DSXchange

I am not sure whether it is a bug or I am doing something funny but this is what I am facing:

I want to fetch a rows from DB2 table to flat file. There is a field in this table with datatype BigInt and length as 19.

I created a parallel job whose structure is


ODBC Stage --------------> Seq File

I got junk value for this particular field. Everyting else fine
I am using User defined SQL

I tried the same functionality by replacing ODBC with DB2 stage, I got valid values

All the things - query, metadata table name etc remain same in both jobs.
In both jobs I am using metadta which is imported from ODBC table definitions.

The issue is I must use ODBC. Reason is that I am developing job in test environment and the rows to be fetched is a table in production Database. I use Data source facility in ODBC to connect from test to prod. This cannot be done in DB2. When I said that I tried the job with DB2 stage, that was in production.

Please explain

DB2 Connect client software should allow you to connect to any DB2 data source anywhere through the DB2 Enterprise stage.

Is this software inbuilt or I have to use third party?
Does your reply imply that there indeed is a problem with ODBC and BigInt and further that I cant use it?

It comes with DB2 client software. It's not part of DataStage.

So what about ODBC? Is this a known issue? I mean is ODBC incompatible with BigInt in parallel jobs?

There are many versions of the ODBC standard. It is a Microsoft standard. The ODBC Enterprise stage documentation (in the Parallel Job Developer's Guide) suggests that it is compliant with version 3.5 of the ODBC standard - read the manual for more detailed information about the grammar and API.

Sorry Ray...do not have access to premium content...
But I have guessed what you have wrote.
Some versions of ODBC are incompatible..So that makes sense