Job actions
The execution of a job can be managed by selecting one of the available actions:
Abort: stops the job after the current operation finishes.
Resume: restarts a Suspended or Error job from the exact point where it stopped.
Resubmit: starts the workflow from the beginning, with a new job ID.
Abort and Resubmit: stops the current job and starts a new one from the beginning.
Change queue: moves the job to a different execution queue.
Abort
Stops a job after the current operation completes. The interruption is not immediate. When Abort is selected, the currently running service task finishes and only then does the workflow stop. This is especially important for triggerable service tasks (i.e., transfers delegated to STENG). For example:
If a file push/pull is in progress, the transfer completes before the job is stopped.
If a job contains multiple steps (e.g., 10 operations) and Abort is triggered during step 4, step 4 completes and then the workflow stops.
Resume
Restarts a job in Suspended or Error status using the same Job ID. Execution resumes exactly where the failure or pause occurred. Use this action when the cause of the issue has been resolved and the workflow can continue safely without reprocessing earlier steps.
Resubmit
Creates a new job with a new Job ID and restarts execution from the beginning. Use Resubmit when:
A workflow has completed and must be re-executed.
A job is in Error or Aborted status and must be run from scratch.
Abort and Resubmit
Stops a running or failed job and restarts the workflow from the beginning with a new Job ID. The original job remains in its final state (Aborted).
Use this option when the workflow must be re-executed entirely and partial results cannot be reused.
Change queue
Assigns the job to a different execution queue. This action affects scheduling and prioritization but does not modify the Job ID or current status.
Use Change queue when a job needs to be executed with a different queue configuration or priority, including jobs in Error status.
Last updated