
                                      $HASP263 WAITING FOR ACCESS
                                       TO JES2 CHECKPOINT VOLUME...


SYSTEM RESET
 (Operator)

                                     $HASP264 WAITING FOR RELEASE
                                            OF CKPT LOCK BY SYSA

                                      $E CKPTLOCK,HELDBY= SYSA

                                      $E MEMBER(SYSA)

                                     $HASP493 SYSB MEMBER SYSA
                                        RESTART IS IN PROGRESS

                                     $HASP406 PRODJOB RESTARTED

                                     $HASP492 SYSB MEMBER-SYSA
                                         RESTART HAS COMPLETED

Re-IPL SYSTEMA