When Teradata MultiLoad encounters an error in a job script, it generates a diagnostic error message and stops with a nonzero return code. At this point, modify the script to correct the error and resubmit the Teradata MultiLoad job. The utility resumes processing at the statement following the last one that completed successfully.
When correcting script errors, make changes at or after the indicated error. Teradata MultiLoad does not repeat the commands that executed successfully, but the job will fail, with additional error messages, if the utility detects changes before such errors.
Any SQL statements used outside the BEGIN MLOAD and END MLOAD commands are not part of a Teradata MultiLoad job. If one of these statements fails and generates an error, it is not treated as a terminating error or a user error. Instead, Teradata MultiLoad continues processing after setting the system variable SYSRC to the last error code. If the Teradata MultiLoad job is processed successfully, the restart log table is dropped at the end of the job.