onoverlap
Defines the action that the scheduler must take on a
job stream instance that is about to start but the previous instance
has not yet completed. The options are to:
- Start the job stream instance anyway.
- Wait for the previous instance to complete.
- Cancel running the new instance altogether.
Syntax
onoverlap parallel|enqueue|donotstart
Arguments
- parallel
- The next instance is started regardless, and the two instances run concurrently. This is the
default behavior for the job stream if you do not use the
onoverlapkeyword. - enqueue
- The next instance is not started until the previous instance has completed its run.
Internally, this option is implemented as a follows dependency on the previous instance of the job stream. This internal dependency counts toward the maximum limit of 40 dependencies allowed for a job or job stream. Therefore, if you use enqueue, you can define a maximum of 39 additional manual dependencies (such as follows, needs, or opens) to avoid exceeding the plan limit.
- donotstart
- If, for any reason, the job stream cannot start within four minutes, the next instance does not start at all.
In the following example, an instance of job stream JS_0415 is run every 10 minutes. In case an
instance has not completed when the next one is to start, the next instance waits for its
completion.
Note: When using onoverlap enqueue, the system automatically generates a
dependency in the plan to track the status of the previous instance. If the total number of
manual dependencies plus the internal onoverlap dependency exceeds 40, the system
returns error AWSJOM138E and the plan might experience delays or synchronization
issues.
SCHEDULE MDM005#JS_0415
ON RUNCYCLE ACCTRC 06/16/2025
( SCHEDTIME 1445 DEADLINE 1530 EVERY 0010 EVERYENDTIME 1515 )
ONOVERLAP ENQUEUE
# Note: Because ENQUEUE is used, ensure the job stream
# below has no more than 39 additional FOLLOWS or NEEDS dependencies.
:
MDM005#JS_0415
END