Archives of the TeradataForum

Message Posted: Thu, 20 Feb 2003 @ 15:17:06 GMT

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

In my opinion, there are three problems with automatically running a RELEASE MLOAD. You touch on the first one in that it hangs you up.

2. You will lose the capability of restarting a job. I feel that a better solution is to check the status and check for the presence of the error and Duplicate (aka Uniqueness Violation) tables after the load and resolving any problems before your next load.

3. If the MLOAD dies in the APPLICATION phase. RELEASE MLOAD does not work. You then need to kill any USIs, perform a RELEAS MLOAD ... IN APPLY, recreate the USIs and recollect statistics.

This is a bad habit of our developers that I have been trying to break for three years now (with some success). You know how developers are :-).

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