CSSCAN and CSALTER To Be Desupported after DB 11.2. (文档 ID 1418321.1)

APPLIES TO:

Oracle Database - Enterprise Edition - Version 11.2.0.1 and later
Information in this document applies to any platform.
***Checked for relevance on 21-Jul-2017***

DETAILS

For releases following Oracle Database 11.2, support for CSSCAN and CSALTER will be discontinued and they will no longer be shipped with the database, since their functionality in respect to migration to Unicode -- the ultimately recommended database character set -- will be provided by the Database Migration Assistant for Unicode (DMU). Through the life of Database 11.2 including all patch sets, we will still provide error correction support for CSSCAN and CSALTER though we will not consider any further feature enhancements to these utilities.

There are scenarios where customers use the pass-through configuration, sometimes inadvertently, to store data in a character set different from the declared database character set. These customers often want to correct the database character set declaration to agree with the database contents but they do not have immediate resources to attempt Unicode migration and adapt their applications to run against a Unicode database. The correction of the database character set declaration, which involves only metadata changes, is currently performed by running CSSCAN with the source and target character set parameters set to the correct database character set and then running CSALTER. In future releases, Oracle will provide a script called CSREPAIR to correct the database character set metadata. The CSREPAIR script will access the DMU repository instead of the CSSCAN repository. It will change the database character set declaration only after the DMU has performed a full database scan with the Assumed Database Character Set property set to the correct database character set and only if no invalid character representation issues have been reported.

There will be no replacement for the CSSCAN functionality to look for data expansion issues before migrating to a non-Unicode character set. This functionality is applicable only if the target character set is multibyte.

See The Database Migration Assistant for Unicode (DMU) Tool (Doc ID 1272374.1) for more information on using DMU.

ACTIONS

Customer Migration Path

• For customers currently using CSSCAN/CSALTER on Oracle Database releases through 11.2, the impact is minimal since we will continue to support the existing versions of CSSCAN and CSALTER. However, we strongly recommend switching to the DMU wherever possible to take advantage of the advanced DMU product features.
• When migrating to Unicode, customers on Oracle Database releases after 11.2 can uptake DMU directly with no compatibility concerns.
• Customers wanting to migrate a database to a non-Unicode character set, which we strongly discourage, may still use the DMU to look for invalid character representation issues in the database. The DMU will not correctly identify data expansion issues (column and data type limit violations), if the migration is not to Unicode. To be able to locate such issues with CSSCAN, the customers will have to perform the migration before upgrading to any release beyond 11.2.
• While we strongly recommend choosing Unicode (AL32UTF8) for the database character set, there is no plan to desupport any of the existing valid database character sets. Non-Unicode databases can continue to be upgraded to the next Oracle Database release without having to change the database character set.