Archives of the TeradataForum
Message Posted: Fri, 02 Mar 2001 @ 19:58:09 GMT
We had some discussion with NCR about this issue (through the PAC), and there are some fairly hairy issues to be dealt with. In short, any table modified in the transaction has to be marked as corrupt and a means provided for dropping it prior to reload. Now think about multi-statement transactions with triggers going in all directions...
/dave 'doc' hough
|Copyright 2016 - All Rights Reserved|
|Last Modified: 28 Jun 2020|