Hello List, another Oracle question. I am now in process of converting the charset in our Oracle database from WE8ISO8859P15 to AL32UTF8. The procedure is mostly similar to the following description -
Character Set Scanner utility though the change from US7ACII to UTF8 could be done with only using the 'alter database command' we will use the character set scanner utility CSSCAN to scan for any change required. Installing the CSSCAN schema. Before using the cssan you must install the schema as a user with DBA privileges. This is done by executing the csminst.sql script located at $ORACLE_HOME/rdbms/admin cd $ORACLE_HOME/rdbms/admin $sqlplus "/ as sysdba" SQL*Plus: Release 9.2.0.6.0 - Production on Tue Oct 17 16:00:16 2006 Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved. Connected to: Oracle9i Enterprise Edition Release 9.2.0.6.0 - Production With the Partitioning, OLAP and Oracle Data Mining options JServer Release 9.2.0.6.0 - Production SQL> START csminst.sql Executing CSSCAN The CSSCAN executable is located at the $ORACLE_HOME/bin. After executing the script you must connect to the Database as a user with DBA privileges. The CSSCAN utility prompts you with options to scan a table, schema or the enter database. We choose to scan the entire database in our case. The CSSCAN displays your current character set version and prompts you to enter the new character set version. $ csscan Character Set Scanner v1.1 : Release 9.2.0.1.0 - Production on Tue Oct 17 16:03:15 2006 Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved. Username: system Password: Connected to: Oracle9i Enterprise Edition Release 9.2.0.6.0 - Production With the Partitioning, OLAP and Oracle Data Mining options JServer Release 9.2.0.6.0 - Production (1) Full database, (2) User, (3) Table: 1 > 1 Enter new database character set name: > UTF8 Enter number of scan processes to utilize (1..32): 1 > After the successful completion of the scan three report files will be created scan.txt, scan.out and scan.err. In my case I have exported the database objects with exceptions as specified in the scan.out and plan to import it back after the database conversion is done. Changing the Character Set at Database Level Shutdown all application services and the database and listener. You must do a clean shutdown of your database and start it up in restricted mode before changing the character set. Also you must not have any active sessions before issuing the alter database command. SQL> alter database character set internal_use UTF8; alter database character set internal_use UTF8 Import Back Database Objects. After a successful character set conversion at the database level now you can import back the objects and rebuild the indexes specified the in scan.out report. I've searched for all the objects that are not convertible using this csscan utility, modified these objects (mostly it was "exceed column size" error and I just enlarged column sizes). Then I switched the charset (alter database character set internal_use UTF8). The database seemed to work fine. So I started the ARS service. But the service did not start. There are two possible explanations - 1. The procedure described above is not right. It is not possible to upgrade the database to Unicode this way. If so, what is a right procedure? 2. I need to change something in ARS configuration files also. ARS can not connect to a database with another char set so easily. If so, what should I correct in ARS? Thank you for your assistance. I tried to find the procedure of switching/upgrading to Unicode within ARS manuals with no success. Environment: Oracle 9.2 + ARS 6.3 on a Win2003 platform. This is our test/development server. I will have to switch our productive one (ARS 6.3 + Oracle 9.2 + Solaris 9) later on also. Kind regards Dmitry Bezhenar
