Previous Topic: Deferred ARCHIVE Command RestrictionsNext Topic: SMS and ARCHIVE


ARCHIVE Command Syntax

The syntax of the ARCHIVE command is as follows:

ARCHIVE DISP=,RETPD=,EXPDT=,ACCESS=,RPT=CIA,CIAOPT=,DEFER,
NEWNAME=,NEWHLQ=,MEMBER=,FAST=

The following parameters honored for PRIMARY2 processing only:

RETPD2=,EXPDT2=
RETPD2=

Specify a one- to five-digit number indicating the number of days the selected data sets are to be retained in the CA Disk archives. For more information, see the section Year 2000 Considerations in the chapter "General Information" and the sysparm description of EXPDTUSEn in the Systems Guide.

If multiple ARCHIVE (and BACKUP) commands are entered, each can have a different RETPD2= or EXPDT2= value.

You cannot specify both the RETPD2 and EXPDT2 parameters.

EXPDT2=

Specify a date constant in any acceptable CA Disk format. For a list of valid dates, see the section Date Formats in the chapter "General Information." For more information, see the section Year 2000 Considerations in the chapter "General Information" and the sysparm description of EXPDTUSEn in the Systems Guide.

CA Disk treats DSNINDEX Julian expiration dates 99000 (except as described in the next paragraph), 99365, 99366 and 99999 as never scratch dates. That is, data sets whose ARCHVOLS or DSNINDEX records use these dates as expiration dates will not be deleted automatically as expired by the CA Disk function IXMAINT.

By default, CA Disk treats Julian expiration date 99000 as a never scratch date. But if you set sysparm UNCATDEL to a value of Y, CA Disk treats Julian expiration date "99000" as a catalog control date. That is, data sets backed up or archived with an expiration date of 99000 are deleted automatically by the CA Disk function IXMAINT when the data sets become uncataloged. What this means is, if your catalog goes bad or someone uncatalogs some data sets, all backups and archives of these data sets are deleted on the next IXMAINT run. Although you can use EXPDT2=99000 for any purpose, it was intended for use with archiving generation data groups (GDGs), the generations of which are managed by the catalog.

For example, to archive all GDG generations, leaving the two most recent on disk, deleting generations that roll off of the catalog, specify sysparm UNCATDEL with a value of Y and:

SCAN REALVOLS
SELECT CRITERIA=(GDGGEN,LE,-2)
ARCHIVE DISP=RECAT,EXPDT2=99000

If multiple ARCHIVE (and BACKUP) commands are entered, each can have a different RETPD2= or EXPDT2= value.

You cannot specify both the RETPD2 and EXPDT2 parameters.

DISP=

The disposition for each data set successfully processed. The default value is DELETE.

Specify RECAT for recatalog to the CA Disk pseudo-volume following deletion. This makes the data set available for auto-restore by the CA Disk Catalog Management hook. The data sets can later be made unavailable for auto-restore by the CA Disk Catalog Management hook by the MERGE command UNCATALOG parameter.

Specify NOCAT to have CA Disk take no catalog action against the data set. This causes CA Disk to delete the data set from DASD, and to leave it cataloged to its original volser.

Specify KEEP if the data set is not to be deleted, nor have its catalog status changed. Even better than specifying DISP=KEEP is to use the BACKUP command.

RETPD=

By default, data sets are assigned a retention period equal to the value of sysparm RETRETPD. To override the default, specify a one- to five-digit number indicating the number of days the selected data sets are to be retained in the CA Disk archives. For more information, see the section Year 2000 Considerations in the chapter "General Information" and the sysparm description of EXPDTUSEn in the Systems Guide.

If multiple ARCHIVE (and BACKUP) commands are entered, each can have a different RETPD= or EXPDT= value.

You cannot specify both the RETPD and EXPDT parameters.

EXPDT=

By default, data sets are assigned a retention period equal to the value of sysparm RETRETPD. To override the default, specify a date constant in any acceptable CA Disk format. For a list of valid dates, see the section Date Formats in the chapter "General Information." For more information, see the section Year 2000 Considerations in the chapter "General Information" and the sysparm description of EXPDTUSEn in the Systems Guide.

CA Disk treats DSNINDEX Julian expiration dates 99000 (except as described in the next paragraph), 99365, 99366 and 99999 as never scratch dates. That is, data sets whose ARCHVOLS or DSNINDEX records use these dates as expiration dates will not be deleted automatically as expired by the CA Disk function IXMAINT.

By default, CA Disk treats Julian expiration date 99000 as a never scratch date. But if you set sysparm UNCATDEL to a value of Y, CA Disk treats Julian expiration date 99000 as a catalog control date. That is, data sets backed up or archived with an expiration date of 99000 will be deleted automatically by the CA Disk function IXMAINT when the data sets become uncataloged. What this means is, if your catalog goes bad or someone uncatalogs some data sets, all backups and archives of these data sets will be deleted on the next IXMAINT run. Although you can use EXPDT=99000 for any purpose, it was intended for use with archiving generation data groups (GDGs), the generations of which are managed by the catalog. For example, to archive all GDG generations, leaving the two most recent on disk, deleting generations that roll off of the catalog, specify sysparm UNCATDEL with a value of Y and:

SCAN REALVOLS
SELECT CRITERIA=(GDGGEN,LE,-2)
ARCHIVE DISP=RECAT,EXPDT=99000

If multiple ARCHIVE (and BACKUP) commands are entered, each can have a different RETPD= or EXPDT= value.

You cannot specify both the RETPD and EXPDT parameters.

ACCESS=

You can override the value of sysparm VSACCESS for this BACKUP command by specifying ACCESS= with a value of C for cluster-level access, or L for logical-level access.

RPT=CIA

This option allows the Cluster Internal Analysis (CIA) report to be produced as a by-product of live mode archive of VSAM clusters. The CIA report will not be produced during simulate runs. See the section CIARPT – Cluster Internal Analysis Report in the chapter "Reports"CIARPT —Cluster Internal Analysis Report for further information.

CIAOPT=

If the RPT=CIA parameter is specified, this parameter can be used to specify which options are to be used. See the section CIAOPT=(CATLG, DETAIL, DUMP) in the chapter "Reports"CIAOPT=(CATLG,DETAIL,DUMP) for further information.

NEWNAME=

An alternative data set name to be assigned to the DSNINDEX record for the archived data set (a relative GDG name is also permitted). The archived version is indexed under this name, which can be up to 44 characters in length.

DEFER

This parameter can be specified on the ARCHIVE command to indicate that the request is to be queued for later processing. For an overview of this function, see Deferred RequestsDeferred Requests.

NEWHLQ=

An alternative high level qualifier can be assigned to each data set selected for archive (a relative GDG name is also permitted). The archived version is indexed under this name, which can be up to 44 characters in length.

MEMBER=

To archive a subset of a partitioned data set, this parameter is used to specify one or more member and/or pattern names to constitute the archived version of the data set. A valid pattern for member level processing is any character followed by the '/' pattern character. This is the only valid pattern character available for this parameter. As as example:

MEMBER=(A/,GEN/)

For detailed information on the '/' pattern character, see the description of Slash (/) Pattern in DSN= in the section Data Set Names and Patterns in the chapter "General InformationSlash (/) Pattern in DSN=."

A maximum of 25 entries is supported and is only valid when an immediate archive is being done (not a deferred archive).

Note: While only the specified members are processed, the entire data set will be deleted. If this particular DSNINDEX record is restored, only the members residing in this index record are recovered. Members residing in the partitioned data set prior to the ARCHIVE will not exist in the data set after RESTORE.

FAST=

Specify YES or NO to indicate whether the data sets selected for processing should use the Fast Archive path.