Data sets that COPY uses

The COPY utility uses a number of data sets during its operation.

The following table lists the data sets that COPY uses. The table lists the DD name that is used to identify the data set, a description of the data set, and an indication of whether it is required. Include statements in your JCL for each required data set and any optional data sets that you want to use.

Table 1. Data sets that COPY uses
Data set Description Required?
SYSIN Input data set that contains the utility control statement. Yes
SYSPRINT Output data set for messages. Yes
DSSPRINT Start of changeOutput data set for messages when making concurrent copies or FlashCopy image copies. End of change No1
Filter A single data set that DB2® uses when you specify the FILTERDDN option in the utility control statement. This data set contains a list of VSAM data set names that DB2 builds, and is used during COPY when you specify the CONCURRENT and FILTERDDN options. No2
Sequential image copies From one to four sequential output data sets that contain the resulting sequential format image copy data sets. Specify their DD names with the COPYDDN and RECOVERYDDN options of the utility control statement. The default is one copy to be written to the data set described by the SYSCOPY DD statement. Yes
Start of changeFlashCopy® image copiesEnd of change Start of changeFor table space or index space level copies, a VSAM data set for the output FlashCopy image copy of each partition or piece.

For a partition level or piece level copy, a VSAM data set for the output FlashCopy image copy of the partition or piece.

End of change		 Start of changeNo3End of change
Note:
  1. Start of changeRequired if you specify CONCURRENT, FLASHCOPY YES, or FLASHCOPY CONSISTENT and the SYSPRINT DD statement points to a data set.End of change
  2. Required if you specify the FILTERDDN option.
  3. Start of changeRequired if you specify either FLASHCOPY YES or FLASHCOPY CONSISTENT.End of change

The following objects are named in the utility control statement and do not require DD statements in the JCL:

Table space or index space
Object that is to be copied. (If you want to copy only certain data sets in a table space, you must use the DSNUM option in the control statement.)
DB2 catalog objects
Objects in the catalog that COPY accesses. The utility records each copy in the DB2 catalog table SYSIBM.SYSCOPY.

Output data set size

Sequential image copies are written to sequential non-VSAM data sets.

Start of changeFlashCopy image copies are written to VSAM data sets.End of change

Recommendation: Use a template for the image copy data set by specifying a TEMPLATE statement without the SPACE keyword. When you omit this keyword, the utility calculates the appropriate size of the data set for you.

Alternatively, you can find the approximate size of the image copy data set for a table space, in bytes, by either executing COPY with the CHANGELIMIT REPORTONLY option, or using the following procedure:

  1. Find the high-allocated page number, either from the NACTIVEF column of SYSIBM.SYSTABLESPACE after running the RUNSTATS utility, or from information in the VSAM catalog data set.
  2. Multiply the high-allocated page number by the page size.

Filter data set size

Recommendation: Use a template for the filter data set by specifying a TEMPLATE statement without the SPACE keyword. When you omit this keyword, the utility calculates the appropriate size of the data set for you.

Alternatively, you can determine the approximate size of the filter data set size that is required, in bytes, by using the following formula, where n = the number of specified objects in the COPY control statement: 

(240 + (80 * n))

JCL parameters

For the output data sets of sequential format image copies, you can specify a block size by using the BLKSIZE parameter on the DD statement for the output data set. Valid block sizes are multiples of 4096 bytes. You can increase the buffer using the BUFNO parameter; for example, you might specify BUFNO=30, which creates 30 buffers.

Start of changeFor the output data sets of image copies created by invoking either the concurrent copy function or FlashCopy function of DFSMSdss, the required specifications for the output data sets might be different than those for sequential format image copies. For example, the BUFNO parameter does not apply to the output data sets for concurrent image copies.End of change

Cataloging image copies

To catalog your image copy data sets, use the DISP=(MOD,CATLG,CATLG) parameter in the DD statement or TEMPLATE that is named by the COPYDDN option. After the image copy is taken, the DSVOLSER column of the row that is inserted into SYSIBM.SYSCOPY contains blanks.

Start of changeFlashCopy image copy data sets are always cataloged. The DISP= parameter is not specified in the FlashCopy template. After the image copy is taken, unless uncommitted work is backed out of the image copy when FLASHCOPY CONSISTENT is specified, the DSVOLSER column of the row that is inserted into SYSIBM.SYSCOPY contains blanks. If uncommitted work is backed out of a FlashCopy image copy, the DSVOLSER column contains the DB2 checkpoint information for each member.End of change

Duplicate image copy data sets are not allowed. If a cataloged data set is already recorded in SYSIBM.SYSCOPY with the same name as the new image copy data set, the COPY utility issues a message and does not make the copy.

When RECOVER locates the SYSCOPY entry, it uses the operating system catalog to allocate the required data set. If you have uncataloged the data set, the allocation fails. In that case, the recovery can still go forward; RECOVER searches for a previous image copy. But even if it finds one, RECOVER must use correspondingly more of the log during recovery.

Recommendation: Keep the ICF catalog consistent with the information about existing image copy data sets in the SYSIBM.SYSCOPY catalog table.