15.11 - Considerations for Aborting a Job - Data Stream Architecture

Teradata Data Stream Architecture (DSA) User Guide

prodname
Data Stream Architecture
vrm_release
15.11
created_date
December 2016
category
User Guide
featnum
B035-3150-026K

You might want to abort a job after the job has been submitted to run. For example, you might have forgotten to include an object for backup that you did not specify in the job definition XML file. Or you might see, while a backup job is running, that data will fill the media if the job is completed.

You can use the abort_job command to abort an actively running job or a job in the queue. An abort command creates two states for the job. While the job state is aborting, devices are being released for other jobs to use. After the job has reached an aborted state, the job releases DSA stream resources and the DSC job slot.

If you abort an actively running job, DSA does not create a save set. Any backed files are rolled back to the state they were in following the last complete, successful run. Third-party backup management software does not keep records of jobs that are not run successfully.

If you try to abort a restore job of a DSC repository backup while the job is in progress, the DSC metadata is corrupted. DSC triggers a command to restore all repository tables to their initial state, which is an empty table. You would therefore lose your DSC repository backup data.