Accents not displayed correctly after migrating Oracle DB to UTF8 on v21

Products

• CA Automic Workload Automation - Automation Engine

Issue/Introduction

In order to prepare for the upgrade to v24, the database Oracle was migrated to UTF8 using Oracle tool, while keeping the Automation Engine on version 21.

There was a change performed in the ucsrv.ini where the codeset was set to AL32UTF8 as shown below:

sqlDriverConnect=ODBCVAR=NNJNIORO...;SP=NLS_LANGUAGE=AMERICAN,NLS_TERRITORY=AMERICA,CODESET=AL32UTF8

After starting the Automation Engine on version 21.x and displaying the folder titles, the accents are no longer displayed correctly as seen below:

Environment

Automation Engine 21.x with Oracle Database migrated to UTF-8

Cause

Incorrect CODESET configuration in ucsrv.ini

Resolution

For version 21.x binaries, it is mandatory to keep the same non UTF8 codeset as before the database was migrated to UTF8 since UTF8 codeset is only supported on version 24.x and superior.

Hence, please revert the change in the codeset parameter in ucsrv.ini (and any other v21 utilities ini files) to before the UTF8 migration and restart the Automation Engine and that will solve the issue.

 

Additional Information

Please note that for Oracle, Automic v21.x supports 3 codesets in ucsrv.ini: 

WE8ISO8859P1, WE8ISO8859P15 or WE8MSWIN1252

When the codeset is updated to WE8MSWIN1252, like here:

sqlDriverConnect=ODBCVAR=NNJNIORO...;SP=NLS_LANGUAGE=AMERICAN,NLS_TERRITORY=AMERICA,CODESET=WE8MSWIN1252,NLS_LENGTH_SEMANTICS='CHAR'

The problem with the accent characters no longer occurred.