The NLS Characterset in our Oracle DB is set as 'AL32UTF8'. When I run my jobs to extract the data i get a warning like below
'Implicit conversion from source type "ustring[3]" to result type "string[3]": Converting ustring to string using codepage ISO-8859-1.'
My taraget data has some weird unrecognisable characters in it. This happens only with the cloumns of string datatype. I get this error when i read the data from source and these are passed along to the target. I am using ODBC to read the source as well as to load into target. Can anyone please help me?
Kumar's suggestion will work for you. It tells Oracle to use that character set as output and it does the NLS translation to UTF8 which is what DataStage would like.
Since you don't have NLS installed in DataStage you cannot use any NLS settings inside the product.