When you start DFSMSrmm, the CDSID ID is compared to the ID in the control data set control record. If the IDs match, DFSMSrmm startup continues. If the control data set does not have an ID, DFSMSrmm creates the ID in the control record from the CDSID. If the IDs do not match, DFSMSrmm startup fails and DFSMSrmm issues a message to the operator to select another parmlib member.

If you do not specify a value for CDSID, you cannot start DFSMSrmm (on toleration systems, a warning message is issued and startup continues as long as the DFSMSrmm control data set does not have a CDSID). See Creating or updating the control data set control record for information about how the DFSMSrmm EDGUTIL utility sets the control data set ID.

Default: None.

This operand is required.

JOB

DFSMSrmm records the accounting information from the JOB statement of the JCL.

STEP

DFSMSrmm records the accounting information from the EXEC statement of the JCL. If you specify STEP and there is no accounting information in the EXEC statement, DFSMSrmm records the JOB statement accounting information.

Default: ACCOUNTING(JOB).

Specify a valid alphanumeric procedure name from 1 to 8 characters. If no name is specified, then no automatic start command is issued. See Steps for automating control data set backup and journal clearing for more information.

Default: None. DFSMSrmm ignores BACKUPPROC on the client system.

Authorization to perform BLP is still dependent on the ICHBLP resource in the RACF FACILITY class, and the ability to use BLP can still be controlled by JES.

RMM

You can use BLP for input from and output to volumes in user status, and for input from volumes in master status.

NORMM

You can use BLP under the normal system controls and DFSMSrmm records the activities you perform on tapes. You can also use BLP for input from and output to volumes in user and master status, and for output to scratch tapes. BLP can be used for reading and writing of master and user status tapes and for output to scratch tapes. BLP read of scratch tapes is not supported.

For scratch tapes written using BLP, DFSMSrmm changes the volume to master status and sets the initialize release action so that the tape is correctly labeled on return to scratch. DFSMSrmm also overrides the logical volume serial number generated by OPEN for BLP output to scratch tapes, so that the correct volume serial number is used for cataloging of data sets.

DFSMSrmm does not allow no label (NL) tapes to be mounted in response to a scratch request. However, you can use BLP to create NL tapes during scratch processing.

Default: BLP(RMM)

DFSMSrmm retains the data set for the catalog retention period if the data set has never been cataloged. DFSMSrmm does not retain the data set if DFSMSrmm detected that the data set was cataloged and then uncataloged during the catalog retention period.

ret_hours is 0-9999 hours. For example, CATRETPD(24) keeps data sets for 24 hours. Set ret_hours to 0 to request that DFSMSrmm does not perform this processing.

Default: CATRETPD(12).

CATSYSID(*) means that all catalogs are fully shared. You must specify an * to specify that catalogs are fully shared so that any data set can be processed by DFSMSrmm on any DFSMSrmm subsystem.

CATSYSID(sys_ID_list) provides a list of DFSMSrmm system IDs that share tape data set user catalogs with this DFSMSrmm subsystem. You can identify up to 16 system IDs for DFSMSrmm subsystems. You must specify a list of system IDs when user catalogs are not shared and must ensure they are synchronized with the DFSMSrmm control data set before you run inventory management vital record processing. Be sure to include IDs for systems that you no longer use but for which you are still retaining tape data sets.

Default: None.

SERVERNAME(servername)

SERVERNAME is a required operand when you specify CLIENT. The servername can either be an IP address, a fully qualified domain name, or a server host name.

An IP address can be specified either as a hexadecimal IPv6 address or a numeric IPv4 address. Specify an IPv6 IP address only when the server is running z/OS V1R12 or later release and TCP/IP on the server system has IPv6 dual mode IP stack. For example, CLIENT(SERVERNAME(1080:0:0:0:8:800:200C:417A)) specifies an IPv6 IP address. See RFC3513 (at http://tools.ietf.org/html/rfc3513) for the syntax of an IPv6 IP address.

When a domain name or a host name is specified, DFSMSrmm uses the domain name system (DNS) to resolve that name into an IP address. The host name can be a maximum of 63 characters. The host name must contain one or more tokens separated by a period. Each token must be larger than one character. The first character in each token must start with a letter. The remaining characters in each token can be a letter, number, or hyphen. For example, CLIENT(SERVERNAME(RMMPLX1.MAINZ.IBM.COM) PORT(1950)) tells DFSMSrmm to start as a client without direct DASD access and to share the tape inventory in access by the RMM server with the host name RMMPLX1 using network IP protocol port 1950.

PORT(PortNumber)

Use this operand to specify the port number to be used for IP communication. The PORT operand is required. Specify a value from 1024 to 65535. Port numbers 1 to 1023 are reserved. Also, the client port number and server port number must match for the systems to communicate.

Default: None.

You can set up authorization so that DFSMSrmm checks for authorization by owner first, and then checks for authorization using the DATASET class and TAPEVOL class. Set up this type of checking by specifying both the OWNER operand and the DSN operand separated by a comma. When the RACF name-hiding function is enabled in the system, this overrides the DFSMSrmm COMMANDAUTH processing. DFSMSrmm processing, when the name-hiding function is enabled, is the same as COMMANDAUTH(DSN).

Default: COMMANDAUTH(OWNER)

Default: DATEFORM(JULIAN)

If you code DISPDDNAME, you must provide the name of the DD card that identifies the data set containing the disposition control statements optionally included in your JCL.

For information about DFSMSrmm disposition processing, see Setting up DFSMSrmm disposition processing.

Default: None.

EDG4054I

The message text provides the device number, volume serial number, volume sequence, the location where the volume is to move, and any message text you defined in the disposition control file.

message_id

You can define any alphanumeric value of up to eight characters.

For information about DFSMSrmm disposition processing, see Setting up DFSMSrmm disposition processing.

Default: Message EDG4054I.

If you do not specify DSNAME, you must specify the data set name in the MASTER DD statement in the DFSMSrmm started procedure. If you specify a name both for DSNAME and MASTER DD, DFSMSrmm ignores the MASTER DD statement.

DFSMSrmm ignores DSNAME on the client system.

When VRSEL and EXPROC are run together in a single EDGHSKP run, volumes that are dropped by VRSEL processing are counted only towards the VRSDROP limit, not to the EXPDTDROP limit.

EXPDTDROP processing counts all processed volumes that are retained only by volume expiration date at the start of the expiration processing run and also which of these are set to pending release. When the EXPROC SYSIN command causes a subset of volumes to be processed, only those volumes are counted.

EXPDTDROP processing is intended to provide limited checking for volumes that are not VRS-retained. The volumes would previously have been through VRSRETAIN limit checking and if retained by VRS, the VRSDROP limit checking as well. It considers how many of the EXPDT-retained volumes expire during the EXPROC run. Those volumes that are not set pending release will be considered by EXPDTDROP limit processing on the next run of EXPROC.

count can be 0 to 2,147,483,647. %age can be 0 to 100.

The default is EXPDTDROP(PERCENT(10),INFO).

To aid analysis of the results of the EXPDTDROP limit checking (when the action is not OFF) you can use the contents of the ACTIVITY file and extended records from the extract file. The EDGJACTP sample generates a detailed report and a summary report of expiration date retained volumes showing why they are set to pending release. See the contents of the EXPDROP and EXPDROPS files.

The GDG option has two operands, CYCLEBY and DUPLICATE.

Duplicate generations are determined within a single VRS sub chain and only if the generation and the generation it duplicates are not already dropped for another reason. For GENERATION based cycles, the duplicate generations are determined in generation (data set name), then creation order. For creation date based cycles, the duplicate generations are determined only in creation order, so they can be detected only when they are created consecutively.

Duplicates are processed within the context of the matching VRS chain or sub chain depending on the DUPLICATE option. DROP is within the context of the VRS chain and all others are within the context of the sub chain. A duplicate generation is considered a duplicate for cycles retention only if the generation it duplicates is retained by the current VRS sub chain.

If you specify YES, DFSMSrmm issues a write-to-operator message during startup only if the date of the last run of expiration processing did not occur within two days of the current date. The message prompts the operator to enter the current date and day of the week. Initialization does not proceed until DFSMSrmm receives a valid reply. If you specify NO for this operand value, DFSMSrmm does not issue a message.

The IPLDATE operand helps prevent you from running expiration processing with an incorrect system date that can result in expiration of unexpired volumes.

Under normal conditions, the operator message requesting the current date is only issued once per IPL of z/OS. If you restart DFSMSrmm, it does not recheck the date.

Additionally, if you run expiration processing daily, a message is not issued. As a result, it is possible under normal conditions to run for a long time and never have the operator prompted to confirm the system date.

Default: IPLDATE(NO)

Specify a value in the range 0-99.

Default: 75.

DFSMSrmm ignores JOURNALFULL on the client system.

If you do not specify JRNLNAME in EDGRMMxx, you can specify a name in the JOURNAL DD statement in the DFSMSrmm started procedure. If you do not specify a journal name in either EDGRMMxx or the started procedure, DFSMSrmm does not provide journaling. If you specify a name in both, DFSMSrmm uses the JRNLNAME value and ignores the JOURNAL DD statement.

DFSMSrmm ignores JRNLNAME on the client system.

NO

Specifies that only the updated copy of the CDS record is to be journaled.

YES

Specifies that the pre-update copy of the CDS record being updated is to be journaled, in addition to the updated copy of the CDS record. Set this option only on a test or recovery system when you plan to use the EDGUPDT utility to duplicate CDS record updates back in the production CDS. As a result of using this option you should plan on providing up to 33% more journal data set space to accommodate the additional records.

Default: JRNLTRAN(NO).

You can override this value when producing individual reports by specifying a parameter to the report program or report utility as described in z/OS DFSMSrmm Reporting.

Default: LINECOUNT(54)

Recommendation: Specify or accept the default value of 10 local tasks on all systems. Most of these tasks are rarely used by DFSMSrmm.

The number of local and server tasks you can use and still successfully start DFSMSrmm is limited by the size of the private region above and below 16MB. To start with more tasks, you will require a larger REGION size.

Default: LOCALTASKS(10)

ADD

Specify this value so new data can be created and no existing data can be destroyed. No existing file on a volume can be re-created, but the last file can have new data added to it. When adding data to the last file, DFSMSrmm checks that the data set name used must match the existing data set name. Select this option when you want the last file on the volume to be extended or a new file added to the volume.

Note: DFSMSrmm enforces the MASTEROVERWRITE(ADD) option on a WORM tape that is in master status. This is done to ensure that you see a message from DFSMSrmm rather than one of a number of symptoms as a result of the tape drive preventing overwrites.

LAST

Specify this value to ensure that when an existing file on a master volume is being written to that only the last file on the volume can be used. The data set name used must match the existing data set name. Select this option when you want the last file on the volume to be used for output.

MATCH

Specify this value to ensure that when an existing file on a master volume is being used for output that exactly the same data set name must be used. Select this option when you want any existing file on the volume to be re-created regardless of whether it is the last file on the volume as long as the same data set name is used.

When you use an existing tape file for output all the files that are higher in sequence are destroyed.

USER

Specify this value to allow any existing file on a master volume to be used for output regardless of the data set names being used and its relative file position on the volume. Select this option when you want validation of master volumes to be just the same as for user status volumes.

When you use an existing tape file for output all the files that are higher in sequence are destroyed.

Default: MASTEROVERWRITE(LAST)

Even if the DASD where the control data set resides is not shared, consider specifying MAXHOLD because it controls how often users on the same system can gain access to the data set when a long task is running. MAXHOLD also influences the amount of virtual storage that DFSMSrmm uses. Increasing the value might improve the performance of individual DFSMSrmm functions, but other users are locked out from the control data set for a longer time. In a shared environment, the device is also reserved for a longer time, possibly impacting users of other data on the volume.

You do not normally need to alter this value. If your system is storage constrained, you might lower the value from 100 to reduce the amount of virtual storage that DFSMSrmm needs, and therefore its demand on real storage.

In a shared DASD environment in which you are using global resource serialization to convert the DFSMSrmm control data set reserve, you could increase the value to reduce the frequency of the release/reserve, which reduces the traffic on the global resource serialization ring. However, this action causes more virtual storage to be used and increases the time between releases of the control data set. The result is that other users wait longer before gaining access to the control data set.

Specify a value between 10 and 500. Under normal conditions, use the default value.

Default: MAXHOLD(100)

MAXRETPD is always used to determine the volume expiration date. The volume expiration date can be ignored when EXPIRYDATEIGNORE is specified on all vital record specifications. MAXRETPD is important if the vital record specification specifies UNTILEXPIRED and the decision is based on the volume expiration date, or if the volume is managed by the EXPDT retention method. For the VRSEL retention method, the preferred method is to put the retention requirements all into the vital record specifications and make MAXRETPD=RETPD. If users are allowed to specify expiration and retention period to override vital record specifications, select a MAXRETPD that covers the maximum retention period they would like to enforce. If this forces 99365 to be reduced, define vital record specifications for any data that should be permanently retained, like DFSMShsm tape data.

Use MAXRETPD to set limits on the values that can be specified for EXPDT and RETPD. If the retention period or expiration date specified in the JCL or management class exceeds the MAXRETPD value, DFSMSrmm overrides it and uses the value in MAXRETPD to determine the expiration date. For data sets on volumes that use the EXPDT retention method, if the expiration date determined with the LASTREF data set attribute exceeds the MAXRETPD value, DFSMSrmm overrides it and uses the value in MAXRETPD to determine the expiration date. The volume label has the JCL-specified value because DFSMSrmm does not change tape labels or system control blocks. The control data set contains the JCL-specified value as well as the expiration date calculated from MAXRETPD. You can display the JCL-specified value for information only.

If the DFSMSrmm ISPF dialog or RMM TSO subcommands are used to specify a retention period or expiration date that exceeds the MAXRETPD value, DFSMSrmm fails the subcommand or panel request.

For more information about how to automatically handle expiration date-protected tapes, see Defining pools: VLPOOL for information on EXPDTCHECK.

Default: MAXRETPD(NOLIMIT)

• Expire after days Non-usage, which is equivalent to LASTREF(extra days)
• Expire after Date/Days, which is equivalent to expiration date / retention period

ALL

The use of all applicable management class attributes is enabled. The MC attributes are exploited as they are appropriate, depending on the retention method. The MC attribute Retention Limit is not applied.

NONE

No management class attributes are used. NONE is the default.

VRSELXDI

The use of management class attributes is enabled. The MC attributes are exploited as they are appropriate, depending on the retention method, with the exception of the MC attribute Expire after Date/Days, which is ignored if the data set is on a volume managed by the VRSEL retention method. The MC attribute Retention Limit is not applied. VRSELXDI is recommended if it is desired that the processing of VRSEL managed volumes is the same as in prior DFSMSrmm releases.

Specify a one to eight character name. Here are examples of MEDIANAME that you might define: CART, ROUND, SQUARE, 3420, 3480, TAPE, OPTICAL, and CASSETTE. You can use any name for a media name because DFSMSrmm does not check that the media name you define is a device type that has been defined to z/OS. Use MEDIANAME to identify different types of physical shelf space for different media or to distinguish different media characteristics such as Cartridge Tape and Enhanced Capacity Cartridge System Tape.

Prior to setting or changing the default media name, check for any VLPOOL commands that are using the media name that you plan to change. If you change the default media name that is used for a VLPOOL command for an existing pool of volumes, you must consider changing the media names for those existing volumes. Refer to Changing pool definitions which describes how to use RMM CHANGEVOLUME volser MEDIANAME subcommand to change the volume media name to match the value in the VLPOOL command. Also check your jobs that run EDGINERS to ensure that MEDIANAME, if coded in the execution parameters, is still consistent with the values you are using.

Default: MEDIANAME(3480)

OPTION  DSNAME(DFRMM.PROD.MASTER)
        JRNLNAME(DFRMM.PROD.JOURNAL)
        MEMBER(S5)

DFSMSrmm ignores the MEMBER operand specified in the second parmlib member.

Default: None.

SET

Specify this value when you want volumes moved as a set. DFSMSrmm moves all volumes in the same multivolume set that are retained by vital record specifications. All the volumes are retained in the same location selected by the vital record specification priority or location priority for the volume.

VOLUME

Specify this value when you want volumes moved as individual volumes. DFSMSrmm moves the volumes without considering the location of the other volumes in the multivolume set.

Default: MOVEBY(VOLUME)

MIXED

Specify MIXED when you want messages to be displayed exactly as they appear in EDGMTAB. This includes mixed case message text.

UPPER

Specify UPPER to ensure that all messages are displayed in upper case only. Any mixed case messages in EDGMTAB are translated to upper case before they are displayed.

This is the value used when DFSMSrmm is inactive and utilities need to issue messages.

Default: MSG(MIXED)

Specify YES and DFSMSrmm notifies the owner using the TSO TRANSMIT command to send an electronic message. The owner information in the DFSMSrmm control data set must include either a valid user ID and node name, or a valid Internet e-mail address, for the message to be sent, and the volume record must specify Notify Owner as a release action. DFSMSrmm uses the Internet e-mail address, if it exists, rather than the user ID and node name.

When you use e-mail addresses for owner notification, DFSMSrmm is dependent on you having an existing SMTP server address space configured to support the sending of e-mails. Refer to z/OS Communications Server: IP Configuration Guide for additional information. The SMTP server can be on the same system as the DFSMSrmm started task, or on another system known through NJE.

The Owner 'SMTP' is now a reserved owner name value that you can use to configure the Node name and SMTP server address space or machine name. You use the NODE operand to identify the node that runs the SMTP server, and the USER operand to identify the SMTP server. You must specify both values. When you do not have the Owner SMTP defined, DFSMSrmm uses the JES node name of the running system and SMTP as the SMTP server address space name.

Specify NO for expired volumes with a release action of notify and the librarian performs the notification. The librarian issues the RMM CHANGEVOLUME subcommand with the CONFIRMRELEASE operand to indicate that the notification has been performed.

Default: NOTIFY(NO).

• Whether DFSMSrmm records tape activity
• How DFSMSrmm interacts with system-managed tape support
• How DFSMSrmm enforces the tape mount validation rules described in How does DFSMSrmm validate tape mounts?.

• MANUAL — Manual mode. When DFSMSrmm is running in manual mode:
  • DFSMSrmm does not record tape volume usage or validate volumes.
  • DFSMSrmm does not participate in system managed tape activity; that is, DFSMSrmm does not:
    • Record system managed tape activity
    • Control change of use for system managed volumes
    • Provide information to OAM during entry or eject, or during volume not in library activities
  • You can use RMM TSO subcommands, the DFSMSrmm ISPF dialog, and inventory management functions for all types of media. Updates to the TCDB for system managed tape volumes are controlled by the SMSTAPE(UPDATE(...)) parmlib option and affect these functions:
    • UPDATE(SCRATCH): Volume status updates if you run inventory management processing.
    • UPDATE(COMMAND): Any TSO command change that affects the information in the TCDB, such as EJECT, status change, and media information.
• RECORD — Record-only mode. When DFSMSrmm is running in record mode, DFSMSrmm performs the actions described for manual mode and also:
  • DFSMSrmm records information about tape volumes used on the system, including details about volume owners and data set names.
  • DFSMSrmm does not validate or reject volumes during OPEN processing.
  • DFSMSrmm actions for system managed tape activity are limited to only recording the system managed tape activity.
  • You can use RMM TSO subcommands, the DFSMSrmm ISPF dialog, and inventory management functions for all types of media. Updates to the TCDB for system managed tape volumes are controlled by the SMSTAPE(UPDATE(...)) parmlib option and affect these functions.
    • UPDATE(SCRATCH): Volume status updates if you run inventory management processing.
    • UPDATE(COMMAND): Any TSO command change that affects the information in the TCDB, such as EJECT, status change, and media information.
    • UPDATE(EXITS): Providing information to OAM during entry or eject activities. Controlling the disposition of TCDB volume records at EJECT time.
• WARNING — Warning mode. When DFSMSrmm is running in warning mode, DFSMSrmm performs the actions described for manual and record-only mode. DFSMSrmm issues warning messages when an action would have failed if DFSMSrmm was running in protect mode.
  • DFSMSrmm validates, but does not reject, volumes during OPEN processing.
  • DFSMSrmm actions for system managed tape activity are limited to only recording the system managed tape activity.
  • You can use RMM TSO subcommands, the DFSMSrmm ISPF dialog, and inventory management functions for all types of media. Updates to the TCDB for system managed tape volumes are controlled by the SMSTAPE(UPDATE(...)) parmlib option and affect these functions:
    • UPDATE(SCRATCH): Volume status updates if you run inventory management processing.
    • UPDATE(COMMAND): Any TSO command change that affects the information in the TCDB, such as EJECT, status change, and media information.
    • UPDATE(EXITS): Providing information to OAM during entry or eject activities. Controlling the disposition of TCDB volume records at EJECT time.
• PROTECT — Protect mode. In protect mode, DFSMSrmm is fully operational. In addition to performing the actions described for manual and record-only mode, DFSMSrmm also:
  • Validates all magnetic tape requests and rejects magnetic tape volume mounts under certain conditions, discussed in How does DFSMSrmm validate tape mounts?.
  • Sets the UNCATALOG operand default to YES. UNCATALOG(YES) specifies that DFSMSrmm should uncatalog data sets under conditions described in the UNCATALOG operand description.
  • Fully participates in system managed tape activity:
    • DFSMSrmm records all system managed tape activity.
    • During OAM exit processing, DFSMSrmm validates all changes to the TCDB and fails those that are not allowed.
    • During cartridge entry processing, DFSMSrmm information overrides information provided by OAM.
    • DFSMSrmm always updates the TCDB when DFSMSrmm information is updated, either by command or by DFSMSrmm processing.
    • During eject processing DFSMSrmm can optionally control the disposition of the TCDB tape volume record. Use SMSTAPE(PURGE) to control the purging or keeping of records in the TCDB during eject processing.

Default: OPMODE(PROTECT).

NONE

Specify to prevent DFSMSrmm from enabling the PDA facility. If NONE is specified at startup, DFSMSrmm does not obtain storage for the trace buffer. If NONE is specified when DFSMSrmm is refreshed, it is equivalent to specifying PDA(OFF) and PDALOG(OFF).

OFF

Specify to disable the trace facility after DFSMSrmm initialization. When the trace facility is disabled, trace data is not accumulated.

ON

Specify to enable the trace facility.

Default: PDA(ON)

Default: PDABLKCT(255)

Default: PDABLKSZ(27) for 3390 DASD devices, PDABLKSZ(22) for 3380 DASD devices, and PDABLKSZ(12) for all other DASD devices

OFF

Specify to disable output to the PDA trace data sets.

ON

Specify to enable output to the PDA trace data sets.

Default: PDALOG(ON)

NO

Specify NO to avoid DFSMSrmm Pre–ACS processing using the EDG_EXIT100 installation exit.

YES

Specify YES to enable DFSMSrmm Pre–ACS processing using the EDG_EXIT100 installation exit.

Default: PREACS(NO)

SET

When you retain by set, if any volume in a set is retained by a vital record specification, all volumes in the set are retained as vital records. DFSMSrmm uses highest retention date of all volumes in the set as the retention date for all volumes retained as vital records in a set. If no volume in a set is retained by a vital record specification, DFSMSrmm performs expiration processing by set. DFSMSrmm does not expire volumes in a set if at least one volume in a set is still not ready to expire because it has not reached its expiration date and you have not specified that you want the expiration date ignored.

VOLUME

When you retain by volume, DFSMSrmm retains a volume based on vital record specifications and on the volume expiration date. DFSMSrmm does not consider other volumes in the set.

DFSMSrmm sets an indicator 'retained by set' in the volume information when a volume is vital record specification retained or not expired only because it is a member of a set.

The location where volumes are retained is determined by the MOVEBY option.

Default: RETAINBY(VOLUME).

EXPDT

Specify EXPDT to set the default retention method for new tape volume sets to be based on EXPDT. Data sets and volumes managed by this retention method are never processed by VRSEL inventory management. When you specify the EXPDT retention method the DFSMSrmm inventory management EXPROC processing always attempts to return volumes to scratch on the same run as the volume is released (this is as if the SCRATCHIMMEDIATE attribute is set for the volume). DFSMSrmm maintains a consistent expiration date and time for all data set records of a multivolume data set, unless the volume set is retained by first file.

EXPDT can be specified either as EXPDT or EXPDT(options). The available options are:

LASTREF(extra_days)

LASTREF specifies the default for the data set record LASTREF attribute. The LASTREF attribute specifies the number of days that the data set will be retained after the data set was last referenced by a read or write operation . LASTREF applies only to data sets on volumes managed by the EXPDT retention method.

extra_days is a decimal number between 0 and 93000. The value must not exceed the maximum retention period (MAXRETPD) specified in the EDGRMMxx parmlib member. An extra_days value of 0 has the same effect as the NOLASTREF operand.

When a data set is added to DFSMSrmm on a volume managed by the EXPDT retention method and neither LASTREF nor NOLASTREF are specified for the data set, then DFSMSrmm uses the default LASTREF value.

DFSMSrmm uses the data set LASTREF value to determine the data set expiration date. The extra days are added to the date of last reference. The data set expiration date is set to the maximum of the date calculated using data set LASTREF value and the date resulting from applying the EXPDT, RETPD, or default RETPD. Any reference to the data set by a read or write operation will change the expiration date.

If neither LASTREF nor NOLASTREF are specified in parmlib, NOLASTREF is used by default.

NOLASTREF

NOLASTREF is the default setting for the data set record LASTREF attribute. NOLASTREF specifies that DFSMSrmm does not consider the data set last reference date when determining the data set expiration date. NOLASTREF applies only to data sets on volumes managed by the EXPDT retention method.

RETAINBY(FIRSTFILE | SET | VOLUME)

RETAINBY specifies how DFSMSrmm is to retain volumes or multivolume sets that are managed by the EXPDT retention method:

FIRSTFILE

The expiration date of the first file is used to set the expiration date of the volume or multivolume set. All volumes in a set will have the exact same expiration date and will be released to scratch in the same run of DFSMSrmm inventory management.

SET

The expiration date is the maximum of all data set expiration dates of all the volumes in the set. All volumes in the set will have the exact same expiration date. Any file on any volume of the set can increment the volume expiration date. All files of a multivolume data set have the same expiration date.

VOLUME

The expiration date is determined separately for each volume in the set. Unless defined differently, the expiration date is the maximum of all data set expiration dates on the volume. Each file on a volume can increment the volume expiration date. All files of a multivolume data set have the same expiration date.

Default: RETAINBY(VOLUME).

Note:

1. In a multivolume set, RETAINBY is assigned only to the first volume in a multivolume sequence. All other volumes added to the set assume the same RETAINBY.
2. For volumes managed by the VRSEL retention method, use the RETAINBY option to obtain a similar function.

VRSEL

Specify VRSEL to set the default retention method for new tape volume sets to be VRSEL. This option enables DFSMSrmm inventory management to attempt to match data sets and volumes to vital record specifications, and if a match is found, to determine if the data set or volume is to be retained by VRS. The VRSEL retention method is controlled by all the other VRS related options in parmlib including OPTION RETAINBY MOVEBY.

DFSMSrmm sets a default retention period as follows:

• If you specify RETPD or EXPDT, the value is used as the data set's new expiration date.
• If you do not specify RETPD or EXPDT, then:
  1. DFSMSrmm uses the EXPDT or RETPD allocation attribute of a data class, if all these are true:
     • The data set is associated with a data class, either explicitly by the DATACLAS keyword on the JCL or implicitly by an automatic class selection routine
     • The data class has an EXPDT or RETPD allocation attribute
     • The Storage Management Subsystem is active.
  2. DFSMSrmm uses the management class Expire after Date/Days as the data set's new expiration date, if all these are true:
     • You do not specify RETPD or EXPDT in the JCL or data class
     • The data set is assigned to a management class
     • The management class Expire after Date/Days is set
     • The EDGRMMxx parmlib option MCATTR allows the exploitation of the management class for this type of volume
     • The Storage Management Subsystem is active.
  3. DFSMSrmm uses the default retention period set in EDGRMMxx, if RETPD or EXPDT are not otherwise specified.

Whenever a new data set is written to tape, DFSMSrmm checks whether the volume's expiration date should be updated, based on whether the new data set has a longer expiration date than the volume on which it is written. DFSMSrmm gets the expiration date for a data set from the job file control block (JFCB) or from the management class at open time. If there is a date in the JFCB or management class, DFSMSrmm compares this date to the current expiration date for the volume. If the date in the JFCB allows the volume to be retained longer, DFSMSrmm uses that date to update the volume's expiration date.

If there is no expiration date in the JFCB or management class, DFSMSrmm uses the EDGRMMxx RETPD value to calculate the new expiration date. If the RETPD value allows the volume to be retained longer, DFSMSrmm uses that date to update the volume's expiration date.

If the use of the management class attributes is enabled by the EDGRMMxx parmlib OPTION MCATTR, you can use the management class Expiration Attributes ('Expire after Date/Days' and 'Expire after Days Non-usage') to set the data set EXPDT and LASTREF attributes, where appropriate, to define more specific default retention periods.

RMM ADDVRS DSNAME('RICK.**') DAYS COUNT(30)

Default: RETPD(5)

CONFIRMMOVE

When a volume moves out of a bin, DFSMSrmm does not reuse this bin until the volume move has been confirmed.

STARTMOVE

A bin can be reused as soon as a volume starts moving out of a bin. Extended bin support must be enabled before you can use this operand. See Enabling extended bin support to enable extended bin support.

Default: REUSEBIN(CONFIRMMOVE).

You must run DFSMSrmm with a scratch procedure. You can modify or replace the DFSMSrmm-supplied sample, EDGXPROC, to support your location procedures. You can use the scratch procedure to take any action you would like. For example, you can code the procedure to trigger the required inventory management expiration processing job, to run inventory management, or to take no action.

When an automated tape library detects a low-on-scratch condition, OAM issues write-to-operator messages CBR3660A, CBR3792E, and CBR3794A. DFSMSrmm intercepts these messages and starts the SCRATCHPROC procedure. This procedure runs DFSMSrmm expiration processing to replenish the automated tape library's scratch volumes.

Default: SCRATCHPROC(EDGXPROC)

See Replenishing scratch volumes in a system-managed library for information about EDGXPROC the DFSMSrmm default procedure.

PORT(PortNumber)

Use this operand to specify the port number to be used for IP communication. The PORT operand is required. Specify a value from 1024 to 65535. Port numbers 1 to 1023 are reserved. The port number must be the same for the client system and the server system to establish a network connection.

Default: None.

SERVERTASKS(number)

Use this operand to specify how many DFSMSrmm tasks should be available on the server to handle socket connections from client systems. DFSMSrmm uses this number to determine how many tasks are to be started for processing all client requests on this server. Specify a value from 1 to 999.

Recommendation: Specify or accept the default value of 10 server tasks on each server system. Depending on the number of CLIENT systems, this value should be increased to allow 3 tasks per CLIENT. Most of the local tasks are rarely used by DFSMSrmm. You do not need more server tasks than the sum of local tasks across your CLIENT systems.

The number of local and server tasks you can use and still successfully start DFSMSrmm is limited by the size of the private region above and below 16MB. To start with more tasks, you will require a larger REGION size.

Default: 10.

IBM recommends that you do not use an SMF record type number nnn, but instead use the IBM assigned record number by specifying YES. The IBM assigned record number is type 42, and the subtype is 22. You cannot mix SMF record types. For example, you cannot specify YES for SMFAUD and a record type for SMFSEC.

If you do not specify either YES or a number, DFSMSrmm does not produce audit records.

Default: No audit records

DFSMSrmm ignores SMFAUD on the client system.

IBM recommends that you do not use an SMF record type number nnn, but instead use the IBM assigned record number by specifying YES. The IBM assigned record number is type 42, and the subtype is 23. You cannot mix SMF record types. For example, you cannot specify YES for SMFAUD and a record type for SMFSEC.

If you do not specify either YES or a number, DFSMSrmm does not produce security records.

Default: No security records

NO

Specify NO to prevent DFSMSrmm from calling the SMS ACS processing to obtain management class and storage group names. DFSMSrmm system-based scratch pooling, and scratch pooling and VRS management values based on the EDG_EXIT100 installation exit are used.

YES

Specify YES to enable DFSMSrmm calls to the SMS ACS processing to obtain management class and storage group names. If values are returned by the SMS ACS routines the values are used instead of the DFSMSrmm and EDG_EXIT100 decisions.

Default: SMSACS(NO)

UPDATE

Use UPDATE to select the system-managed tape functions DFSMSrmm provides. The UPDATE operand has 3 subparameters: EXITS, SCRATCH, and COMMAND. You can specify one or more of the subparameters. When DFSMSrmm is running in PROTECT mode, DFSMSrmm ignores the UPDATE operand and performs processing as if you specified EXITS, SCRATCH, and COMMAND. When DFSMSrmm is running in WARNING or RECORD mode, DFSMSrmm does not update TCDB information unless you request the update. You can specify one or more of the values. When you specify a value, DFSMSrmm performs the updates to the TCDB.

EXITS

Specify EXITS when you want DFSMSrmm volume status information to override the OAM volume status during entry processing, and you want to use the DFSMSrmm VNL exit.

SCRATCH

Specify SCRATCH when you want DFSMSrmm to update the volume status in the TCDB during expiration processing. This can be:

• When volumes are returned to scratch status.
• When scratch volumes are changed to master status during expiration volume replacement processing.

COMMAND

Specify COMMAND when you want to use the RMM TSO subcommands or the DFSMSrmm API to update the TCDB. This controls change of status, TDSI and owner information, eject processing, and manual cartridge entry processing.

Default: None, if you are running in MANUAL, RECORD, or WARNING mode, but DFSMSrmm ignores the update options and forces them to UPDATE(EXITS,SCRATCH,COMMAND) when DFSMSrmm is running in PROTECT mode. See the description of the OPTMODE parameter in this topic for more information on how OPMODE relates to and affects System-Managed Tape Library Support.

PURGE

Use PURGE to control how DFSMSrmm affects the TCDB volume records during EJECT processing. The default is PURGE(ASIS) in all operating modes except MANUAL mode. In manual mode, DFSMSrmm provides no support for eject processing.

ASIS

Specify ASIS when you do not want DFSMSrmm to determine the TCDB volume record dispositions at eject time. Specifying ASIS allows the eject requestor or Library defaults to control the TCDB volume record disposition

NO

Specify NO to request that DFSMSrmm prevent TCDB records being deleted.

YES

Specify YES when you want DFSMSrmm to force the TCDB volume records to be purged at eject time.

Default: ASIS.

If you are running multiple z/OS systems and sharing the control data set and journal, specify a unique SYSID for each system.

If you have unshared catalogs, you can specify a list of system IDs using the CATSYSID operand. Select the SYSID values from the list of IDs which include current IDs and previously used IDs.

Default: DFSMSrmm uses the system's SMF identification.

The TPRACF(NONE) option is assumed for any volume serial number containing special characters. To protect tape volumes that use special characters in the volume serial number, use RACF generic TAPEVOL profiles that are outside of DFSMSrmm control.

DFSMSrmm honors all other TPRACF options for volume serial numbers that are alphanumeric including national characters. You can also use RACF generic DATASET profiles to protect data created on tape volume. DFSMSrmm honors the TPRACF setting when running in all modes.

If you set TPRACF(PREDEFINED) or TPRACF(AUTOMATIC), DFSMSrmm ensures that all nonscratch tapes are protected by a discrete RACF TAPEVOL profile by checking that a RACF profile exists whenever a data set is written on a tape. If a profile does not exist, DFSMSrmm creates one. Therefore you do not need to use RACF installation exits to set the JCL PROTECT=YES option or specify PROTECT=YES in your JCL. You can use generic data set profiles for all tape data sets without changes to JCL or installation procedures, if you used the VLPOOL command with the RACF(Y) operand, because DFSMSrmm creates a TVTOC when you use the RACF TAPEDSN option.

For both TPRACF(PREDEFINED) and TPRACF(AUTOMATIC), DFSMSrmm ensures TAPEVOL profiles are always deleted during recycling of scratch volumes, if they exist. When you use a volume for output and there is no RACF protection for the volume when you close the data set, DFSMSrmm creates a profile to protect the volume. The owner of the volume is given RACF access to the profile. If TAPEVOL and TAPEDSN are active, DFSMSrmm creates a TVTOC and adds the first file data set entry.

For TPRACF(CLEANUP), DFSMSrmm ensures that TAPEVOL profiles and discrete tape DATASET profiles are deleted during recycling of scratch volumes and existing TAPEVOL profiles are deleted when volumes are deleted from the DFSMSrmm CDS. When you use this option, DFSMSrmm never creates any RACF profiles for you. This processing is only provided for VLPOOLs with RACF(Y).

TPRACF(CLEANUP) is intended to be used when you are changing how tape data sets are protected. For example, you no longer wish to use TAPEVOL profiles and are enabling the use of DATASET profiles, TPRACF(CLEANUP) can be used for this occasion.

For more information on RACF processing and security options, see Authorizing DFSMSrmm users and ensuring security.

• When you specify TPRACF(NONE), DFSMSrmm does not create RACF tape profiles for any volumes in your installation.
  If you also specify one of these VLPOOL values:

  • VLPOOL RACF(N)

    DFSMSrmm does not create tape profiles or TVTOCS for volumes in the pool that you are defining with the VLPOOL command.

  • VLPOOL RACF(Y)

    DFSMSrmm does not create tape profiles or TVTOCS for volumes in the pool that you are defining with the VLPOOL command.

• When you specify TPRACF(PREDEFINED) or TPRACF(P), DFSMSrmm creates predefined RACF tape profiles for any volumes in your installation.
  If you also specify one of these VLPOOL values:

  • VLPOOL RACF(N)

    DFSMSrmm does not create tape profiles or TVTOCS for volumes in the pool that you are defining with the VLPOOL command.

  • VLPOOL RACF(Y)

    DFSMSrmm creates RACF tape profiles for volumes in the pool that you are defining with the VLPOOL command. The RACF tape profiles are created using RACF options that you have specified and the RACF tape profiles that DFSMSrmm creates when RMM ADDVOLUME, CHANGEVOLUME, or DELETEVOLUME subcommands are issued.

    DFSMSrmm deletes TAPEVOL profiles during recycling of scratch tapes if the profiles exist.

    DFSMSrmm creates a profile when a volume that is used for output does not have a RACF tape profile defined.

    The owner of the volume is given RACF access to the profile.

    If TAPEVOL and TAPEDSN are active, DFSMSrmm creates a TVTOC and adds the first file data set entry.

• When you specify TPRACF(AUTOMATIC) or TPRACF(A), DFSMSrmm creates tape profiles for any volume in your installation.
  If you also specify one of these VLPOOL values:

  • VLPOOL RACF(N)

    DFSMSrmm does not create tape profiles or TVTOCS for volumes in the pool that you are defining with the VLPOOL command.

  • VLPOOL RACF(Y)

    DFSMSrmm creates RACF tape profiles for volumes in the pool. Table 1 shows what DFSMSrmm processing takes place when you specify RACF(Y) for a pool.

  The processing is the same as TPRACF(PREDEFINED) with these exceptions:
  • Scratch tape volumes do not have TAPEVOL profiles created.
  • IF TAPEVOL and TAPEDSN are active, RACF automatically creates TAPEVOL and data set profiles for the ADSP and PROTECT=YES users.
  • TPRACF(AUTOMATIC) also supports the environment where an installation uses RACF installation exits to create TAPEVOL profiles.

Default: TPRACF(NONE)

• EXPIRE(days) — Use the EXPIRE(days) option to set the volume expiration date to the current date plus days for volumes to be purged. Use the EXPIRE(days) parameter when you use the EXPDT retention method, using days as a way to delay expiration of the volume. When using the VRSEL retention method, you can optionally use this operand in combination with vital record specifications that use the UNTILEXPIRED retention type. First, the EXPIRE(days) is applied to set a new volume EXPDT. Next, by running VRSEL you can then optionally extend retention using the extra days retention type. For example, specify EXPIRE(0) when using a VRS with extra days retention or you can use a non-zero EXPIRE(days) value and avoid using an extra days retention VRS.

  days is the number of days for which DFSMSrmm retains the volume before considering it for release. The value is a one to four digit decimal number and is added to today’s date to compute the new expiration date. If the value exceeds the maximum retention period (MAXRETPD) it is reduced to the MAXRETPD value. The default value for days is 0. That is, TVEXTPURGE(EXPIRE) is the same as TVEXTPURGE(EXPIRE(0)).

• NONE — DFSMSrmm take no action for volumes to be purged.
• RELEASE — DFSMSrmm releases volume to be purged according to the release actions set for the volume. You must run expiration processing to return a volume to scratch status.

Default: TVEXTPURGE(RELEASE)

• N — Do not use DFSMSrmm catalog processing. DFSMSrmm does not uncatalog data sets under any circumstances.
• S — Only uncatalog data sets when the volume on which they reside is returned to scratch status.
• Y — always uncatalog data sets. Use DFSMSrmm catalog processing. DFSMSrmm uncatalogs data sets when:
  • A volume is returned to scratch status, DFSMSrmm uncatalogs all the data sets on the volume.
  • The RMM DELETEVOLUME FORCE subcommand is issued for a volume, DFSMSrmm uncatalogs all the data sets on the volume.
  • The RMM CHANGEVOLUME DSNAME subcommand is issued for a volume, DFSMSrmm uncatalogs all the data sets on the volume. If the data set name specified on the RMM CHANGEVOLUME subcommand command matches the data set name on the volume, then DFSMSrmm only uncatalogs subsequent data sets.
  • The RMM DELETEDATASET subcommand is issued for a data set, DFSMSrmm uncatalogs the data set. Also, DFSMSrmm uncatalogs all data sets recorded on the same volume with higher data set sequence numbers.
  • A tape data set is overwritten, DFSMSrmm uncatalogs the data set. Also, all data sets recorded on the same volume with higher data set sequence numbers are uncataloged.

If you set UNCATALOG(S) or UNCATALOG(Y), DFSMSrmm uncatalogs data sets even when DFSMSrmm is running in manual mode, record mode, or warning mode.

You should use the UNCATALOG(N) option during early implementation of DFSMSrmm on your system, when DFSMSrmm is running at the same time as your existing management software.

If you use the EDG_EXIT100 exit to request suppression of data set recording for a volume, you should ensure that any data sets that are cataloged, but not recorded by DFSMSrmm are uncataloged by some other mechanism. To leave nonexistent data sets cataloged could lead to later processing problems.

If you use the UNCATALOG(N) option to prevent DFSMSrmm from uncataloging tape data sets, you should ensure that data sets are uncataloged by some other mechanism. To leave nonexistent data sets cataloged could lead to later processing problems.

Ensure Integrated catalog facility catalogs are shared if catalog control is required or if you specify the UNCATALOG(Y) or (S) operand.

Default: The default is UNCATALOG(N) unless DFSMSrmm is running in protect mode. DFSMSrmm uses the UNCATALOG default of Y when DFSMSrmm is running in protect mode.

Default: VRSCHANGE(VERIFY)

VRSDROP processing is intended to provide limited checking for volumes that are already VRS-retained. It considers how many of these existing VRS-retained volumes are removed from VRS retention. Those volumes dropped from VRS-retention can be set pending release, depending on VRS release options and the volume expiration date. They can also become EXPDT retained and on the next run of EXPROC, they will be considered by EXPDTDROP limit processing.

count can be 0 to 2,147,483,647. %age can be 0 to 100.

The default is VRSDROP(PERCENT(10),INFO).

To aid analysis of the results of the VRSDROP limit checking you can use the contents of the ACTIVITY file. The EDGJACTP sample generates a detailed report and a summary report of data sets dropped from VRS retention showing the reasons why. See the contents of the VRS and VRSS files.

Default: VRSEL(NEW)

Default: VRSJOBNAME(2).

VRSMIN is intended to prevent the unintended scratching of tapes in case of accidental deletion of vital record specifications.

count can be 0 to 2,147,483,647. Specify a count value of 0 to disable this checking.

The default value is 1,FAIL.

VRSRETAIN processing is intended to provide limited checking for volumes containing newly created data sets that have not yet been processed by inventory management VRSEL. The data sets on the volumes have been created since the start of the last completed VRSEL run. It considers how many of these volumes become VRS-retained during the new VRSEL run. Those volumes that become VRS-retained will be considered by the VRSDROP limit checking in future VRSEL runs. Those volumes that are not retained by VRS can be set pending release, depending on VRS release options and the volume expiration date. They can also become EXPDT retained and on the next run of EXPROC, they will be considered by EXPDTDROP limit processing.

countcan be 0 to 2,147,483,647. %age can be 0 to 100.

The default is VRSRETAIN(PERCENT(80),INFO).

To aid analysis of the results of the VRSRETAIN limit checking (when the action is not OFF) you can use the contents of the ACTIVITY file and extended records from the extract file. The EDGJACTP sample generates a detailed report and a summary report of newly assigned volumes showing how they are processed. See the contents of the VRSRETN and VRSRETNS files.