Reinitializing After Terminating a Teradata MultiLoad Job
When a Teradata MultiLoad job terminates either before or after the application phase, and job is abandoned, use the following procedure to restore access to the target tables.
1 Log on to BTEQ with a user ID that has drop table privileges on Teradata Database associated with the aborted Teradata MultiLoad job.
2 Use a RELEASE MLOAD statement specifying all of the target tables identified in the TABLES clause of the BEGIN MLOAD command for the aborted Teradata MultiLoad job.
Note: An unsuccessful RELEASE MLOAD statement indicates that the Teradata MultiLoad job was in the application phase when it terminated. In this case, the Teradata MultiLoad job cannot be abandoned. Restart the job and let it run to completion. Use the After a Job Abort or Client System Failure procedure in “Restarting a Paused Teradata MultiLoad Job” on page 43.
Further, if the tables had permanent journals, to recover the table, use an insert or select operation (with locking for access) to copy data from the target table into a temporary table. Then drop the target table and rename the temporary table with the original name of the target table.
3 After successfully executing the RELEASE MLOAD statement, use Teradata SQL DROP TABLE statements to drop the work tables and error tables identified in the WORKTABLES and ERRORTABLES options of the BEGIN MLOAD command for the aborted Teradata MultiLoad job.
Note: If the tables were not explicitly defined, see the SYSPRINT output following the UTY0812/UTY0815/UTY0817 message sequence.
4 Use a DROP TABLE statement to drop the restart log table identified in the LOGTABLE command for the aborted Teradata MultiLoad job.