We are running a process connecting a DB2 EE V7.1 to a DB2 for z/OS V7.

Tables in the host (DB2 for z/OS) are updated from the client (DB2 EE Windows) and some fields containg arabic characters.

The DB2CODEPAGE in Windows is 1256 and the CCSID in the z/OS is 420.

The connection is thru TCP/IP and DRDA.

This process is running without problems, tables are being updated and selected from both environments successfully and displaying the arabic characters.

Now we migrated the DB2 EE in the Windows platform to version 8 and the following problem surfaced:

If a table is updated in the mainframe the arabic characters display OK in both environments.

If the same table is updated in the Windows box, the characters in the Windows are displayed OK, but in the mainframe they appear "unlinked".

This problem affects all the tables.

DB2CODEPAGE and CCSID are set as before. The problem is the same with Windows 2000 server or Windows XP Professional.

DB2 in the mainframe is V7 at PUT0405 running under z/OS V1.4

DB2 in the Windows box is:

DB21085I Instance "DB2" uses "32" bits and DB2 code release "SQL08020" with level identifier "03010106".
Informational tokens are "DB2 v8.1.7.664", "s040914", "WR21342", and FixPak

Any help on this subject will be appreciated


PS.: Of course... I am in charge of solving the problem and I do not know the arabic language... Business as usual...