Invalid Character warnig in datastage

rumu · Post by **rumu** » Mon Oct 27, 2008 5:34 am

Dear All,

I am getting the following warning in datastage:

ODBC_Enterprise_3,0: Invalid character(s) ([xFF]) found converting string (code point(s): StateProvinceID[x00][x00][x00][x00][x01][x00][x00][x00][x00][xFF][xFF][xFF][xFF][xFF][xFF]) from codepage ibm-1051_P100-1995 to Unicode, substituting.

My job design is as mentioned here:

Input :One Seq file stage and one ODBC stage .Through ODBC I am accessing sqlserver table.
Join stage:Joining(inner) the seq file and ODBC data based on one key column.

Output:ODBC stage ---another sql server table.

I am getting the above warning in the input odbc stage .
the field 'StateProvinceID' is of nvarchar in source table.

In my project the following NSL mapping is used:

Parallel job default NLS map ISO-8859-1, default locale OFF

Could you please let me why this warning is coming?

regards,
rumu

hemanthakumar · Post by **hemanthakumar** » Wed Oct 29, 2008 3:13 am

try with UTF-8 IN NLS setting in each stage.

rumu · Post by **rumu** » Mon Nov 03, 2008 5:24 am

Hi Hemanthkumar,

Thanks for your reply..
I already tried this option but issue remains same.

regards,
rumu

shershahkhan · Post by **shershahkhan** » Mon Nov 03, 2008 2:03 pm

I don't know how it the characters works in SQL Server but i was facing the same type of issue in Teradata, to solve the issue use the character set ibm-1051_P100-1995, use the one that is used in SQL server and that will solve the issue

toshea · Post by **toshea** » Thu Nov 06, 2008 12:14 pm

Perhaps the character set in your job NLS map does not match the character set in your ODBC data source. If you have your NLS map set to ibm-1051_P100-1995, make sure your data source in .odbc.ini is set to an equivalent character set.