That's true, but you might have a "MARK" option at use in your mload script, so in that case ET/UV tables are nice to have for later analyze.
cheers
-vh-
Hi Sankalp,
Nice question. To all intents and purposes, a software or automation or technology is made to serve business. A software or automation that takes care of all business cases or eventualities is a very good software. An architect or designer has understood and looked at all the events or cases from all corners, leaving no chances of faults or discrepancies(take care of every business case).
So what happens, if I have to load from sources having less, medium , high corrupted source files(duplicate) say having billions of rows? We can think of several scenarios.
Mload has error,log,uv , work tables for reasons. Hats off to the genius who made it.
Do you agree with me ? :)
Cheers,
Raja
Raja K Thaw
My wiki: http://en.wikipedia.org/wiki/User:Kt_raj1
Street Children suffer not by their fault. We can help them if we want.
no doubt.
Hi Sankalp,
of course the UV table is needed and any errors in this table will not pause the job (ERRLIMIT is only for the ET table).
This is a cut&paste from the MLoad manual:
The second error table, called the application error table, is the one specified as tname4
(default UV_tname1) in the BEGIN MLOAD command. It provides information about:
• Uniqueness violations
• Field overflow on columns other than primary index fields
• Constraint errors
Dieter
As I am under a impression that Mload does support duplicate entries, correct me if I am wrong. Then why we need a UV table? population on UV will pause the job.