ABACKUP command

To perform aggregate backup, enter the ABACKUP command. If you only want to verify the content of an aggregate backup without actually having the data sets backed up, specify the VERIFY parameter. This is like performing a test run of aggregate backup. After aggregate backup verification processing completes, the ABACKUP command stops. No data sets are backed up.
Note: If you have specified the FILTEROUTPUTDATASET parameter on the ABACKUP command, verification processing will place the results of the data set filtering into the data set specified. These results consist of a list of data sets selected for ABACKUP processing. This list can be useful when developing the data set filters in the selection data set.

When you are ready to have your data sets backed up, you can issue the ABACKUP command and specify the EXECUTE parameter. After successful data set verification processing, your data sets are backed up.

The following topics describe in more detail what occurs when you issue the ABACKUP command with the VERIFY or EXECUTE parameter.

The agname and EXECUTE | VERIFY parameters are required on the ABACKUP command.

For agname, specify the name of the aggregate group you defined earlier. See Defining an aggregate group.

The VERIFY parameter lists all input tape volumes that are required at the aggregate backup site if a subsequent ABACKUP EXECUTE command is submitted. It also performs verification processing of both migrated and nonmigrated data sets, detecting validation errors before data movement actually takes place.

For unittype in UNIT(unittype), specify the tape unit type to be allocated for the aggregate backup data, control, and instruction/activity log files.

After successful aggregate backup of an application, the optional MOVE parameter causes ABACKUP to delete all data sets in the application. The different data sets are affected in the following manner:
Note: If users with restricted RACF® command authority issue the ABACKUP command with the MOVE parameter, they must have ALTER access authority for each data set belonging to the aggregate group, or aggregate backup verification fails with a related error message.

If a user has comprehensive RACF command authority and issues the ABACKUP command with the MOVE parameter, no RACF checking of individual data sets is required.

The optional FILTEROUTPUTDATASET(dsname) parameter allows DFSMShsm to direct a fully-qualified list of data set names from the filtering criteria from the INCLUDE/EXCLUDE lists, as well as the names of ACCOMPANY, ACCOMPANYEXCLUDE, ALLOCATE, and ALLOCATEEXCLUDE data sets, to a separate output data set. If the user has restricted ABACKUP command authority, then that user must have a minimum of update authority to append to the existing RACF-protected filter output data set or allocation authority to create a new filter output data set. For dsname, substitute a data set name of up to 44 characters. FILTEROUTPUTDATASET can be abbreviated FODS.

When the FILTEROUTPUTDATASET parameter is specified along with the PROCESSONLY parameter, the filter output data set contains the data sets that pass the filtering criteria and also match the category designated in the PROCESSONLY parameter.

When you use the optional OPTIMIZE(option) parameter with an ABACKUP command it overrides the value specified by the SETSYS ABARSOPTIMIZE command. The OPTIMIZE parameter allows you to specify the performance level used by DFSMSdss during dump processing of level 0 DASD data sets. You can specify that DFSMSdss read one, two, or five tracks at time, or one cylinder at a time.

The optional PROCESSONLY parameter directs DFSMShsm to process only the specified categories. You can specify one or more of the categories L0, ML1, ML2, and USERTAPE. The categories can be specified in any order. Specification of all categories will be accepted and is equivalent to not specifying the PROCESSONLY parameter.

Restriction: During ABACKUP, the PROCESSONLY parameter is processed only for data sets that exist (cataloged) during the creation of the control file. If the data set is deleted between the time that data set filtering is complete and the time that there is a successful search (locate) for it during the building of the control file, there is no way to determine where the data set (L0, ML1, ML2, or USERTAPE) resides, nullifying any meaning the PROCESSONLY option would have had.
PROCESSONLY can be abbreviated to PONLY. The following are recognized aliases for category names:
Category
Aliases
L0
LEVEL0, PRIMARY
ML1
MIGRATIONLEVEL1
ML2
MIGRATIONLEVEL2

The optional, and mutually exclusive, STACK and NOSTACK parameters allow you to override the specified SETSYS ABARSTAPES value, or its default value of STACK. The STACK option causes ABACKUP to place the ABACKUP output files on a minimum number of tape cartridges. The STACK option only applies to output being directed to tape. Using the STACK option and attempting to direct the ABACKUP output files to DASD causes ABACKUP to fail.

The SKIP option causes DFSMShsm to skip data sets on volumes that are protected by PPRC or XRC. That is, these data sets are not backed up.

You might find it helpful to use the following DFSMShsm commands to manage and query an ABACKUP command:

Related reading

For more information about the commands listed in this topic, see DFSMShsm Storage Administration Reference.