DSNJU003 (change log inventory)

Environment

Execute the change log inventory utility only as a batch job when Db2 is not running. Changing a BSDS for a data-sharing member by using DSNJU003 might cause a log read request from another data-sharing member to fail. The failure occurs only if the second member tries to access the changed BSDS before the first member is started.

Authorization required

The authorization ID of the DSNJU003 job must have the requisite RACF® authorization.

Required and optional data sets

DSNJU003 recognizes DD statements with the following DD names:

SYSUT1

Specifies and allocates the bootstrap data set. This statement is required.

SYSUT2

Specifies and allocates a second copy of the bootstrap data set. This statement is required if you use dual BSDSs.

Dual BSDSs and DSNJU003: With each execution of DSNJU003, the BSDS timestamp field is updated with the current system time. If you run DSNJU003 separately for each copy of a dual copy BSDS, the timestamp fields are not synchronized, and Db2 fails at startup. If you change the contents of the BSDS copy by running DSNJU003, Db2 issues error message DSNJ122I. Therefore, if you use DSNJU003 to update dual copy BSDSs, update both BSDSs within a single execution of DSNJU003.

SYSPRINT

Specifies a data set for print output. This statement is required. The logical record length (LRECL) is 125.

SYSIN

Specifies the input data set for control statements. This statement is required. The logical record length (LRECL) is 80.

Running DSNJU003

Execute the utility with the following statement, which can be included only in a batch job:

//EXEC   PGM=DSNJU003

DSNJU003 utility control statements

DSNJU003 uses multiple statements that you submit in separate jobs. The statements are:

• NEWLOG
• DELETE
• CRESTART
• NEWCAT
• DDF
• CHECKPT
• HIGHRBA
• DELMBR
• RSTMBR

Each change log inventory control statement can be entered in a single input record or multiple input records. If the length of a control statement is greater than 80 bytes, it must be entered in multiple records. The following example shows a change log inventory statement that spans multiple records:

SYSIN DD *
DDF LOCATION=USIBMSTODB22,PORT=33275,RESPORT=34522,ALIAS=MYALIAS1:33720,
MYALIAS2:33722

DSNJU0003 option descriptions

NEWLOG

Declares one of the following data sets:

• A VSAM data set that is available for use as an active log data set.

  Use only the keywords DSNAME=, COPY1, and COPY2.

• An active log data set that is replacing one that encountered an I/O error.

  Use only the keywords DSNAME=, COPY1, COPY2, STARTRBA=, and ENDRBA=.

• An archive log data set volume.

  Use only the keywords DSNAME= ,COPY1VOL=, COPY2VOL=, STARTRBA=, ENDRBA=, UNIT=, CATALOG=, STRTLRSN=, and ENDLRSN=.

  If you create an archive log data set and add it to the BSDS with this utility, you can specify a name that Db2 might also generate. Db2 generates archive log data set names of the form DSNCAT.ARCHLOGx.Annnnnnn where:

  • DSNCAT and ARCHLOG are parts of the data set prefix that you specified on installation panels DSNTIPA2 and DSNTIPH.
  • x is 1 for the first copy of the logs, and 2 is for the second copy.
  • Annnnnnn represents the series of low-level qualifiers that Db2 generates for archive log data set names, beginning with A0000001, and incrementing to A0000002, A0000003, and so forth.

  For data sharing, the naming convention is DSNCAT.ARCHLOG1 or DSNCAT.DSN1.ARCLG1.

  If you do specify a name by using the same naming convention as Db2, you receive a dynamic allocation error when Db2 generates that name. The error message, DSNJ103I, is issued once. Db2 then increments the low-level qualifier to generate the next data set name in the series and offloads to it the next time Db2 archives. (The active log that previously was not offloaded is offloaded to this data set.)

  The newly defined active logs cannot specify a start and end LRSN. When Db2 starts, it reads the new active log data sets with an RBA range to determine the LRSN range, and updates the start and end LRSN in the BSDS for the new log data sets. The start and end LRSN for new active logs that contain active log data are read at Db2 start-up time from the new active log data sets that are specified in the change log inventory NEWLOG statements. For new archive logs that are defined with change log inventory, the user must specify the start and end RBAs. For data sharing, the user must also specify the start and end LRSNs. Db2 startup does not attempt to find these values from the new archive log data sets.

DSNAME= data-set-name

Specifies a log data set.

data-set-name can be up to 44 characters long.

COPY1

Makes the data set an active log copy-1 data set.

COPY2

Makes the data set an active log copy-2 data set.

STARTRBA= startrba

Identifies a hexadecimal number of up to 20 characters. If you use fewer than 20 characters, leading zeros are added. startrba must end with '000'; otherwise Db2 returns a DSNJ4381 error message. You can obtain the RBA from messages or by printing the log map.

On the NEWLOG statement, startrba gives the log RBA of the beginning of the replacement active log data set or the archive log data set volume that is specified by DSNAME.

On the CRESTART statement, startrba is the earliest RBA of the log that is to be used during restart. If you omit STARTRBA, Db2 determines the beginning of the log range.

On the CHECKPT statement, startrba indicates the start checkpoint log record.

STARTRBA is required when STARTIME is specified.

On the HIGHRBA statement, startrba denotes the log RBA of the highest-written log record in the active log data sets.

ENDRBA= endrba

endrba is a hexadecimal number of up to 20 characters. If you use fewer than 20 characters, leading zeros are added. endrba must end with '000' or Db2 returns a DSNJ4381 error message.

On the NEWLOG statement, endrba gives the log RBA (relative byte address within the log) of the end of the replacement active log data set or the archive log data set volume that is specified by DSNAME.

On the CRESTART statement, endrba is the last RBA of the log that is to be used during restart, and it is also the starting RBA of the next active log that is written after restart. Any log information in the bootstrap data set, the active logs, and the archive logs with an RBA that is greater than endrba is discarded. If you omit ENDRBA, Db2 determines the end of the log range.

The value of ENDRBA must be a multiple of 4096. (The hexadecimal value must end in 000.) Also, the value must be greater than or equal to the value of STARTRBA. If STARTRBA and ENDRBA are equal, the next restart is a cold start; that is, no log records are processed during restart. The specified RBA becomes the beginning RBA of the new log.

On the CHECKPT statement, endrba indicates the end checkpoint log record that corresponds to the start checkpoint log record.

COPY1VOL= vol-id

vol-id is the volume serial of the copy-1 archive log data set that is specified after DSNAME.

COPY2VOL=vol-id

vol-id is the volume serial of the copy-2 archive log data set that is specified after DSNAME.

UNIT=unit-id

unit-id is the device type of the archive log data set that is named after DSNAME.

CATALOG

Indicates whether the archive log data set is to be cataloged.

NO

Indicates that the archive log data set is not to be cataloged. All subsequent allocations of the data set are made using the unit and volume information that is specified on the statement.

YES

Indicates that the archive log data set is to be cataloged. All subsequent allocations of the data set are made using the catalog.

Db2 requires that all archive log data sets on disk be cataloged. Select CATALOG=YES if the archive log data set is on disk.

STRTLRSN= startlrsn

On the NEWLOG statement, startlrsn identifies the LRSN in the log record header of the first complete log record on the new archive data set. startlrsn is a hexadecimal number of up to 20 characters. If you use fewer than 20 characters, leading zeros are added. In a data sharing environment, run the print log map utility to find an archive log data set and start and end RBAs and LRSNs.

ENDLRSN=endlrsn

endlrsn is a hexadecimal number of up to 20 characters. If you use fewer than 20 characters, leading zeros are added. In a data sharing environment, run the print log map utility to find an archive log data set and start and end RBAs and LRSNs.

For the NEWLOG and CHECKPT statements, the ENDLRSN option is valid only in a data sharing environment. For the CRESTART statement, the ENDLRSN option is valid in both data sharing and non-data sharing environments. This option cannot be specified with STARTRBA or ENDRBA.

On the NEWLOG statement, endlrsn is the LRSN in the log record header of the last log record on the new archive data set.

On the CRESTART statement, in a data sharing environment, endlrsn is an LRSN value that is to be used as the log truncation point. A valid log truncation point is any LRSN value for which there exists a log record with an LRSN that is greater than or equal to the specified LRSN value. Any log information in the bootstrap data set, the active logs, and the archive logs with an LRSN greater than endlrsn is discarded. If you omit ENDLRSN, Db2 determines the end of the log range.

In a non-data sharing environment, endlrsn is the RBA value that matches the start of the last log record that is to be used during restart. Any log information in the bootstrap data set, the active logs, and the archive logs with an RBA that is greater than endlrsn is discarded. If the endlrsn RBA value does not match the start of a log record, Db2 restart fails. If you omit ENDLRSN, Db2 determines the end of the log range.

On the CHECKPT statement, endlrsn is the LRSN of the end checkpoint log record.

STARTIME=startime

Enables you to record the start time of the RBA in the BSDS. This field is optional.

startime specifies the start time in the following timestamp format:

yyyydddhhmmsst

In this format:

yyyy

Indicates the year (1989-2099).

ddd

Indicates the day of the year (0-365; 366 in leap years).

hh

Indicates the hour (0-23).

mm

Indicates the minutes (0-59).

ss

Indicates the seconds (0-59).

t

Indicates tenths of a second.

If fewer than 14 digits are specified for the STARTIME or ENDTIME parameter, trailing zeros are added.

If STARTIME is specified, the ENDTIME, STARTRBA, and ENDRBA options must also be specified.

ENDTIME= endtime

Enables you to record the end time of the RBA in the BSDS. This field is optional.

endtime specifies the end time in the same timestamp format as the STARTIME option. The ENDTIME value must be greater than or equal to the value of STARTIME.

DELETE

Deletes either CCSID information or log data set information from the bootstrap data sets. To delete CCSID information, specify the CCSIDS option. To delete all information for a specified log data set or volume, specify the DSNAME option.

CCSIDS

Deletes CCSID information from the BSDS. CCSID information is stored in the BSDS to ensure that you do not accidentally change the CCSID values.

Use this option under the direction of IBM® Support when the CCSID information in the BSDS is incorrect. After you run a DSNJU003 job with the DELETE CCSIDS option, the CCSID values from the application defaults load module are recorded in the BSDS the next time Db2 is started.

CRESTART

Controls the next restart of Db2, either by creating a new conditional restart control record or by canceling the one that is currently active.

CREATE

Creates a new conditional restart control record. When the new record is created, the previous control record becomes inactive.

SYSPITR=log-truncation-point

Specifies the log RBA (non-data sharing system) or the log LRSN (data sharing system) that represents the log truncation point for the point-in-time for system recovery. Before you run the RESTORE SYSTEM utility to recover system data, you must use the SYSPITR option of DSNJU003. This option enables you to create a conditional restart control record to truncate the logs for system point-in-time recovery. You can also specify a value of FFFFFFFFFFFFFFFFFFFF to cause a point-in-time recovery to occur without log truncation.

log-truncation-point specifies the log RBA, log LRSN, or log FFFFFFFFFFFFFFFFFFFF. In a non-data sharing environment, log-truncation point is the RBA value that matches the start of the last log record that is to be used during restart. If the RBA value does not match the start of a log record, Db2 restart fails. In a data sharing environment, log-truncation point is an LRSN value that is a valid log truncation point. A valid log truncation point is any LRSN value for which there exists a log record with an LRSN that is greater than or equal to the specified LRSN value. Use the same LRSN value for all members of the data sharing group that require log truncation.

You cannot specify any other option with CREATE, SYSPITR.

ENDTIME= log-truncation-timestamp

Specifies an end time value that is to be used as the log truncation point. A valid truncation point is any UTC timestamp for which there exists a log record with a timestamp that is greater than or equal to the specified timestamp value. Any log information in the bootstrap data set, the active logs, and the archive logs with a timestamp greater than the ENDTIME is discarded. If you do not specify ENDTIME, Db2 determines the end of the log range.

You cannot specify any other option with CREATE, ENDTIME.

SYSPITRT= log-truncation-timestamp

Specifies the timestamp value that represents the point-in-time log truncation point for system recovery. Before you run the RESTORE SYSTEM utility to recover system data, you must use the SYSPITR or SYSPITRT option of DSNJU003. The options enable you to create a conditional restart control record to truncate the logs for system point-in-time recovery.

Log-truncation-timestamp specifies a timestamp value that is to be used as the log truncation point. A valid log truncation point is any UTC timestamp for which there exists a log record with a timestamp that is greater than or equal to the specified timestamp value. Any log information in the bootstrap data set, the active logs, and the archive logs with a timestamp greater than SYSPITRT is discarded. If you omit SYSPITRT, Db2 determined the end of the log range. Use the same timestamp value for all members of the data sharing group that require log truncation.

You cannot specify any other option with CREATE, SYSPITRT.

Note: The startime keyword specifies the start time in the yyyydddhhmmsst timestamp format. See the STARTIME option for details about the timestamp.

CANCEL

On the CRESTART statement, deactivates the currently active conditional restart control record. The record remains in the BSDS as historical information.

No other keyword can be used with CANCEL on the CRESTART statement.

On the CHECKPT statement, deletes the checkpoint queue entry that contains a starting RBA that matches the parameter that is specified by the STARTRBA keyword.

Attention: This statement can override DB2®'s efforts to maintain data in a consistent state. Do not use this statement without understanding the conditional restart process.

CHKPTRBA= chkptrba

Identifies the log RBA of the start of the checkpoint record that is to be used during restart.

If you use STARTRBA or ENDRBA, and you do not use CHKPTRBA, the DSNJU003 utility selects the RBA of an appropriate checkpoint record. If you do use CHKPTRBA, you override the value that is selected by the utility.

chkptrba must be in the range that is determined by startrba and endrba or their default values.

If possible, do not use CHKPTRBA; let the utility determine the RBA of the checkpoint record.

CHKPTRBA=0 overrides any selection by the utility; at restart, Db2 attempts to use the most recent checkpoint record.

FORWARD=

Indicates whether to use the forward-log-recovery phase of Db2 restart, which reads the log in a forward direction to recover any units of recovery that were in one of the following two states when Db2 was last stopped:

• Indoubt (the units of recovery had finished the first phase of commit, but had not started the second phase)
• In-commit (had started but had not finished the second phase of commit)

YES

Allows forward-log recovery.

If you specify a cold start (by using the same value for STARTRBA and ENDRBA), no recovery processing is performed.

NO

Terminates forward-log recovery before log records are processed. If a very old unit of recovery exists, specify this option to avoid a lengthy restart. Db2 does not go back in the log to the beginning of units of recovery to complete forward phase of Db2 restart. Instead, it marks them as bypassed and completed in the log. Database writes that are pending at the end of the log, including updates from other units of recovery, are written out during the forward phase of restart. However, Db2 skips page rewrites for non-GBP dependent page sets for committed URs and may skip page rewrites for any uncomitted URs that exist. The missing log apply for such URs might cause inconsistencies. Any updates that must be rolled-back, such as for an in-flight or in-abort unit of recovery, are done during the backout phase of restart. However, if the backward phase is not skipped, it might encounter log apply errors for in-flight or in-abort URs because forward log apply was skipped for those URs.

Important: FORWARD=NO does not process any log records during forward log phase in restart, including log records for committed, in-flight or in-abort, and indoubt URs. If you want to avoid processing log records for long running in-doubt URs, but you want to process log records for all other URs, then you can use the following approach instead:

CRESTART CREATE,STARTRBA=rba-value

For rba-value, specify either the second prior checkpoint at the time of the restart, or possibly a checkpoint before the oldest in-flight or in-abort UR. If you want to use this approach and are unsure which option to choose, contact IBM Support.

BACKOUT=

Indicates whether to use the backward-log-recovery phase of Db2 restart, which rolls back any units of recovery that were in one of the following two states when Db2 was last stopped:

• Inflight (did not complete the first phase of commit)
• In-abort (had started but not finished an abort)

YES

Allows backward-log recovery.

If you specify a cold start (by using the same value for STARTRBA and ENDRBA), no recovery processing is performed.

NO

Terminates backward-log recovery before log records are processed.

CSRONLY

Performs only the first and second phases of restart processing (log initialization and current-status rebuild). After these phases, the system status is displayed, and restart terminates. Some parts of the log initialization are not performed, including any updating of the log and display of STARTRBA and ENDRBA information.

When Db2 is restarted with this option in effect, the conditional restart control record is not deactivated. To prevent the control record from remaining active, use the DSNJU003 utility again with CRESTART CANCEL, or with CRESTART CREATE to create a new active control record.

NEWCAT

Changes the VSAM catalog name in the BSDS.

VSAMCAT= catalog-name

Changes the VSAM catalog name entry in the BSDS.

catalog-name can be up to eight characters long. The first character must be alphabetic, and the remaining characters can be alphanumeric.

DDF

Updates the LOCATION, LUNAME, and other DDF related information values in the BSDS. If you use this statement to insert new values into the BSDS, you must include at least the LOCATION in the DDF statement. To update an existing set of values, you need to include only those values that you want to change. The DDF record cannot be deleted from the BSDS after it has been added; it can only be modified.

LOCATION= location-name

Changes the LOCATION value in the BSDS.

location-name specifies the name of your local Db2 site.

PORT

Identifies the TCP/IP port number that is used by DDF to accept incoming connection requests. This value must be a decimal number between 0 and 65535, including 65535; 0 indicates that DDF TCP/IP support is to be deactivated. Set PORT to 0 if DDF needs to use only VTAM®, and will not use TCP/IP.

If Db2 is part of a data sharing group, all the members of the Db2 data sharing group must have the same value for PORT.

If the same port number is specified for the TCP/IP port and secure port, Db2 accepts only secure TCP/IP connection requests that are protected by SSL.

RESPORT

Identifies the TCP/IP port number that is used by DDF to accept incoming DRDA two-phase commit resynchronization requests. This value must be a decimal number between 0 and 65535, including 65535; zero indicates that DDF's TCP/IP support is to be deactivated. If RESPORT is non-zero, it must not use the same value that is supplied for PORT or SECPORT.

For data sharing Db2 systems, RESPORT must be uniquely assigned to each Db2 member, so that no two Db2 members use the same TCP/IP port for two-phase commit resynchronization.

Db2 uses the resync port only for transaction completion states after failures, so no user data or authentication protocols are ever exchanged on the secure port. You can enable SSL security for the resync port by defining a network policy on all Db2 requesters and Db2 servers.

SECPORT

Identifies the TCP/IP port number that is used by DDF to accept inbound secure DRDA connection requests. This value must be a decimal number between 0 and 65535, including 65535; zero indicates that DDF's secure connection support for TCP/IP is deactivated.

If the same port number is specified for the TCP/IP port and secure port, Db2 accepts only secure TCP/IP connection requests that are protected by SSL. For more information, see Configuring the Db2 server for SSL.

ALIAS= alias-name :alias-port :alias-secport

Specifies one or more alias names for the location. An alias name is a name besides the location name that connect processing can accept. Specifying an alias name does not change the location identifier for a database object.

Important: ALIAS applies to DRDA connections only.

You can specify and modify as many as eight location aliases by using the DSNJU003 utility. To specify more than 8 aliases, use the MODIFY DDF command with the ALIAS option.

alias-name specifies from 1 to 16 characters for the location name. alias-name cannot be one of the valid DSNJU003 keywords.

:alias-port specifies a TCP/IP port number for the alias that can be used by DDF to accept distributed requests. This value must be a decimal number between 1 and 65535, including 65535. The value must be different from the values for the PORT, RESPORT, and SECPORT options and any value that was specified for alias-port or alias-secport of any other defined alias. The value can be the same value that is specified for :alias_secport of the same location alias. Specify a value for alias-port when you want to identify a subset of data sharing members to which a distributed request can go.

:alias-secport specifies a secure TCP/IP port number for the alias that can be used by DDF to accept secure distributed requests using SSL. This value must be a decimal number between 1 and 65535, including 65535. The value must be different from the values for the SECPORT, PORT, and RESPORT options, and any value that was specified for alias-port or alias-secport of any other defined alias. The value can be the same value that is specified for :alias_port of the same location alias. Specify a value for alias-secport when you want to identify a subset of data sharing members to which a secure distributed request can go.

You can add or replace aliases by respecifying the ALIAS option. The new list of names replaces the existing list.

Related information:

• -MODIFY DDF (Db2)

IPNAME=ipname

Identifies and associates an IPNAME value with DDF.

ipname can be up to 8 bytes in length and must be an alphanumeric string, beginning with a letter. When you specify this option, DDF activates only its TCP/IP communications support, regardless of whether or not there is a value for LUNAME. Only inbound and outbound DRDA protocol communications over TCP/IP are allowed. Db2 makes no attempt to activate SNA/APPC communications support.

The value specified must be either unique to this Db2 subsystem within an enterprise, or if the Db2 subsystem is configured to be a member of a data sharing group, unique to the data sharing group for which this Db2 subsystem is a member. All members of a data sharing group must be defined with the same IPNAME value if all the members are to activate only their TCP/IP communications support. If some members of a data sharing group activate their SNA/APPC (as well as TCP/IP communications support), then the IPNAME value chosen for the TCP/IP only members must match the GENERIC value specified for the members which activate their SNA/APPC as well as their TCP/IP communications support.

Db2uses the IPNAME value as the network-ID portion of a unit-of-work identifier. A unit-of-work identifier has traditionally been made up of a network-ID, an LU name, a 6-byte unique identifier created from a timestamp, and a 2-byte current commit count value. When running with an IPNAME value, the LU name portion of the unit-of-work identifier is created from a 4-byte character representation of the hexadecimal notation of the value specified for the TCP/IP resync port (RESPORT).

Do not confuse IPNAME with a TCP/IP external such as a hostname or domain name. The value you give IPNAME is only used internally by Db2 or in a DRDA exchange with another DRDA server. It cannot be referenced by any TCP/IP external, such as ping. Also, the hostname of the system upon which the Db2 is running is a poor choice for IPNAME because more than one Db2 could run on the same system, and the IPNAME value given to any Db2 or Db2 data sharing group must be unique within an enterprise.

IPV4= ipv4-address

Identifies and associates a constant IPv4 IP address with DDF to accept incoming connection requests to this specific subsystem only. This address must be entered in dotted decimal form. If an IP address is not specified, Db2 will automatically determine the IP address from TCP/IP.

When Db2 is a member of a data sharing group, it is strongly recommended that you refer to a dynamic virtual IP address (DVIP). A group IP address, GRPIPV4, should also be specified.

IPV6= ipv6-address

Identifies and associates a constant IPv6 IP address with DDF to accept incoming connection requests to this specific subsystem only. This address must be entered in colon hexadecimal form. If an IP address is not specified, Db2 will automatically determine the IP address from TCP/IP.

When Db2 is a member of a data sharing group, it is strongly recommended that you refer to a dynamic virtual IP address (DVIP). A group IP address, GRPIPV6, should also be specified.

GRPIPV4

Identifies and associates a constant IPv4 IP address with the data sharing group for which this DDF is a member. The IP address is used to accept incoming connection requests that can be serviced by any member of the data sharing group. This address must be entered in dotted decimal form. An associated IPv4 subsystem/member address must also be specified in order to identify the IP address associated with this specific member of the group. If an IP address is not specified, Db2 will automatically determine the IP address from TCP/IP.

It is strongly recommended that you refer to a sysplex distributor owned distributing dynamic virtual IP address (DVIPA).

GRPIPV6

Identifies and associates a constant IPv6 IP address with the data sharing group for which this DDF is a member. The IP address is used to accept incoming connection requests that can be serviced by any member of the data sharing group. This address must be entered in colon hexadecimal form. An associated IPv6 subsystem/member address must also be specified in order to identify the IP address associated to this specific member of the group. If an IP address is not specified, Db2 will automatically determine the IP address from TCP/IP.

It is strongly recommended that you refer to a sysplex distributor owned distributing dynamic virtual IP address (DVIPA).

LUNAME= luname

Changes the LUNAME value in the BSDS.

luname specifies the LUNAME value. The LUNAME in the BSDS must always contain the value that identifies your local Db2 subsystem to the VTAM network.

PASSWORD=

The DDF password follows VTAM convention, but Db2 restricts it to one to eight alphanumeric characters. The first character must be either a capital letter or an alphabetic extender. The remaining characters can consist of alphanumeric characters and alphabetic extenders.

password

Optionally assigns a password to the distributed data facility communication record that establishes communications for a distributed data environment. The PRTCT=password option on the APPL definition statement is used to define Db2 to VTAM.

GENERIC= gluname

Replaces the value of the Db2 GENERIC LUNAME subsystem parameter in the BSDS.

gluname specifies the GENERIC LUNAME value.

NOPASSWD

Removes the archive password protection for all archives that are created after this operation. It also removes a previously existing password from the DDF record. No other keyword can be used with NOPASSWD.

NGENERIC

Changes the Db2 GENERIC LUNAME to binary zeros in the BSDS, indicating that no VTAM generic LU name support is requested.

NOALIAS

Indicates that no alias names exist for the specified location. Any alias names that were specified in a previous DSNJU003 utility job are removed.

NOIPV4

Removes the constant IPv4 address from the BSDS. The NGRPIPV4 keyword must also be specified to ensure that the associated group address, if any, is also removed.

NOIPV6

Removes the constant IPv6 address from the BSDS. The NGRPIPV6 keyword must also be specified to ensure that the associated group address, if any, is also removed.

NGRPIPV4

Removes the constant data sharing group IPv4 address from the BSDS.

NGRPIPV6

Removes the constant data sharing group IPv6 address from the BSDS.

NOIPNAME

Removes the IPNAME value from the DDF record. No other keyword can be used with NOIPNAME.

NOLUNAME

Removes the LUNAME value from the DDF record. No other keyword can be used with NOLUNAME.

CHECKPT

Allows updating of the checkpoint queue with the start checkpoint and end checkpoint log records.

Attention: This statement can override the effort ot Db2 to maintain data in a consistent state. Do not use the statement without understanding the conditional restart and checkpoint processing processes.

TIME= time

On the CHECKPT statement, specifies the time that the start checkpoint record was written.

On the HIGHRBA statement, TIME specifies when the log record with the highest RBA was written to the log.

time specifies the time value. For timestamp format, see the STARTIME option description.

HIGHRBA

Updates the highest-written log RBA in either the active or archive log data sets.

Attention: This statement can override the effort of Db2 to maintain data in a consistent state. Do not use the statement without understanding the conditional restart process.

OFFLRBA= offlrba

Specifies the highest-offloaded RBA in the archive log.

offlrba is a hexadecimal number of up to 20 characters. If you use fewer than 20 characters, leading zeros are added. The value must end with hexadecimal X'FFF'.

DELMBR

Deactivates or destroys a member of a data sharing group.

DEACTIV

Marks a member of a data sharing group for deactivation. Deactivation is the first step in deletion of a member from a data sharing group.

Before the member can be deactivated, it must be quiesced and have no outstanding work. The logs and BSDS must exist.

DESTROY

Completes the deletion of a member from a data sharing group.

After a member is destroyed, its member ID can be reused, and the logs and BSDS can be deleted.

RSTMBR

Restores a deactivated member of a data sharing group to the quiesced state.

MEMBERID= member-id

Specifies the data sharing group member that is to be deactivated, destroyed, or restored.

member-id is a number in the range 1 - 32. This number is the member ID that shown in the output from the DISPLAY GROUP command or the DSNJU004 (print log map) utility.

Related information:

• DSNJU004 (print log map) output
• DSN7100I