Backup and Recovery › The Backup Process

The Backup Process

Ensure that the VM directory entry for the CA ACF2 for VM service machine includes MDISK entries for all the minidisks that contain the primary and alternate sets of databases.

Automatic backup processing is done as follows:

1. As the backup begins, CA ACF2 for VM sends all users defined in the NOTIFY operand a message indicating that it is taking a backup. It also notifies these users about the status of the backup, such as completion of step, completion of the entire backup, or abnormal termination.
2. CA ACF2 for VM tries to locate the service machine minidisk at the address specified in the MDISK operand of the BACKUP VMO record. This minidisk must be accessed read and write. If CA ACF2 for VM cannot find this minidisk or it is not accessed in read and write mode, CA ACF2 for VM writes the backup copies to the service machine 0191 disk. If the 0191 disk is not accessed in read and write mode, CA ACF2 for VM terminates with an error message.
3. CA ACF2 for VM copies each database and creates a sequential CMS file. The filename is determined through the @DDSN macro. The filetype is DATUT1. After it copies each database, it erases the previous backup copy and renames the DATUT1 file to BACKUPDB. This process ensures that any previous backup copy remains intact until the new copy is created. The minidisk containing the backup files must be large enough to contain all three files, plus a copy of the largest file.
4. The output data from the backup is in sequential format. You cannot use it as a database for the CA ACF2 for VM service machine. (CA ACF2 for VM supplies a facility for rebuilding databases. See the Post Backup Service Machine for more details.)