FDRARCH JCL Requirements

The following JOB control statements are required to execute the archive maintenance utility.

STEPLIB or JOBLIB DD statement

If FDR is not in the system link list (LNKLST), specifies the program library in which FDRARCH resides. The library must be APF authorized.

EXEC statement

Specifies the program name (PGM=FDRARCH), region requirement (REGION=), and optional PARM= operand. Because the region requirements of FDRARCH may vary depending on the operation being performed, we recommend specifying REGION=0M to get the largest possible below-the-line region.

If a PARM field is specified, FDRARCH uses the data specified as the first control statement, which must be a valid FDRARCH statement. For example,

If FDRARCH is invoked from a user program, Register 1 must follow IBM's convention for passing data from the PARM field.

ARCHIVE DD statement

Specifies the Archive Control File or Extended Archive Control file to be processed; it must be on DASD. Do not specify any DCB parameters. The data set name must contain an index level of “ARCHIVE” in other than the first level index unless DISABLE=DSNCK is specified. The Archive Control File must reside on a DASD volume that has been initialized by the ABR VTOC maintenance utility, FDRABRM, for ABR processing (see VTOC Maintenance Utility (FDRABRM)) unless USERINDEX=YES is specified; USERINDEX=YES is used mainly with FDRAPPL Application Backup, described in Working-with-FDRAPPL.

If creating a new Archive Control File, a SPACE= parameter with secondary allocation and RLSE are supported and recommended. Estimate the number of data sets that will need to be recorded at any one time and calculate the tracks required, figuring on 348 per track on a 3390. However, multi-volume data sets require one entry per volume, and VSAM clusters require one entry for each cluster (on each volume) plus one entry for each component on each volume (for example, 3 for a single volume KSDS). Be sure to add a liberal amount for obsolete entries between reorganizations. The Archive Control File does not take secondary allocations once it has been formatted; you can increase its size by using the DUMP and RESTORE statements of FDRARCH.

If creating a new Extended Archive Control File, the ARCHIVE DD points to the Index component of the EACF cluster. Unlike ACFs which can be variable in size, the Index component must be the maximum size of a BDAM data set, 4369 cylinders (65,535 tracks). Any allocation less than the maximum will result in FDRARCH failing the request. 

For an existing Archive Control File, DISP=SHR can be used. By default, FDRARCH and FDRABR use enqueue and reserve macros to serialize access to the file. For example:

If the ARCHIVE DD statement is not present and ENABLE=DYNARC is specified on the control statement, the Archive Control File is dynamically allocated using the ARCDSN data set name in the FDR Global Options (see ARCDSN in ABR-Options). This is normally the name of the ACF used for ABR auto-recalls. You should always use an ARCHIVE DD statement when referencing an Application Control File.

For EACF, if the ARCHIVE DD statement is not present and ENABLE=DYNARC is specified on the control statement, the Extended Archive Control File is dynamically allocated using the EARCDSN data set name in the FDR Global Options (see EARCDSN in ABR options). This is normally the name of the EACF used for ABR auto-recalls. You must always use an ARCHIVE DD statement when referencing an Extended Application Control File. 

For an existing Archive Control File, DISP=SHR can be used. By default, FDRARCH and FDRABR use enqueue and reserve macros to serialize access to the file.

ABRWORK DD statement

(Optional) Used by the REORG statement if UNCAT=YES is specified or defaulted. If this DD statement is present and references a DASD data set, the data set contains a list of all the Archive Backup data sets active at the time of the REORG, their use counts, and catalog status. In addition, the DCB characteristics is set to:

If this DD statement is not present, FDRARCH uses a temporary data set that is dynamically allocated on a volume within the unit name specified by the WORKUNIT operand (default SYSALLDA).

EACF Data Component DD statements

If you are executing the FORMAT statement to format an Extended Archive Control File (EACF), you must specify between one (1) and fourteen (14) DD statements that point to the Data components of an EACF cluster. The ddnames are any names of your choice and are specified in the ADDNAME= operand of the FORMAT; the ARCHIVE DD statement in this case points to the Index component of the EACF cluster to be formatted. Unlike the Index component, Data components of an EACF can be any size greater than 1 cylinder and less than or equal to the maximum size of 4369 cylinders. 

We recommend allocating the maximum sized EACF Data components to allow for the maximum number of EACF data sets without having to REORG and/or MERGE Data components in the future. 

MERGE DD statements

If you are executing the MERGE statement of FDRARCH, you must supply two or more DD statements pointing to the input Archive Control Files. The ddnames are any names of your choice, and are specified in the MERGEDDNAMES= operand of the MERGE statement; the ARCHIVE DD statement points to the output control file.

SORT DD statements

(Optional) If you are executing the REORG statement with SORT=YES or other operands requesting sorting (see SORT= under REORG), you may need to specify DD statements required by your system SORT product (for example, SORTLIB, SORTWKnn).

In particular, a SORTOUT DD statement is required, allocating a temporary or permanent data set with space several times as large as the used space within the Archive Control File being reorganized. If the ACF is very large, this may require a SORTOUT with over 65535 tracks; this can be allocated only as a multi-volume data set, or as a SMS-managed, Extended Format data set.

However, any or all of these SORT DD statements may be omitted if you have requested dynamic allocation of the SORT data sets via the SORTALLOC= operand of REORG.

SYSIN DD statement

Specifies the control statement data set, usually an input stream or DD * data set. It can be omitted if the only control statement is specified by PARM= on the EXEC statement.

SYSPRINT DD statement

Specifies the output message data set. This is a required DD statement and usually is a SYSOUT data set.

SYSUDUMP DD statement

Specifies the abend dump data set. Usually specifies a SYSOUT data set. Although not required, we strongly urge you to always include this DD statement, so that we can help you diagnose error conditions. If you have a debugging aid product on your system that would prevent the desired dump, please add the appropriate one of these statements to the JCL so that a fully-formatted dump is produced.

TAPE1 DD statement

Specifies the file to be used for the backup copy of the Archive Control File. This DD statement is used to output the backup copy of the Archive Control File if the DUMP or REORG statements are executed. It is used to input a backup copy of the Archive Control File if the RESTORE statement is executed. It is not required for other functions.

TEMPARCH DD statement

(Optional) Used by the REORG statement if both SORT=YES and SIMULATE are specified. It is not required, but if present, it will contain an image of what the Archive Control File would have looked like if the REORG had really been done; it can be used as input to reports in later steps. It should be large enough to hold all records remaining after the simulated REORG; we suggest you make it as large as the Archive Control File being reorganized; do not specify RLSE on the DD statement. See ARCHIVE DD Statement for other considerations on allocation.