The dbexport and dbimport Utilities

The dbexport utility unloads a database into ASCII files; the dbimport utility creates a database from ASCII files. You can use the -d option of the dbexport utility to specify that only the TEXT or BYTE data descriptor is written for a TEXT or BYTE data column that is stored on an optical platter. Migrating the descriptor permits a TEXT and BYTE data object that is stored on an optical platter to be shared in multiple databases, eliminating the need to store duplicate copies of it on an optical platter. The TEXT or BYTE data descriptor contains information about the location and size of the data. The descriptor holds all the information that is necessary to retrieve a TEXT or BYTE data object from an optical platter. If you do not specify the -d option, dbexport exports both the descriptor and the TEXT or BYTE data object.

If you want a different cluster arrangement in the destination database and you can afford the space on an optical platter, you can choose to duplicate the TEXT and BYTE data in a different cluster.

The following dbexport command exports the stores_demo database to the exstores_demo directory, exporting only descriptors for tables that contain TEXT or BYTE data columns. (The -c instructs the program to ignore all errors except fatal errors, -o specifies the output directory, and -q suppresses the echo of SQL statements.)

dbexport -c -d -o exstores_demo -q stores_demo

TEXT or BYTE data descriptors remain valid from one Optical Subsystem to another because each volume on an optical platter contains an internal tracking label. In addition to TEXT or BYTE data, each WORM volume contains a label with the following information:

• Family name
• Family number (also referred to as the optical-storage subsystem serial number)
• Volume number
• Optical-storage-subsystem identifier

Each optical-storage-subsystem device, whether a stand-alone drive or jukebox, has a unique optical-storage-subsystem identifier. The optical-storage subsystem uses the identifier and the family name to uniquely identify the family within the optical-storage subsystem. Using the identifier and the family name becomes an important issue in a distributed database environment where more than one optical-storage subsystem might be available.

For instructions on how to use dbexport and dbimport, see the IBM Informix: Migration Guide.