Source volume definition parameters

Use source volume definition parameters to define source data sets.

The group/pair parameters that are required in Step 2 are described in the following list.
Delete Existing Target Data Sets
This parameter causes zDMF to delete any data sets that have a High-Level Qualifier (HLQ) equal to the one defined in the TARGET DSN parameter of the SET statement, and that have a second-level qualifier equal to the name of the group.

This parameter can be used to clean up any target data sets that were created in a previous activation of the same group and target HLQ combination. The deletion is done after the source data set selection and the validation phase of Activation, but before any new target data sets are allocated for the group. Refer to Target volume definition parameters for a complete description of how target data set names are generated. Valid input characters are Y and N:

  • Y specifies that an existing target data set can be replaced. The existing target data set is erased, and the new target data set is allocated.
  • N specifies that an existing target data set cannot be replaced. The migration fails if the target data set name exists.
    Note: This parameter affects all data sets defined for the group. N (or No) is the default setting.
Early Data Set Completion
Use this parameter to automatically divert any data sets in the group that are not currently allocated, immediately after the group enters mirror state.
  • N (NO): All data sets remain in mirror state until you enter the DIVERT command for the group.
  • Y (YES): Unallocated data sets are diverted automatically, immediately after synchronization, and are then eligible for completion processing as if the group itself was diverted.
  • D (DROPOPEN): Any open file is dropped unconditionally from the migration and the group is set to Pending Completion, and then Complete state shortly after.
  • W (WAITOPEN): Any open file causes the group to be left in mirror state and must be manually diverted when the file becomes available.
Data Set Allocation Sequence
Valid input characters for this option are D, S, and N:
  • D (DATASET), specifies to process data sets in ascending name sequence.
  • S (SIZE), specifies to process data sets in descending size sequence. The largest data sets are processed first and the smallest are processed last.
  • N (NONE), specifies to process data sets in the order that they are selected for processing. This order might appear random.
Note: N is the default setting.
VSAM Sphere
This parameter specifies whether, for any VSAM cluster that is copied, all associated AIX® clusters and paths are copied also. Individual names of sphere components do not need to be specified. Only the base cluster is eligible to be migrated if SPHERE(N) is specified; individual AIX clusters are ineligible.

Valid input characters are Y and N:

  • Y specifies that all associated AIX clusters and PATHs are to be copied.
  • N specifies that only the selected base cluster is copied.
Note: Y is the default setting; however, when migrating non-VSAM data sets, zDMF ignores the SPHERE parameter.

If SPHERE(Y) is specified, then the selection of the base cluster is determined by the source selection mask, however, multiple related components (KSDS|ESDS + AIX + PATH) are migrated in a single action, and they might not have a common naming structure.

Tolerate Source Allocation Failure
Valid input characters are Y and N:
  • Y indicates that GROUP activation continues and the SOURCE data set is skipped. All skipped data sets contain a reason code/short message that indicates the reason it was skipped. This message can be displayed by using the data set display commands in the TSO monitor or MVS™ console commands.
  • N indicates that GROUP activation ends if any SOURCE data set defined by a MASK or fully qualified data set name on the SET command fails to be selected because it is an invalid data set type or has invalid device or catalog information that makes it impossible to migrate.
Note: The Tolerate Source Allocation Failure parameter that is displayed in the TSO Monitor Option 1 is identical to the TOLERATE_SOURCE_VALIDATE_FAILURE CLI parameter (see TOLERATE_SOURCE_VALIDATE_FAILURE | TOL_SRC_SEL_FAIL | TSSF).
Build Data Set Exclude List
This parameter is a method to eliminate data sets from the selection list.

Valid input characters are Y and N:

  • Y specifies to allow the creation of the exclude list; you are asked to enter the masks later.
  • N specifies not to create the exclude list.
Tip: You can also specify multiple data set names or data set masks to exclude after you create the group, by modifying the group definition. For more information, see Modifying a group definition to exclude multiple data sets.
Migrate Only Specified Volumes
Use this parameter to migrate selected volumes for data sets that spread across multiple volumes. If you have a data set that occupies more than one volume, you are able to migrate only specific volumes without having to migrate all of them.
Note: Setting this parameter to Yes prevents consolidation of all volumes; see information about the Consolidate All parameter in Target volume definition parameters.

Valid input characters are Y and N:

  • Y allows the migration of one or more volumes that are specified in the Source Volume(s) parameter.
  • N allows the migration of all volumes, including the volumes that are not specified on Source Volume(s) parameter.
Note: Setting this parameter to Yes is not recommended because zDMF moves all of the volumes where data to be migrated might be located.

Consider the following example:

You have 10 volumes in the same storage pool that all need to be migrated by zDMF (VOL001-VOL010). To reduce the total number of data sets per group, you migrate them in two groups (GROUP1: VOL001-VOL005 and GROUP2: VOL006-VOL010). A multi-volume data set (DSN1) is on volumes VOL005 and VOL006. If you specify MIGRATE_ONLY_SPECIFIED_VOLUME(YES), DSN1 is in included in both GROUP1 and GROUP2. If GROUP1 is active and you activate GROUP2, DSN1 is excluded from GROUP2 and when both groups are terminated, the portion of DSN1 on VOL006 is not migrated. With MIGRATE_ONLY_SPECIFIED_VOLUME(NO), regardless of which GROUP is activated first, all of the volumes of DSN1 are moved.

Additionally, when you move data sets with MIGRATE_ONLY_SPECIFIED_VOLUME(YES), if the last volume is not included in the migration, the data set is not eligible for active termination and the application must be stopped.

MIGRATE_ONLY_SPECIFIED_VOLUME(YES) is intended to be used when a portion of a data set is moved by a volume level migration product such as TDMF.

Source Data Set Name/Mask
Enter the source data set name for this required parameter. To specify multiple data sets as source, use the wildcard (*) at the end of the data set name; for example: PROD.PARM*

Both the source and target data sets must resolve to compatible devices. These devices might be within the same storage array, or they might be on different storage arrays. All extents for the source data sets must be on devices that are currently online and accessible. The maximum number of data sets for a group is 50,000.

For a detailed listing of both supported and unsupported data set types, see Supported storage subsystems and data set types.

Source Volume(s)
Specify the source data volume serial number. If a volume is specified, the volume VTOC is used as the source for data set names to be included in the group. Data sets in the VTOC are then checked to make sure that they are cataloged; if they are, then they are included in the group. If you select a VSAM data set, ensure that either the data set mask or data set name identifies the component name in the VTOC. Only data sets returned by the catalog search and having a matching volume serial number are selected for migration with the current group.

You can specify up to 16 source volumes in a group. If you specified fewer than 16 volumes and want to add more volumes later on, refer to Modifying a group definition to add more source volumes.

Include by data set name
This parameter specifies a data set with records that contain data sets or masks to be included in the migration. This data set must be a fixed record length physical sequential or PDS with a member name specified. This data set is read at VERIFY and ACTIVATE and each record that does not start with a wildcard character (*) is scanned for a valid data set name or mask. The data set name or mask is then included as part of the group. If no records contain valid data set names or the data set cannot be found or opened, an error is indicated. This parameter can be used as the only method of defining data sets to be migrated or it can be used in addition to the Source Data Set Name/Mask and Source Volume(s) parameters, with the exception that any volumes that are coded do not affect the inclusion of data sets named in the Include by data set name file.
Data Set Exclude Mask
  • Define one or more data sets to be excluded. This parameter can end with a wildcard character (*). If the SOURCE parameter causes several data sets to be selected, the EXCLUDE parameter can be used to eliminate some of them from the selection list.
  • A comma-delimited list of up to 127 data set names or masks can be specified.
  • Use the IMport command to include an installation-defined set of data sets/masks by specifying a data set name (up to 44 characters long) and a member name to import (up to 8 characters long). You will be allowed to append more data sets to the end of the list after the import. If you append, make sure a comma (,) is the last character of the imported data.
  • Examples of data set exclude masks:
    PROD.PARMLIB
PROD.PARMLIB,SYS2.PARMLIB,SYS3.PAR*
PROD.PAR*,PROD.PARMLIB.**
Exclude by data set name
This parameter specifies a data set with records that contain data sets or masks to be excluded from the migration. This data set must be a fixed record length physical sequential or PDS with a member name specified. This data set is read at VERIFY and ACTIVATE and each record not starting with a wildcard character (*) is scanned for a valid data set name or mask. The data set name or mask is then added to the exclude list for the group. If no records contain valid data set names or the data set cannot be found or opened, an error is indicated. This parameter can be used as the only method of defining data sets to be excluded from the migration or it can be used in addition to the Data Set Exclude Mask parameter.
Exclude RLS data sets
This parameter is used to exclude any VSAM data sets that can be opened in RLS mode according to the RLSDATA LOG parameter from the migration group. If the RLSDATA LOG parameter is not NULL according to a LISTCAT, then the data set is deemed as a potential RLS data set and is excluded from the migration group.

Valid input characters are Y and N.

Exclude PDS data sets
This parameter is used to exclude any PDSE data sets from the migration.

Valid input characters are Y and N:

Exclude BDAM data sets
This parameter is used to exclude any DSORG type DA data sets from the migration.

Valid input characters are Y and N.

Exclude by storage class name
This parameter is used to exclude data sets in a particular storage class from the migration.
Set EATTR to extended OPT
This parameter allows data sets that DFSMS defines as able to exist in EAS space to be allocated in EAS space, regardless of the current source data set setting or data class designation.

Valid input characters are Y and N.

Convert to Extended Format

This parameter specifies that the selected source data sets should be converted to extended format during migration. This parameter can be used with the ENCRYPT(YES) parameter to both convert to extended format and encrypt a data set during a migration. If this parameter is set to Y all data sets selected for the group will be examined to determine if they can be converted. Some data sets can be converted due to their block size or their existing format.

Data sets that cannot be converted due to their format are identified with message GZD4104E during the activation phase. This can cause the group to terminate in activation unless TOLERATE_SOURCE_VALIDATION_FAILURE is set to YES. If TOLERATE_SOURCE_VALIDATION_FAILURE is set to Y the data set is skipped and activation continues. This allows the user to migrate the data sets that can be converted in the group.

Data sets that cannot be converted due to their block size are identified by the message GZD4230I during the activation phase. These data sets cannot be converted due to their existing block size, because adding 32 bytes to the existing block size results in fewer blocks per track, which is not supported.

For example, a 512-byte block size data set would result in 544-byte block size data set. At a block size of 512 the track will hold 49 blocks; however, when the block is expanded to 544 bytes, the track will only hold 48 blocks, thereby reducing the number of blocks per track which is not valid for zDMF. In this example, the data set would be skipped with a GZD4230I message, and the group would either terminate or continue depending on the TOLERATE_SOURCE_VALIDATION_FAILURE setting.

VSAM data sets have physical block sizes that are multiples of 512 and there are 28 valid physical block sizes for VSAM. Of those 28 physical block sizes, three are not able to be converted. Those are 512, 1536, and 18,432 . Sequential data set physical block sizes are in 1-byte increments from 4 to 32760. 16.7% of those possible block sizes cannot be converted. Looking at sequential target data sets during mirror can be misleading. When the target data set is allocated, the rnd-of-file block is set at the same location as the source. Adding records beyond that location are not read when using standard access methods (such as browse) to read the target data set. During divert, the new end-of-file block number is copied from the source to the target, and the target data set is renamed to the source data set name. After divert you should be able to see the blocks you added.

Valid input characters are Y and N.

Rotate Encryption Keys

This parameter is used to rotate from one encryption key to another. Setting this parameter to Y causes zDMF to obtain the current keylabel for the data set from the VVDS and compare it to the keylabel in the DFP segment of the data set profile, a keylabel specified in the zDMF group control cards, or a keylabel defined in the data sets current data class. If one of the three data set keylabel sources are different than the current keylabel in the VVDS then zDMF will rotate from the key in the VVDS to the new key from one of the three key sources. If all three key sources are null, the data set will be decrypted.

A data set that is not already encrypted will not be processed by a group with ROTATE(YES). These data sets will either be skipped or will cause the group to fail, depending on the TOLERATE_SOURCE_VALIDATE_FAILURE option. If this option is set to YES, data sets that are not already encrypted will cause the data set to be skipped. If this option is N, any data set included in the group that is not encrypted will cause the group to fail at activation. Therefore, it is incompatible to specify both ROTATE(YES) and ENCRYPT(YES) and this combination will fail during group promotion.

Note: If the data set security profile contains a DFP segment with a valid key, the data set cannot be decrypted.

Valid input characters are Y and N.