Originally posted by: paul.brett
If a map has the following example DBLOOKUP() function:
=DBLOOKUP ("select * from Test1", "Test.mdq", "Test_DB2")
...this syntax means that the MDQ file is read at build time, and the database connection values for native DB2 should be read from the MDQ, and stored in the compiled map.
The MDQ may contain:
<DATABASE name="Test_DB2">
<TYPE>DB2</TYPE>
<PLATFORM>MVS</PLATFORM>
<QUALIFIEDNAMES>false</QUALIFIEDNAMES>
<SURROUNDCHARS></SURROUNDCHARS>
<PLAN>DBUTIL</PLAN>
<SUBSYSTEM>DH07</SUBSYSTEM>
<DESIGNSOURCE>MainframeDB2V9</DESIGNSOURCE>
<USER>FRED</USER>
<PASSWORD.encrypted>$@k``[^ZZ</PASSWORD.encrypted>
</DATABASE>
A quick review of the compiled map code shows that this is NOT happening, and it is most likely the map will fail to connect to DB2 (depending on the inclusion of SDSNEXIT dataset in the JCL and machine specific configuration). Compiled map contains:
"DBInfoBlock DB2-MVS MainframeDB2V9"
...at the end, which is the connection info for an ODBC connection (from the MDQ file for Windows for the purposes of importing Typetrees).
IBM WebSphere TX Software Developers are reviewing the compiler code, to see why the wrong database connection data is being compiled into the map.
Thank-you.
Paul
Follow me on Twitter
#IBM-Websphere-Transformation-Extender#IBMSterlingTransformationExtender#DataExchange