This blog post describes how you can convert your client database running with a single-byte character set database, WE8ISO8859P1, to a Unicode Character Database (UCD), AL32UTF8, by using the Oracle Data Migration Assistant for Unicode (DMU) tool.
To migrate Oracle® Database 18.104.22.168 and R12.1.3 E-Business Suite® (EBS) application file system character set databases from WE8ISO8859P1 to AL32UTF8, complete the following steps:
Perform the following tasks before starting the migration:
Take a complete backup of the DB and applications. Count the invalid details and disable DB force logging.
Purge RECYCLEBIN and use the
GATHER AUTO option to gather statistics on e-Business schemas. Check the
optimizer_features_enable setting and correct it as needed.
Apply patch 28956851:R12.TXK.B (R12.1.3). Refer to Oracle Doc ID 393861.1.
Run AutoConfig on the Application node
Create the /admin/out/appsutil.zip file on the application node by running the following command:
Copy appsutil.zip to RDBMS ORACLE_HOME and run the following command:
$unzip -o appsutil.zip
Run adautocfg.sh on the DB node.
Run adautocfg.sh on the application node.
Perform the following steps to prepare to use the DMU:
Execute the following commands as a SYS user:
SQL> spool prvtdumi.log SQL> @?/rdbms/admin/prvtdumi.plb
Run the following command to create a password file:
NONPROD [oradi@nordlrbbnebsd01 dbs]$ orapwd file=$ORACLE_HOME/dbs/orapw$ORACLE_SID password=*****
Download and install JDK if required.
To launch DMU on a VNC session and create a database connection to the DMU tool, run the following command:
$ sh dmu.sh
Save the connection.
Install the DMU repositorycheckbox.
Click Next and then Finish.
The preceding message displays after the repository installs successfully.
Now scan the data dictionary and the application schemas:
The following screenshot shows the scan in progress:
The following screenshot confirms the complete database scan:
After scanning, you need to look at the data using the Database Scan Report as illustrated in the following image:
Notice that your actual data shown in the DMU is in the following columns:
To perform bulk cleansing, perform the following steps:
Right-click on the database connection and select the Bulk Cleansing option from the pop-up menu.
Run the Migrate to character length semantics option and check the results of the bulk cleansing. You also have the option to select a different parameter to clean the dirty data.
You can also refer to some tips for the issues and fixes related to DMU Doc ID 2018250.1.
You need to clear all the warnings before proceeding to the conversion steps. Sometimes you need to open a service request (SR) to confirm some issue or data purge during cleansing.
After fixing all the issues, right-click on the Database connection, select the Convert Database option, and click Convert.
Accept all the defaults by clicking the Next. The Migration Status tab also shows the conversion status when it completes.
Migration in progress:
Migration is complete as indicated by the Status panel:
This section includes steps 3.1.4 through 3.3 from Doc ID 393861.1 Update Database Tier AutoConfig Context Files, Appendix A:
On the DB node, execute the following command and verify NLS_LANG is in the DB env file:
$ perl txkSetUnicodeCharSet.pl
On the Apps tier, execute the following commands:
$ ls $ORACLE_CONFIG_HOME/reports/Tk2Motif_AL32UTF8.rgb /u01/app/appdi/DEV/inst/apps/MKS_mksdbcs/ora/10.1.2/reports/Tk2Motif_AL32UTF8.rgb $ ls $ORACLE_HOME/guicommon/tk/admin/Tk2Motif_UTF8.rgb /u01/app/appdi/DEV/apps/tech_st/10.1.2/guicommon/tk/admin/Tk2Motif_UTF8.rgb $ cd $ORACLE_CONFIG_HOME/reports/ $ cp Tk2Motif_AL32UTF8.rgb $ORACLE_CONFIG_HOME/reports/Tk2Motif_AL32UTF8_backup.rgb $ grep -i fontMapCs Tk2Motif_AL32UTF8.rgb !Tk2Motif*fontMapCs: iso8859-2=EE8ISO8859P2 Tk2Motif*fontMapCs: iso8859-1=UTF8 $ vi Tk2Motif_AL32UTF8.rgb $ grep -i fontMapCs Tk2Motif_AL32UTF8.rgb !Tk2Motif*fontMapCs: iso8859-2=EE8ISO8859P2 Tk2Motif*fontMapCs: iso8859-1=AL32UTF8
Execute the following command and verify NLS_LANG is in the Application env file:
$ perl txkSetUnicodeCharSet.pl
Regenerate Forms and Reports through adadmin.
Regenerate Messages through adadmin.
Update the ICX:Client IANA profile option to
Start services and validate the application. It should work fine for you.
The Oracle DMU tool provides a streamlined solution for migrating a legacy single-byte character set database to a UCD. The DMU tool helps ensure all character data migrates correctly quicker and easier. It also ensures that the migration completes without any data loss and validates the data integrity of the databases.
If the Oracle EBS application has more than one language group, UCD is the only choice. Oracle recommends using Unicode as the database character set for maximum compatibility and extensibility. The UCD character set supports all frequently used languages. It also supports symbols like the Euro sign, smart quotes, and scientific and musical notations. The IT industry is using UCDs more and more every day.
Use the Feedback tab to make any comments or ask questions. You can also start a conversation with us.