Building a job

This space will step you through the process of constructing a single FATSCOPY batch job to achieve a specific task. It is not intended to be read start-to-finish. We will ask you step-by-step questions about the details of your task. As you answer the questions and click on links, you will be guided through the tutorial where we explain which FATSCOPY keywords you will need to add to your batch job stream.

(If you need to go back to where a link took you from, use ALT-Left Arrow.)

We will cover the FATSCOPY parameters needed for most tasks. To create another job for a different task, you should go through the tutorial again from the beginning.

The symbol ¤ is used to mark keywords which may have had their defaults changed, and require special care. 

For example: there is a parameter called TMSDATA. The delivered default value is TMSDATA=COPY. In this tutorial, we tell you to add a TMSDATA= parameter only if you want to change the value to NOCOPY. But if your site’s installation has changed the default to NOCOPY, then you would need to add the TMSDATA= parameter if you wanted the value to be COPY.

 This is not a reference document. If you want to look up precisely how a specific keyword works or what the limits are for its parameters, you should refer to the FATSCOPY User Manual.

 After you have finished this tutorial, you may wish to refer to some additional FATSCOPY resources which are available.

• The FATSCOPY page on the BMC Support site.
• The FATSCOPY User’s Manual, which contains documentation on all keywords and describes their parameters
• FATSCOPY ISPF panels, which allow you to create and submit FATSCOPY jobs without directly coding JCL and control statements. These panels are described in the Getting started with FATSCOPY, an online demonstration of using FATSCOPY batch jobs and ISPF panels, at BMC Support.
• The FATSCOPY Concepts.

You can create a FATSCOPY job to copy one file, one volume, or hundreds of files or volumes. When you are getting ready to set up a FATSCOPY job, you should think about the criteria you will use to specify what input files you want to copy and where you want the output files to go.

• Which do you have: a list of volumes you want to copy, or a list of file names (or filename patterns)?
• Do you want to limit the files selected by certain criteria such as expiration date, creation date, file location, file sequence number, creating job name, last-used job name, data set size, etc.?
• Where do you want the output files to go? Do you want to stack them onto as few tapes as possible? Do you want to add them to a tape which already has files on it?
• Do you want the output files to be cataloged? Should they be renamed?
• Do you want to expire the input files after they have been copied?

To use FATSCOPY, you must have a supported tape management system on your z/OS system. FATSCOPY currently supports IBM’s DFSMSrmm, CA Technologies’ CA 1 and TLMS, BMC's Control-M/ Tape, and ASG’s Zara. FATSCOPY depends on tape management information for some of the data set selection process and to obtain scratch volumes for the output files. FATSCOPY will update most of the tape management information for the output files, which can be copied from the tape management information for the input files.

FATSCOPY will copy all types of tape files whose locations are not recorded in external databases. Some common types of files are DB2 data files, SMF logs, DFDSS datasets, and ABR backups. In addition, FATSCOPY can copy certain special types of files, such as HSM migration files and ABR migration and application files, and update their databases to reference the copied files. Other special files, which are recorded in external databases, are not good candidates for copying with FATSCOPY at this time. Some of these include CA-SAR, SAS, CA-Dispatch, and Tivoli Storage Manager archive tapes. Some special files which are recorded in external databases can be copied with an “image copy”, which is an exact bit-for- bit copy of an input tape including the volume serial number; OAM tapes can be copied with an image copy. Contact BMC Support if you have questions about a specific file type.

You use FATSCOPY control statements to specify what input files you want to copy (or simulate copying), where the output files should go, and what should be done with the input files after a successful copy. A FATSCOPY job:

• always contains either a SIM or COPY control statement (to tell FATSCOPY how to produce the output files, and some miscellaneous details);
• usually contains one or more SELECT control statements (to specify what input files you want to copy);
• may contain EXCLUDE control statements (to further limit the selections);
• may contain a RENAME control statement (to change the names of the output files).

We’ll first show you examples of three complete FATSCOPY jobs. You don’t need to understand the details in these examples. They are here just to illustrate some of the things you can do with FATSCOPY, and what some typical complete jobs look like.

You can find more examples in the JCL library provided with FATSCOPY, which you can use as models for building jobs.

Preview Example 1: Simple SIM

A “Sim” job performs a simulation, which is where FATSCOPY goes through the data set selection process you specify, so that you can see which data sets FATSCOPY would actually copy in a Copy job with the same selection parameters specified.

In this example,

• EXPIRED=YES means both expired and unexpired files are eligible for selection. MULTIFILE=YES means that tapes containing multiple files will be eligible for selection.
• All data sets from volume CCR009 will be selected.
• Any data sets on volume CCR001 which were created on or after January 1, 2010 will be selected (CRDATE.GE.2010001).
• A report about the data sets selected or bypassed will be written to the SELRPT data set.
• A report showing what the copy results using these SELECTs would look like will be written to the COPYRPT data set.
• Any error messages will be written to the ERRORRPT data set.
• The optional CHECKPT keyword and DSNTABLE DD statement causes a disk data set named JAT.FATSCOPY.RESTART to be allocated and cataloged. The selection results of this simulation are saved in this data set; it can be used in a “Restart” job to perform a copy of these data sets without re-specifying all of the FATSCOPY control statements.

No tapes are actually mounted, and no files are copied. The SELRPT data set shows the selection results, and the COPYRPT data set contains a report on the simulated copy results. Click here to see what some sample Select and Simulation Reports look like.

Preview Example 2: Simple COPY

A “Copy” job actually copies data sets. You might want to “stack” data sets from several underutilized tapes onto a single tape or migrate data sets from one type of tape device to another.

In this example,

• EXPIRED=NO means that files that are marked as “expired” in the tape management system will not be copied.
• All cataloged tape data sets with the first two index levels “JAT.TEST” in their data set names will be selected.
• A report about the data sets selected or bypassed will be written to the SELRPT data set.
• A report on the copy results will be written to the COPYRPT data set.
• Any error messages will be written to the ERRORRPT data set.

Click here to see what a sample Copy Report looks like.

Preview Example 3: Another COPY

This shows an example of a COPY with more complex selection criteria, and more parameters specifying where you want the output files to go. This shows an example of how you can “stack” data sets onto as few output tapes as possible.

In this example,

• The output files will not be cataloged (CAT=NO). You might use this when you are making copies for a backup site, and want the original files to remain cataloged on your system.
• All the output data sets will be “stacked” onto as few tapes as possible (EXPDTGROUP=999).
• Only data sets on single-file tapes will be eligible for copying (MULTIFILE=NO); if they’re already “stacked”, they won’t be copied.
• Files that are marked as “expired” in the tape management system will not be copied (EXPIRED=NO).
• To be selected, the first 2 index levels of a data set name must be “JAT.TEST”. Furthermore, the EXCLUDE statements (which must precede the SELECT statement) specify that any data set with a 3^rd index level “DEPT2” will not be selected, nor will any with a 4^th index level “SALES”. So,
  • JAT.TEST.DEPT3.PAYROLL will be selected
  • ABC.TEST.DEPT3.PAYROLL will not be selected (1^st index level doesn’t match)
  • JAT.TEST.DEPT2.PAYROLL will not be selected (due to 1^st EXCLUDE)
  • JAT.TEST.DEPT3.SALES will not be selected (due to 2^nd EXCLUDE)
• A report about the data sets selected or bypassed will be written to the SELRPT data set.
• A report on the copy results will be written to the COPYRPT data set.
• Any error messages will be written to the ERRORRPT data set.
• The optional AUDIT DD statement causes a disk data set named JAT.FATSCOPY.AUDITDSN to be allocated and cataloged. A record of the files copied is saved in this data set. The FATAUDIT program, which is provided with FATSCOPY, can be used to display a readable Audit Report.

We will now go through a series of Questions to help guide you through the process of constructing your own FATSCOPY batch job. Text in lowercase shows where you have to substitute your own value.

This section provides information about the following topics: