Previous topic |
Next topic |
Contents |
Contact z/OS |
Library |
PDF
Making jobs eligible to restart z/OS JES2 Commands SA32-0990-00 |
|
When a checkpoint data set resides on a coupling facility structure instead of on a DASD volume, neither the software checkpoint lock nor the hardware lock are set. Therefore, neither lock has to be released. However, the jobs from the failed processor must be requeued for execution on an active member of the MAS. When a JES2 member of a MAS configuration fails while
holding the software checkpoint lock, other members within the configuration
cannot access the checkpoint data set. JES2 issues the message:
To make jobs from a failed JES2 member eligible for restart on
another member when both the MVS™ system and the JES2 subsystem
fail, the hardware lock (if held) must be reset and the software checkpoint
lock must be reset. If the JES2 subsystem fails and the MVS system
continues to run jobs, reset the software lock and restart the JES2
subsystem. After resetting both the hardware and the software checkpoint
lock, jobs become eligible for restart:
Using operator commands: To make jobs eligible for restart
without the AUTOEMEM option (when the JES2 checkpoint data set resides
on DASD) an operator must:
Note: When a checkpoint resides on a coupling facility structure,
you must still enter the $E MEMBER(member) to requeue jobs.
Using the AUTOEMEM Option: In a sysplex, JES2 members of
a MAS can take advantage of the AUTOEMEM option to make jobs eligible
for restart, with minimal operator intervention. The option can be
specified on the MASDEF initialization statement (or dynamically through
the $T MASDEF command):
|
Copyright IBM Corporation 1990, 2014
|