Archives of the TeradataForum

Message Posted: Tue, 13 Jul 2004 @ 12:31:16 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Chakravarthy:

Normally in the script, every target table would have its own work table defined, as well as two error tables. In other words, you need 4 worktables and 8 error tables to load 4 target tables. Since your script does not define all the ancillary tables, they are created automatically in the same database as the target table with default names starting WT_ for worktables and ET_ and UV_ for the error tables. You should see these tables being created in the output from Mload. However, it does not appear that this is your problem based on the error message. If you had a syntax error, the Mload script would not have run in the first place.

This message indicates that a restart is attempting to be performed. It appears that Mload aborted for some reason. Why did it abort? What is the sequence of events before receiving this error message?

My guess is that someone deleted one or more of the error tables and did not delete the log table for the Mload run. If this is the case, than the tables are still locked for loading and it is an inconsistent state for a restart to be performed.

If this is a test environment, then you could issue the RELEASE MLOAD command for your tables and no harm is done if the abort occurred in the acquisition phase. There has been a recent discussion thread on the use of RELEASE MLOAD and I would recommend that you look at it in the archives.

Michael Larkins
Certified Teradata Master

	<Prev	Next>		<<First	<Prev	Next>	Last>>