Spooler Utilities Reference Manual

Introduction to the Spooler Subsystem
Spooler Utilities Reference Manual—522295-003
2-19
The Spooler and Batch Jobs
When a batch job is printed, each member job is delivered to the device in the
sequence in which it is linked in the batch. Optional batch headers can be applied to
identify the beginning and end of the batch output.
A batch job can be created in the spooler only when a job is opened that originates
from a process that has a JOBID attribute. The JOBID attribute may be established
using a DEFINE or a RUN parameter. When such a job is opened, a new batch job
may be created or the job may be linked to an existing open batch job. To be linked to
an existing open batch job, all of the new job’s key attributes must match those of the
open job in the batch job.
The key attributes that determine whether a spooler job is part of a batch job are
described in Table 2-2.
Other job attributes have no affect on whether the job is linked to a batch job. Some
attributes might have no meaning for a job that is part of a batch job. For example, if
every job within a batch job contains a different report name, only the report name
associated with the first job in the batch job is printed on the header pages for that
batch job. Attributes that have meaning only for the first job in a batch job are the
report name, the batch name, and the selection priority.
The method that the spooler uses to determine whether a job becomes a normal job,
the first job in a batch job, or a job to be linked to an existing batch job is shown in
Figure 2-6
.
Table 2-2. Key Attributes
Key Attribute Description
JOBID This identifier is the same for all jobs in a batch job. A job that is
not part of a batch job does not have this identifier. JOBID can be
set through a TACL RUN command or a DEFINE for the
application generating the batch job.
FORM Each job in a batch job must use the same print device and
paper.
OWNER Each job in a batch job must be owned by the same user.
Device where the job is
printed
The device where the job is printed must be the same for all
members of a batch job. This does not mean that location is a key
attribute; jobs can have different locations and still be members of
the same batch as long as their locations map to the same
device*
GMOM identity of the
process that produced
the job
This ensures that batches created from different batch executives
will be different, even though all of the other key attributes match.
*Locations that do not have a single device mapping (for example, groups or dummy locations) are considered to
be mapped to a
null device. Because locations can be mapped to a null device, jobs whose key attributes and
locations map to the null device can be in the same batch job even though their actual locations may vary. Batch
jobs that specify the null device are delivered to the location that is established for the first member of the batch
job.