DB2 Loader and Null indicator
Posted: Thu Apr 23, 2009 8:59 am
Howdy all,
We have the following problem with Datastage and DB2.
Using DB2/Enterprise targets, data needs to be loaded with the DB2 loader with ASCII delimited format = false (aka fixed length). The problem is that, for null values of timestamp columns, the Datastage outputs the data as a string of 26 asterix characters (25 is the timestamp mask length and 1 is the null indicator character).
However, the DB2 does not seem to use the same character for nulls and, when attempting to load data it would treat it as any other character, thus failing the timestamp mask validation.
We figure we need to either change the default DB2 loader null indicator or change the null indicator from DataStage.
The official Datastage guides do specify a place where to configure the null characters: the mainframe tab. However, that is enabled only for mainframe jobs, and we do not use them.
Any thoughts?
$ L3191N The field in row "1", column "407" which begins with
"**************************" does not match the user specified DATEFORMAT,
L3191N The field in row "1", column "407" which begins with
"**************************" does not match the user specified DATEFORMAT,
TIMEFORMAT, or TIMESTAMPFORMAT. The row will be rejected.
Kind regards,
Adi.
We have the following problem with Datastage and DB2.
Using DB2/Enterprise targets, data needs to be loaded with the DB2 loader with ASCII delimited format = false (aka fixed length). The problem is that, for null values of timestamp columns, the Datastage outputs the data as a string of 26 asterix characters (25 is the timestamp mask length and 1 is the null indicator character).
However, the DB2 does not seem to use the same character for nulls and, when attempting to load data it would treat it as any other character, thus failing the timestamp mask validation.
We figure we need to either change the default DB2 loader null indicator or change the null indicator from DataStage.
The official Datastage guides do specify a place where to configure the null characters: the mainframe tab. However, that is enabled only for mainframe jobs, and we do not use them.
Any thoughts?
$ L3191N The field in row "1", column "407" which begins with
"**************************" does not match the user specified DATEFORMAT,
L3191N The field in row "1", column "407" which begins with
"**************************" does not match the user specified DATEFORMAT,
TIMEFORMAT, or TIMESTAMPFORMAT. The row will be rejected.
Kind regards,
Adi.