JCL requirements

JOB statement

JOB indicates the start of JCL.

EXEC statement

The EXEC statement must be in the following form:

SUF=suffix

Specifies a single-character control program nucleus identifier of the IMS system to be recovered.

There is no default value.

IMSID=imsid

Specifies a subsystem identifier (one to four characters) of the IMS system to be recovered. This parameter must be specified when the IMS is not an extended recovery facility (XRF) complex or has DBCTL standby configuration. If the IMS has DBCTL standby configuration, both IMSID and RSENAME parameters are required. In other cases, the IMSID parameter is mutually exclusive with the RSENAME parameter.

There is no default value.

RSENAME=rsename

Specifies the Recoverable Service Element (RSE) name (one to eight characters) of the IMS XRF complex to be recovered. This parameter must be specified if the IMS system is an XRF complex or has DBCTL standby configuration. If the IMS has DBCTL standby configuration, both IMSID and RSENAME parameters are required. In other cases, the RSENAME parameter is mutually exclusive with the IMSID parameter.

There is no default value.

IMSPLEX=imsplex

Specifies the IMSplex name of the IMS system that is to be recovered, which must be an IMSplex group name identifier (one to five characters). This parameter must be specified only if both of the following two conditions are met:

• The DBRC=Y parameter is specified or defaulted
• The targeted IMS belongs to an IMSplex group represented by the IMSplex name

Note: When IMS DEDB Fast Recovery is run with these conditions, the automatic RECON loss notification is available.

You cannot give an IMSplex name by the SCI Registration exit routine (DSPSCIX0) that is provided by DBRC. For the details of the SCI Registration exit routine (DSPSCIX0), see IMS Exit Routines.

If a name is specified in the EXEC parameter, the name given in the exit overrides it.

If the targeted IMS does not belong to any IMSplex group, this parameter must not be specified. If it is, the job terminates abnormally with code U3512, according to the return code from DBRC.

There is no default value.

DBRCGRP=dbrcgrp

Specifies the DBRC group ID (one to three characters) of the IMS system that is to be recovered. The DBRC group ID is an identifier that is assigned to a group of DBRC instances that make an access to the same DBRC RECON data set in an IMSplex. This parameter must be specified only if all the following three conditions are met:

• The DBRC=Y parameter is specified or defaulted
• The targeted IMS was running through SCI
• The DBRC group ID is registered to identify the DBRC RECON data set in an IMSplex

The default is DBRCGRP=001 (It is the same as the default value of DBRC.)

The job will end abnormally with code U3512, according to the return code from DBRC as in the following cases:

• The DBRCGRP parameter is specified though the IMS system that is to be recovered belongs to the IMSplex group but does not belong to the DBRC group.
• The DBRCGRP parameter is not specified though the IMS system that is to be recovered belongs to the DBRC group.

If the DBRCGRP parameter is specified, you must also specify the IMSPLEX parameter. If you specify DBRC=Y or if you omit the DBRC parameter, and you specify the DBRCGRP parameter but do not specify the IMSPLEX parameter, message DFR3001E showing RSN=1B is issued, and the job ends with a return code of 16.

If the IMSPLEX parameter is specified, either of the following cases is possible:

• DBRC group ID is given by the SCI Registration exit routine (DSPSCIX0) provided by DBRC.
• The value specified by the DBRCGRP parameter can be overridden by the SCI Registration exit routine (DSPSCIX0) provided by DBRC. For the details of the SCI Registration exit routine (DSPSCIX0), see IMS Exit Routines.

DBRC=

Specifies whether to use the DBRC RECON data set to determine the last-used OLDS. The default is DBRC=Y.

Use DBRC=N only when all DBRC RECON data sets are unusable or when IMS DEDB Fast Recovery terminates abnormally because of a DBRC function failure. In either case, you must recover the DBRC RECON data sets, even if they are usable, after running IMS DEDB Fast Recovery and before the IMS cold start.

AUTO=

Specifies whether the automatic determination function is to be used for determining the last OLDS (AUTO=Y) or whether the operator will determine it manually by replying to the WTOR message (AUTO=N). This parameter is valid only when DBRC=N. The default is AUTO=Y.

LCHKPT=

Specifies whether to start recovery log processing from the Fast Path checkpoint (LCHKPT=Y) or from the checkpoint that is required for backout of DL/I databases (LCHKPT=N). The default is LCHKPT=Y.

If there are no DL/I database definitions for the IMS to be recovered, log processing starts from the Fast Path checkpoint regardless of this parameter.

Specify LCHKPT=N if the IMS to be recovered contains DL/I database definitions and the Backout Required PSB list, which lists databases that need backout, is required. If you specify LCHKPT=Y to reduce log processing time when there are DL/I database definitions, some of the databases that need backout might be left off the list. Even in this case, however, all the PSBs that need backout are listed.

CIDUMP=

Specifies whether to write a dump of the recovered CIs into the RCIDUMP data set when the data set is specified in the execution JCL.

CIDUMP=Y

Writes a dump of the recovered CIs. Even if the output operation is terminated abnormally because, for example, the output data set does not have enough space, the recovery process continues without writing the dump.

CIDUMP=A

Writes a dump of the recovered CIs. If the output is terminated abnormally because, for example, the output data set does not have enough space, IMS DEDB Fast Recovery also terminates abnormally.

CIDUMP=N

Does not write a dump of the recovered CIs.

The default is CIDUMP=Y. If CIDUMP=Y or CIDUMP=A is specified but the RCIDUMP data set is not in the JCL, the job terminates abnormally with an abend code of U3007.

T=

Specifies whether Universal Coordinated Time (UTC) is to be shown along with local time in the report header.

T=L

Only local time appears in each report header.

T=U

UTC appears along with local time in each report header.

The default is T=L.

CFCONF=

Specifies whether to suspend the IMS DEDB Fast Recovery execution when at least one connection to a CF structure, which belongs to a block-level data-sharing VSO area, fails. Specify CFCONF=Y to suspend the execution with a WTOR message (DFR3601A) when the connection to a CF structure fails. Specify CFCONF=N to allow its execution.

This parameter is effective only if the failed IMS system is using block-level data-sharing VSO areas.

The default is CFCONF=N.

CFRET=

Specifies the response to a CF connectivity error. CFRET=Y keeps a block-level data-sharing VSO area available on DBRC even if the recovery fails because of a CF connectivity error. CFRET=N makes the area unavailable with this kind of error.

This parameter is effective only if the failed IMS system is using block-level data-sharing VSO areas. The default is CFRET=Y.

The CFRET=Y execution generates Area Recovery Retry control statements into the RETLIST DD data set. The Area Recovery Retry control statements contain the names of the areas that were not made unavailable and were not recovered, and they are the input of AREASLCT DD statement for the next execution. You can selectively run IMS DEDB Fast Recovery again for the incomplete areas by specifying the area names in the AREASLCT DD statement. If the failed IMS system is using block-level data-sharing VSO areas and CFRET=Y is specified for IMS DEDB Fast Recovery execution, you must specify the RETLIST DD statement for output.

Note: If there is no other connector for a block-level data-sharing VSO area and the recovery is completed successfully with IMS DEDB Fast Recovery, the structures related to the area are deleted from CF. If IMS DEDB Fast Recovery runs for the area again, IMS DEDB Fast Recovery recognizes the deleted structures as a connectivity error, because there is no structure to be recovered. The IMS DEDB Fast Recovery execution with CFRET=N and DBRC=Y specified makes the area unavailable on DBRC. Be careful for the second or later execution. Especially if IMS DEDB Fast Recovery terminates abnormally after the recovery of some block-level data-sharing VSO areas, specify CFRET=Y in the next execution to keep the areas available.

MSDB=

Specifies whether to automatically generate the input JCL for recovering the main storage database (MSDB) for the MSDB Dump Recovery utility. To generate it, specify MSDB=Y or MSDB=O. For the default, MSDB=N, the JCL is not generated. The specifications MSDB=Y and MSDB=O are different in how the MSDBCPn DD statements will be specified in the JCL for the MSDB Dump Recovery utility (n represents from 1 to 4 when IMS system is an XRF complex, from 1 to 2 in other cases). The following lists describe the difference between MSDB=Y and MSDB=O.

MSDB=Y

The MSDBCPn DD name and data set name pair specified for the JCL for IMS DEDB Fast Recovery will be used without change for the MSDBCPn DD statement.

MSDB=O

Among the MSDB checkpoint data sets that contain 1) checkpoint IDs that are older than (or the same as) the Fast Path checkpoint ID of the relevant IMS and 2) valid data, IMS DEDB Fast Recovery selects the latest one. The name of the data set selected is used for all MSDBCPn DD statements.

Note: The MSDB Dump Recovery utility selects the older MSDB checkpoint data set (of the later pair in an XRF complex) to recover an MSDB. If you specify MSDB=O, IMS DEDB Fast Recovery specifies the later data set for all MSDBCPn DD statements in the output JCL for MSDB recovery. This shortens the recovery time, because the MSDB Dump Recovery utility always selects the later one.

MSDB=Y or MSDB=O can be specified with DBRC=Y or DBRC=N, AUTO=Y. If DBRC=N, AUTO=Y is specified, the following considerations apply:

• If the automatic determination function for the use of OLDS (AUTO=Y) does not work because the recovery checkpoint for DEDB is in an SLDS, the JCL creation function is deactivated when IMS DEDB Fast Recovery is terminated.
• If the recovery checkpoint for DEDB is in an OLDS but an SLDS is required for the MSDB Recovery, the JCL is generated without SLDS specification and warning message DFR3715W is issued.

OLC=

Specifies the scope of the online change of the IMS system to be recovered.

OLC=L

The online change applies locally to the IMS system, and IMS DEDB Fast Recovery decides the active ACBLIB data set on the basis of the record in the MODSTAT or the MODSTAT2 data set.

OLC=G

The online change, including the ACB library member online change function, applies globally to the IMS system—that is, the IMS system uses online change libraries consistently with other IMS systems across the IMSplex to which it belongs. On the basis of the record that is in the OLCSTAT data set, IMS DEDB Fast Recovery decides the active ACBLIB data set and then checks the availability of the ACB library members.

The default is OLC=L.

CSLG=xxx

This parameter is the same as the CSLG= parameter that is specified in the EXEC parameter or in the DFSPBxxx IMS.PROCLIB member of the IMS system that is to be recovered. Specify 1- to 3- character suffix of the DFSCGxxx IMS.PROCLIB member that contains parameters related to the Common Service Layer (CSL).

If OLC=G is specified, this parameter, the DFSDF parameter, or both must also be specified.

If OLC=L is specified, or if the OLC= parameter is omitted, this parameter will be ignored.

There is no default value.

DFSDF=yyy

This parameter is the same as the DFSDF= parameter that is specified in the EXEC parameter or in the DFSPBxxx IMS.PROCLIB member of the IMS system that is to be recovered. Specify a 1- to 3- character suffix of the member name DFSDFyyy IMS.PROCLIB where the Common Service Layer (CSL) is specified as a section.

If OLC=G is specified, this parameter, the CSLG parameter, or both must also be specified.

If OLC=L is specified, or if the OLC= parameter is omitted, this parameter will be ignored.

There is no default value.

ACBLIB=

Specifies whether to enable the automatic determination function. When this function is enabled, IMS DEDB Fast Recovery searches for active application control blocks (ACBs) and uses the detected active ACBs.

ACBLIB=C

IMS DEDB Fast Recovery detects active application control blocks automatically.

If the IMS management of ACBs is enabled, IMS DEDB Fast Recovery uses the active ACBs in the IMS directory.

If active ACBs are managed in ACBLIB data sets, IMS DEDB Fast Recovery uses the active ACBs found in the ACB library that the IMSACBA DD or IMSACBB DD statement specifies.

ACBLIB=F

IMS DEDB Fast Recovery uses the active ACBs found in the ACB library that the IMSACBA DD statement specifies.

Use ACBLIB=F only when the IMS management of ACBs is enabled and IMS directory data sets cannot be used.

The default is ACBLIB=C.

DD statements

STEPLIB DD

Specifies the library containing the IMS DEDB Fast Recovery load modules. Because IMS DEDB Fast Recovery runs as an MVS authorized program, this library must be registered as an APF authorized library. For an explanation of how to register, see the MVS Initialization and Tuning Reference.

The IMS.SDFSRESL (the library containing IMS load modules) of the IMS to be recovered must also be specified in the STEPLIB statement. IMS DEDB Fast Recovery uses the IMS module (DFSVC000) in the data set:

• By checking the version and release in the module, IMS DEDB Fast Recovery verifies and determines the version and release of the IMS being used.
• To display messages to the system console of the IMS to be recovered, IMS DEDB Fast Recovery uses the routing code and the descriptor code of the system console contained in the module.

IMS DEDB Fast Recovery also uses IMS DBRC modules.

A user-written logger exit routine (DFSFLGX0) is invoked during the execution of IMS DEDB Fast Recovery if the exit routine is present. DFSFLGX0 is called once with an initialization call, once with a write call for each log buffer of data that is written, and once with a termination call.

If you want to use automatic RECON loss notification, specify the data set that contains the dynamic allocation parameter lists for the DBRC RECON data sets in the IMSDALIB concatenation or the STEPLIB concatenation.

To use IMS-managed ACBs in an IMS managed ACB environment, specify the data set that contains the IMS Catalog Definition exit routine (DFS3CDX0) to the STEPLIB concatenation. To allocate the IMS directory data sets dynamically, specify the data set that contains the DFSMDA member with the TYPE=CATDSHLQ statement.

PROCLIB DD

When OLC=G is specified in the EXEC parameter, specify the IMS.PROCLIB data set which includes the DFSCGxxx member, DFSDFyyy member (xxx is the value specified in the CSLG= EXEC parameter, and yyy is the value specified in the DFSDF= EXEC parameter), or both. IMS DEDB Fast Recovery dynamically allocates the OLCSTAT data set by using the name that is specified in the OLCSTAT= parameter in the member. You do not need to specify the OLCSTAT DD statement in the JCL.

If the OLCSTAT= parameter is specified in both DFSCGxxx and DFSDFyyy, one in DFSCGxxx is used.

To use IMS-managed ACBs in an IMS managed ACB environment, specify the IMS.PROCLIB data set that includes the DFSDFxxx member (xxx is the value specified for the DFSDF= parameter of the EXEC statement). This member must contain the ACBMGMT=CATALOG parameter in the CATALOG imsid section or the CATALOG section. If both sections are defined in the member, IMS DEDB Fast Recovery uses definitions in the CATALOG imsid section.

IMSDALIB DD

Specifies the input data set containing the dynamic allocation parameter lists for log data sets.

If you want to use automatic RECON loss notification, specify the input data set that contains the dynamic allocation parameter lists for the DBRC RECON data sets in the IMSDALIB concatenation or the STEPLIB concatenation.

IMSACBA DD

IMSACBB DD

Specifies the ACB libraries of the IMS system to be recovered.

IMS DEDB Fast Recovery uses these libraries to initialize DEDB basic control blocks (DMCB and DMAC).

If the IMS management of ACBs is enabled, this parameter is ignored.

Note: If some of the databases specified in IMS definition are not actually used in the IMS subsystem, the NOTINIT status for those databases appears in a Database Status list. Even if ACBLIBs specified in the execution JCL are different from the ones used by the failed IMS, IMS DEDB Fast Recovery tries to redo DEDBs (apply the uncommitted update to the CI and write the updated CI to DASD) by using the libraries specified in the JCL when the libraries contain related necessary members. Because IMS DEDB Fast Recovery cannot detect this error, you must be careful in specifying the libraries to be used.

IMSBSDS DD

Specifies the IMS bootstrap data set. Use this DD statement to run IMS DEDB Fast Recovery in an IMS-managed ACBs environment without accessing DBRC RECON data sets.

This DD statement is useful when the IMS management of ACBs is enabled but the IMS catalog database is not registered to DBRC RECON data sets.

Usage notes:

• The final qualifier of the IMS bootstrap data set must be BSDS.
• If this DD statement is supplied, the DFSMDA member with the TYPE=CATDSHLQ statement is not used.
• This DD statement is mutually exclusive with ACBLIB=F in the EXEC parameter.

MODSTAT DD

MODSTAT2 DD

When OLC=L is specified or defaulted in the EXEC parameter, specify the MODSTAT data set of the IMS system to be recovered. The MODSTAT2 DD statement is necessary only if the IMS system is an XRF complex.

IMS DEDB Fast Recovery uses this data set to determine the name of the ACB library that was in use when IMS failed. For the IMS XRF complex, two data sets are necessary for the active and alternate IMS systems. For the recovery of warm standby IMS system in a DBCTL environment, specify the MODSTAT data set of the active IMS system, and not of the standby IMS system.

If OLC=G is specified in the EXEC parameter, these data sets are not required and will be ignored even if they are specified.

OLCSTAT DD

When OLC=G is specified in the EXEC parameter, specify the OLCSTAT data set of the IMS system that is to be recovered. Because IMS DEDB Fast Recovery dynamically allocates the OLCSTAT data set that is specified in the OLCSTAT= parameter in the DFSCGxxx IMS.PROCLIB member or the DFSDFyyy IMS.PROCLIB member (xxx is the value specified in the CSLG= EXEC parameter, and yyy is the value specified in the DFSDF= EXEC parameter), you do not need to specify this DD statement.

IMS DEDB Fast Recovery uses this data set for the following purposes:

• To determine the name of the ACB library that was in use when the IMS system failed.
• To check the status of the ACBLIB member OLC when the IMS system failed. If, however, the OLCSTAT data set is formatted by an IMS release earlier than IMS 10.1, the status of the ACBLIB member OLC is not checked.

DFSWADSn DD

Specifies the write-ahead data sets (WADSs) (where n is 0 - 9). You can specify all WADSs used during IMS online execution, but you must include the WADS that was in use when IMS failed. To avoid executing IMS DEDB Fast Recovery against the active IMS system, you must specify DISP=OLD in the DD statement.

WADSs that are defined as large format data sets and WADSs that are allocated in the extended addressing space (EAS) of an extended address volume (EAV) can be specified.

IMS DEDB Fast Recovery uses WADS to build log blocks not yet written in the OLDS that was in use when IMS failed, writes them at the end of that OLDS, and closes it.

IMS DEDB Fast Recovery can allocate the data set dynamically without the DD statement. See How log data sets are allocated.

DFSOLPnn DD

DFSOLSnn DD

Specifies the online log data sets (OLDSs) that IMS DEDB Fast Recovery is to read to determine the recovery start checkpoint. nn is 00 - 99. DFSOLPnn indicates the primary OLDS, and DFSOLSnn indicates the secondary OLDS. The OLDS ddnames are the same as the ddnames used at IMS online execution.

OLDSs that are defined as large format data sets, OLDSs that are defined as DFSMS striped extended-format data sets, and OLDSs that are allocated in the extended addressing space (EAS) of an extended address volume (EAV) can be specified.

If the transaction that did not reach a sync point at IMS failure had already created DL/I database update log records or Fast Path log records, IMS DEDB Fast Recovery creates and writes a sync point failure log record into the OLDS to invalidate those log records.

In a two-phase commit process environment such as DBCTL, for a transaction whose status at IMS failure was indoubt, IMS DEDB Fast Recovery creates a sync point commit or abort log record and writes it into the OLDS to resolve the in-doubt status. Resync control statements determine whether IMS DEDB Fast Recovery writes a sync point commit log record or a sync point abort log record. See Resync control statement.

IMS DEDB Fast Recovery can allocate the data set dynamically without the DD statement. See How log data sets are allocated.

IMSLOGR DD

IMSLOGR2 DD

ssssssss DD

Specifies the system log data sets (SLDSs) that IMS DEDB Fast Recovery reads to determine the recovery start checkpoint.

SLDSs that are defined as large format data sets and SLDSs that are defined as DFSMS striped extended-format data sets can be specified.

If DBRC=Y, specify IMSLOGR (primary SLDS) and IMSLOGR2 (secondary SLDS). If multiple SLDSs are required, concatenate the data sets in the order they are used (the latest-used data set comes last) on each IMSLOGR and IMSLOGR2 DD statement.

If DBRC=N, specify DD statements with any ddnames (ssssssss DD) to allocate the data sets. DD statements must be eight characters.

IMS DEDB Fast Recovery can allocate the data set dynamically without the DD statement. See How log data sets are allocated.

RECON1 DD

RECON2 DD

RECON3 DD

Specifies the DBRC RECON data set. These statements are required when DBRC=Y.

The DBRC RECON data set in parallel access mode can be specified.

The DBRC RECON data set that is allocated in the extended addressing space (EAS) of an extended address volume (EAV) can be specified.

If you want to use automatic RECON loss notification, specify the input data set that contains the dynamic allocation parameter lists for the DBRC RECON data sets in the IMSDALIB concatenation or the STEPLIB concatenation. Do not specify the RECONn DD statement (n=1 to 3).

IMS DEDB Fast Recovery uses this data set to determine the names of the IMS log data sets (OLDS or SLDS) to be used during IMS recovery. This data set is not necessary when DBRC=N.

IMS DEDB Fast Recovery updates the DBRC RECON data set for the following conditions:

• The DEDB area data set that could not be recovered because of an I/O error during the IMS DEDB Fast Recovery execution is registered as unavailable in the RECON data set. The nonrecoverable DEDB area data set that could not be recovered because of the uncommitted database changes is registered as unavailable in the RECON data set.
• If the DEDB Area Data Set Create utility was in process when IMS failed, IMS DEDB Fast Recovery updates the RECON data set to reset its in-process status to enable re-execution.
• If the DEDB area or the area data set was being stopped when IMS failed, IMS DEDB Fast Recovery continues processing and registers the area or the area data set as stopped in the RECON data set.
• In the IMS XRF complex, if the I/O toleration Extended Error Queue Element (EEQE) and the I/O toleration buffer existed when IMS failed, IMS DEDB Fast Recovery uses the buffer to recover the VSAM control interval (CI) in the DEDB area and then deletes the I/O toleration EEQE from the RECON data set.
• In a two-phase commit process environment such as DBCTL, if a transaction has an in-doubt status at IMS failure, and in-doubt EEQEs have been already created, IMS DEDB Fast Recovery deletes the following EEQEs:
  • For DEDB areas, it deletes all in-doubt EEQEs.
  • For a DL/I database, it deletes only the in-doubt EEQEs related to the transaction that IMS DEDB Fast Recovery commits.
• When IMS DEDB Fast Recovery closes the OLDS that was in use when IMS failed, it updates the OLDS information in the RECON data set.

DEDBSNAP DD

Specifies the output data set for a snap dump of the damaged DEDB VSAM CIs that IMS DEDB Fast Recovery reads for recovery. Snap dumps are generated only when this statement is specified.

RCIDUMP DD

Specifies the output data set for an image of the DEDB VSAM CIs updated by IMS DEDB Fast Recovery. It must be a sequential data set with a blocked variable record format. For the way to specify parameters for the output, see the explanation of the CIDUMP= parameter. If a BLKSIZE parameter on the RCIDUMP DD statement is omitted from the execution JCL, IMS DEDB Fast Recovery sets it to 32,760 bytes. If BLKSIZE is specified, the specified value is used. IMS DEDB Fast Recovery sets the maximum value of a variable logical record length to the value that is 4 bytes smaller than the block size. The value you specify for BLKSIZE in the RCIDUMP data set in the JCL must be equal to or greater than the maximum length of the CI expected to be written plus 40, because you must consider the maximum length of the recovered CI and the length of the control information written with the CI. For the format of the RCIDUMP data set, see RCIDUMP data set format. You can print an image of the output VSAM CI as a snap dump using the Recovered CI Print program. See Recovered CI Print program.

RCISUMM DD

Specifies the output data set for the number of the DEDB VSAM CIs recovered by IMS DEDB Fast Recovery. It lists the number of the CIs recovered in each area and the total number recovered.

RCIBMAP DD

Specifies the output data set for the bitmap of the DEDB VSAM CIs recovered by IMS DEDB Fast Recovery. It is written for each recovery buffer in each area.

SYSPRINT DD

Specifies the output data set for the recovery status lists that IMS DEDB Fast Recovery generates. This statement is required.

RSYLIST DD

Specifies the output data set for the skeleton of Resync control statements. If IMS DEDB Fast Recovery runs in a two-phase commit process environment such as DBCTL, specify the DD statement. Otherwise, the DD statement is not necessary, because there are no cases of resync.

IMS DEDB Fast Recovery generates the skeleton only in the following case:

• Some transactions have an in-doubt status, and
• No Resync control statement is specified in RESYNCTL DD, or the specified Resync control statements cannot resolve all in-doubt status transactions.

RESYNCTL DD

Specifies the input data set for Resync control statements that determine whether the in-doubt transaction is committed or aborted in the sync point process. It is mandatory if there are any in-doubt transactions. It is necessary only in the two-phase commit process environment. For more information, see Resync control statement.

RETLIST DD

Specifies the output data set to generate Area Recovery Retry control statements, which become the input for the data set specified by the AREASLCT DD statement for the next execution. This is required if IMS DEDB Fast Recovery runs with CFRET=Y specified. For the recovery of the block-level data-sharing VSO area, if an error in connectivity to a CF structure occurs, IMS DEDB Fast Recovery gives up the recovery for the area without marking the area as needing recovery. Area Recovery Retry control statements generated in the data set contain the names of the areas. For Area Recovery Retry control statements, see Area Recovery Retry control statement.

AREASLCT DD

Specifies the input data set containing Area Recovery Retry control statements. It is used to select the areas to be recovered. The selection can be applied only for DEDB areas. For Area Recovery Retry control statements, see Area Recovery Retry control statement.

MSDBIN DD

Specifies the input data set that contains the skeleton JCL for the MSDB Dump Recovery utility. This data set is required when you specify MSDB=Y or MSDB=O. For the details of the skeleton JCL, read Skeleton JCL for generating MSDB Dump Recovery utility JCL.

MSDBOUT1 DD

MSDBOUT2 DD

Specifies the output data sets to which the JCL for the MSDB Dump Recovery utility will be written. One of these is required when you specify MSDB=Y or MSDB=O. You can specify either or both of them. If you specify both, the same JCL is written in both data sets.

//MSDBOUTn DD SYSOUT=(A,INTRDR)

By specifying MSDBOUTn (n=1 or 2) as is done here, you can submit the MSDB Dump Recovery utility job. By specifying MSDBOUTn as a data set, you can have the output JCL written in it. In the IMS DEDB Fast Recovery internal process, the DCB specification is RECFM=FB, LRECL=80.

There is no log data set contention between IMS DEDB Fast Recovery and the MSDB Dump Recovery utility, because IMS DEDB Fast Recovery writes the JCL after all log data set processes are over.

MSDBCP1 DD

MSDBCP2 DD

MSDBCP3 DD

MSDBCP4 DD

Specifies the input data sets for MSDB checkpoint. These are required when you specify MSDB=Y or MSDB=O. The DD statements point to the MSDB checkpoint data sets. If the IMS system to recover is not an XRF complex, MSDBCP3 and MSDBCP4 are not required.

DBRCIN DD

Specifies the input data set that contains the skeleton JCL for the DBRC commands to notify backout elements. For more information, see Skeleton JCL for generating DBRC commands.

DBRCOUTL DD

DBRCOUTV DD

DBRCOUTN DD

Specifies the output data sets to which the DBRC commands with the skeleton JCL will be written. DBRCOUTL is required when you specify a DBRCIN DD statement. DBRCOUTV and DBRCOUTN are optional. The difference is as follows:

DBRCOUTL

All commands are written in this data set, though some might be incomplete because of lack of parameters.

DBRCOUTV

Of the commands written in DBRCOUTL, only the complete commands are written in this data set.

DBRCOUTN

Of the commands written in DBRCOUTL, only the incomplete commands are written in this data set.

Note: The time stamp of the X'5607' log record is in Universal Coordinated Time (UTC), and contains the signed packed-decimal local time zone offset for when the time stamp was created. The UOR time for DBRC commands is converted to local time and local time zone offset from the time stamp of X'5607' log record.

For more information, see Skeleton JCL for generating DBRC commands.