18.4.2 Options

Organization of this subsection
(1) -m [host-name:]first-HiRDB-filename-in-master-directory-RDAREA
(2) -M{x|r|s}
(3) -p process-results-output-filename
(4) -i
(5) -j lock-release-wait-time
(6) -E EasyMT-MT-attributes-definition-filename
(7) -B EasyMT I/O buffer sectors count
(8) -z log-point-information-file-name
(9) -J
(10) -w pause-period,read-count
(11) -f control-information-filename
(12) -b {[host-name:]backup-file-name[,backup-file-name]...|[host-name:]device-symbolic-name[,device-symbolic-name]|[host-name:]device-group-name|[host-name:]object-name|[host-name:]policy-name}
(13) -a
(14) -u unit-identifier[,unit-identifier]...
(15) -s server-name[,server-name]...
(16) -r RDAREA-name[,RDAREA-name]...
(17) -k {u|i|e|m|o|n}
(18) -v volume-name[,volume-name]...
(19) -N EasyMT-filename
(20) -G barlist-filename
(21) -g differential-backup-group-name
(22) -d backup-type
(23) -K HiRDB-file-system-area-name-storing-differential-backup-management-file
(24) -o differential-backups-history-filename
(25) -L differential-backup-management-file-size
(26) -q generation-number
(27) -S {backup-file-initial-size,increase-value | '(backup-file-initial-size, increase-value) [,(backup-file-initial-size,increase-value)]...'}

(1) -m [host-name:]first-HiRDB-filename-in-master-directory-RDAREA

[Figure]<identifier:pathname> ((up to 167 characters))

Specifies the name of the first HiRDB file in the master directory RDAREA. Specify the name of the host containing the master directory RDAREA and the path name, separated by a colon (:). This is the name specified in the pd_master_file_name operand in the system common definitions.

Example

-m host01:/hirdb_s/rdarea/rdmast

Rules
  1. You can omit the host name if the master directory RDAREA is located at the server machine where the database copy utility is executed (where the pdcopy command is entered). In this case, specify only the path name.
  2. If you specify the host name, make sure that it is a host name specified in the -x option of the pdunit operand in the system definition.
  3. If you are using the system switchover facility, make sure that the host name of the primary system is specified.

(2) -M{x|r|s}

Specifies the backup acquisition mode. The following table shows available backup acquisition modes:

-M OptionDescriptionDatabase recovery method
x
(Reference/update-impossible mode)
This mode does not allow an RDAREA under backup processing to be referenced or updated. Use the pdhold command to place the RDAREA subject to backup processing in the shutdown and closed status.Only the backup copy is needed to restore the database to the status existing when the backup copy was made.
r
(Reference-possible mode)
This mode allows an RDAREA under backup processing to be referenced but does not allow it to be updated.
s
(Updatable mode)
This mode allows an RDAREA under backup processing to be referenced or updated. To use this mode, the RDAREA subject to backup processing must be created in a character special file.To restore the database, you need the backup copy and the system log obtained at the synchronization point immediately before the backup*.
Notes
  1. To back up the master directory RDAREA with -M x specified, start HiRDB using the pdstart -r command, then back up the master directory RDAREA using the database copy utility.
    If you specify -M r, you can back up the master directory RDAREA with the pdcopy command by starting HiRDB with the normal pdstart command.
  2. If RECOVERY PARTIAL or RECOVERY NO is specified in CREATE TABLE for a LOB column (including when the RECOVERY operand is omitted), -M s does not back up the user LOB RDAREA containing the LOB column nor the data dictionary LOB RDAREA storing the object.
  3. If -M s is specified and another transaction is updating a page subject to backup processing, the utility waits until the transaction issues a COMMIT. Therefore, pdcopy may result in a time-out.
  4. If you are executing pdcopy in the updatable mode using an LVM for the database, you must place the target RDAREAs in backup hold status.

* The database copy utility process results listing contains the name and generation number of the system log file required for restoring the RDAREA. For details about the database copy utility process results listing, see Section 18.6 Database copy utility process results listing.

Notes
  1. To back up all RDAREAs during HiRDB operation (in units of systems), specify r or s as the backup acquisition mode (-M option). You cannot specify x for the following reason: to specify x, you need to place all RDAREAs subject to backup processing in the shutdown and closed status using the pdhold command. However, the master directory RDAREA cannot be placed in the shutdown and closed status. Therefore, you cannot specify x in the -M option to back up all RDAREAs during HiRDB operation.
  2. You cannot make a backup copy specifying -M s while a UAP or utility is executing in the no-log mode or pre-update log acquisition mode. To back up data after completion of a UAP or utility in the no-log mode or pre-update log acquisition mode, you need to specify -M r or -M x.
  3. When you make a backup copy with -M s specified in either of the following cases, you must first use the pdlogswap -w command to swap system logs:
    • When pdmod has been executed (to exclude configuration modification processing from the system log file)
    • After operating in the no-log mode (to exclude no-log processing from the system log file)
  4. For a shared RDAREA, -M s cannot be specified to acquire a backup.

(3) -p process-results-output-filename

[Figure]<pathname>

Specifies the name of the file to which the processing result of the database copy utility is to be output.

Example

-p /usr/pdcopy/list01

Rules
  1. Specify the path name of the server machine at which the database copy utility is executed (where the pdcopy command is entered).
  2. If the specified path name is not found or this option is omitted, the database copy utility outputs its processing result to the server machine where the utility is executed. The utility displays the output destination in the KFPR26022-I message.
  3. Regardless of the specification of this option, the utility outputs all error messages to the system log file and standard output, and the final processing result to the standard output. However, error messages may not be output to the system log file and to the standard output in the same order.

(4) -i

Specifies that utilization status is to be output to the process results output file. This option is applicable to the following RDAREAs:

The following information is output to the process results output file:

TypeInformationDescription
PagePage count
  • Number of available pages
  • Number of unused pages
  • Number of used pages
    [Figure](a) Number of pages storing index1
    [Figure] (b) Number of pages storing table1
Ratio
  • Ratio of unused pages to available pages
  • Ratio of used pages to available pages
  • Ratio of pages storing index to used pages1
  • Ratio of pages storing table to used pages1
Area length1Length
  • Sum of free areas in used pages
  • Sum of free areas in pages storing index
  • Sum of free areas in pages storing table
  • Area length of all pages storing index
  • Area length of all pages storing table
Ratio
  • Ratio of sum of free areas to the area length of all pages storing index
  • Ratio of sum of free areas to the area length of all pages storing table
System information fragmentation rate2RatioExtent of fragmentation of system information in data dictionary LOB RDAREAs and user LOB RDAREAs3

1 This information is not output for data dictionary LOB RDAREAs or user LOB RDAREAs.

2 This information is output only for data dictionary LOB RDAREAs and user LOB RDAREAs.

3 A value greater than 1% can have adverse effects on access performance, in which case you need to reorganize the database using the database reorganization utility. For a data dictionary LOB RDAREA, reorganize the dictionary tables related to stored procedures (SQL_ROUTINES, SQL_ROUTINE_RESOURCES, SQL_ROUTINE_PARAMS).

(5) -j lock-release-wait-time

[Figure]<unsigned integer> ((0-200,000))

Specifies in seconds the maximum wait time for monitoring lock release.

Lock-release wait time is the time from when a lock request enters a wait status, until the status is released. If the lock is not released within the specified time, pdcopy terminates with an error.

If you specify a value of 0, the utility waits until the lock is released, without monitoring the lock-release wait time. If you omit this operand, the utility assumes the value of the pd_lck_wait_timeout operand in the system definitions.

Notes
  1. If -M r is specified and the specified RDAREA is already locked by another user, pdcopy terminates with an error.
  2. If -M s is specified and a requested page is being released by another user, pdcopy is placed in lock-release wait status until the user's transaction is terminated. In this case, specify a lock-release wait time that is greater than the execution time of the transaction conducting the page release. Page release occurs mainly in the following cases:
    • DROP TABLE
    • DROP INDEX
    • PURGE TABLE statement
    • INSERT or UPDATE statement executed on a column with an index
    • DELETE statement executed on a column with an index (with duplicate key)
    • DELETE statement executed after LOCK statement; or, UPDATE statement executed after LOCK statement resulting in a change to row length

(6) -E EasyMT-MT-attributes-definition-filename

[Figure]<path name>

Specifies the name of the EasyMT MT attributes definition file. This file must be connected to the server machine where the database copy utility is executed (where the pdcopy command is entered). The -E option is applicable to a backup file that has been created with e or m specified in the -k option. The following attributes take effect:

When the -B option is specified, it takes precedence over the number of I/O buffer sectors specified in the MT attributes definition file.

The utility checks the information specified in this file when EasyMT is executed.

(7) -B EasyMT I/O buffer sectors count

[Figure]<unsigned integer> ((1 255))

Specifies the number of I/O buffer sectors to be used for magnetic tape input/output operations. A larger value results in better performance, but it requires more memory.

Rules
  1. The -B option is applicable to a backup file that has been created with e or m specified in the -k option.
  2. If the -E and -B options are both omitted, the utility assumes EasyMT's default values.

(8) -z log-point-information-file-name

[Figure]<pathname>

This option is applicable to an operation that does not involve unloading of the system log. Specify the name of the file used to store the log point information.

Example

-z /usr/pdcopy/logp01

Rules
  1. Specify the path name of the server machine where the database copy utility is executed (where the pdcopy command is entered).
  2. If this log point information file is specified in the pdlogchg command, any unneeded system log file created before the log point is placed in the unload completed status. When this option is specified, the utility also stores the log point information in the backup file.
Notes
  1. For a HiRDB/Parallel Server, for RDAREAs subject to backup processing, specify only the names of those RDAREAs that are located at the same server.
  2. The server containing the RDAREAs subject to backup processing may be active when the utility is executed.
  3. If a backup copy is to be made by starting HiRDB with the pdstart -r command, the -z option cannot be specified. To back up all data by specifying the -z option, start HiRDB with pdstart, then specify pdcopy -M r. The utility obtains a log point information file for each server; therefore, for a HiRDB/Parallel Server, it is impossible to back up all data at the same time. Additionally, the master directory RDAREAs cannot be placed in the shutdown or closed status. Therefore, if the processing is started with pdstart, pdcopy -M x -a cannot be specified.

(9) -J

Specifies that when a skip target error is detected while copying RDAREAs, the next RDAREA is to be copied immediately without terminating the pdcopy command.

The -J option is mutually exclusive with the -g, -K, and -d options. If they are specified, specification of the -J option is ignored.

The table below describes the skip target errors. Note that a skip may not have occurred if the KFPR26061-W message was not issued before or after the error message.

Skip target errorMessage issued for a skip target error
A name specified in the -r option is not defined as an RDAREA, or a list RDAREA was specified.KFPR26006-E
-M x was specified, but the RDAREAs to be copied include the master directory RDAREA.KFPR26111-E
An open or close error was detected before copy processing started.KFPR26012-E
-M s was specified, but the server containing an RDAREA to be copied is not active.KFPR26015-E
recovery partial or recovery no was specified during table definition and -M s was specified, but the RDAREAs to be copied include a user LOB RDAREA.KFPR26030-E
-M x was specified, but an RDAREA to be copied is not in shutdown and closed status.KFPR26110-E

(10) -w pause-period,read-count

Because execution of pdcopy involves transfer of a large amount of data, the CPU usage rate increases and other online jobs may be affected adversely. Specifying the -w option pauses the utility for a specified period of time after reading a specific amount of data, thereby minimizing adverse effects on other online jobs.

Although specifying this option reduces the CPU usage rate, the pdcopy processing time increases.

Figure 18-2 provides an overview of the processing when the -w option is specified.

Figure 18-2 Overview of processing when the -w option is specified

[Figure]

pause-period[Figure]<unsigned integer> ((10[Figure]60000))

Specifies the pause period during data read operations in units of 10 milliseconds (example: 10, 20, 30,...). If the specified value is not a multiple of 10 milliseconds, it is rounded up.

The utility performs as many read operations as specified in read-count and then pauses for the specified period.

read-count[Figure]<unsigned integer> ((1[Figure]10000))

Specifies the number of read operations to be performed before the processing is paused.

The utility performs as many read operations as specified here and then pauses for the specified period.

Rules
  1. In determining the pause period and read count, take into account the total size of the data to be read by pdcopy and the total number of read operations.
  2. For a HiRDB/Parallel Server, the processing specified with the -w option is performed for each server. Read operations are counted for each server and pause processing is performed for each server. Check the server that contains the target RDAREA before specifying the pause period and read count.
Determining the processing time when the -w option is specified
  1. Obtain the data read count for each RDAREA.
    [Figure]For a data directory RDAREA
    [Figure](4096 [Figure] number of records in the HiRDB file system)/65536[Figure]
    + (number of HiRDB files constituting the RDAREA - 1)
    [Figure]For other RDAREA
    [Figure]Page length [Figure] (number of records in the HiRDB file system
    - (number of unused segments [Figure] segment size))/65536[Figure]
    + (number of HiRDB files constituting the RDAREA - 1)
  2. Obtain the total number of data read operations.
    [Figure]For a HiRDB/Single Server
    Sum of the numbers of data read operations for all target RDAREAs
    [Figure]For a HiRDB/Parallel Server
    Largest sum of the numbers of data read operations for all target RDAREAs among all applicable servers
  3. Obtain the processing time.
    [Figure](total number of data read operations obtained in 2/number of read operations)[Figure]
    [Figure] pause period (milliseconds)
    + pdcopy processing time

(11) -f control-information-filename

[Figure]<pathname>

Specifies the name of the control information file.

Example

-f /usr/pdcopy/cont01

This control statement must contain the options beginning at (12) as follows. You can also specify these options directly in the pdcopy command.

Notes
  1. Create the control statement file at the server where the database copy utility is executed (where the pdcopy command is entered).
  2. Specify in the control statement file at least one set of a backup file name, options for the RDAREAs subject to backup processing, a file type, a volume name, and an EasyMT file name. Note that each set of this information must be specified on a single line (which is up to 32,768 bytes). You can specify a maximum of 16 lines of information.
  3. If you are using the differential backup facility with the -a or -q option specified, you can specify only one control statement line.
Format of the control statement file

 -b flag-argument {-a|-u flag-argument|-s flag-argument|-r flag-argument}
                [-k flag-argument][-v flag-argument][-N flag-argument]
                [-S flag-argument][-G flag-argument]
[-b flag-argument [{-u flag-argument|-s flag-argument|-r flag-argument}]  
                 [-k flag-argument][-v flag-argument][-N flag-argument]
                 [-S flag-argument]][-G flag-argument]
[-b flag-argument [{-u flag-argument|-s flag-argument|-r flag-argument}]
                 [-k flag-argument][-v flag-argument][-N flag-argument]
                            [-S flag-argument]][-G flag-argument]
                       .
                       .
                       .

Notes
  1. For details about each option flag, see the explanation of the -b option as follows.
  2. When specifying multiple lines, you can specify only those option flags specified on the first line also on the subsequent lines (except for the -a option). For example, if you specify the -u option on the first line, you can specify the -u option on any subsequent line.

(12) -b {[host-name:]backup-file-name[,backup-file-name]...|[host-name:]device-symbolic-name[,device-symbolic-name]|[host-name:]device-group-name|[host-name:]object-name|[host-name:]policy-name}

[Figure]<identifier: path name> ((up to 167 characters when -k i is specified))

Specifies the backup file name, object name of JP1/OmniBack II, or policy name specified with NetBackup.

Rules
  1. If you are creating a backup file at the server machine where the database copy utility is executed (where the pdcopy command is entered), you can omit the host name, in which case you need to specify the path name only.
  2. If you are specifying a host name, make sure that it is a host name specified in the -x option of the pdunit operand in the system definition.
  3. If the server machine used to execute the database copy utility (where the pdcopy command is entered) has a disk agent of JP1/OmniBack II or a NetBackup client, you can omit the host name.
  4. If you are using a control statement file, make sure that there is no duplication of backup file names on a line.
  5. If you execute backup processing with an existing backup file name specified, that backup file will be overwritten.
  6. A maximum of 1,120 RDAREAs can be backed up into a single backup file.
  7. To back up RDAREAs into a file under a host using the system switchover facility, make sure that the host name of the primary system is specified. The utility creates a backup file under the host of the running system.
  8. When using a disk agent of JP1/OmniBack II or a NetBackup client at a host that uses the system switchover facility, make sure that the host name of the primary system is specified. The disk agent of JP1/OmniBack II or NetBackup client will be the running system.
  9. If you are using NetBackup and making a backup copy using the same policy name, do not change the RDAREAs to be backed up. If the policy name is the same but some of the RDAREAs are different, the system may not be able to restore the RDAREAs.
  10. The following table describes the relationship between the -b and -k options:
    -k option specification-b option specification
    -k uYou can specify one or more regular file names or one MT special file name. The second and subsequent files will be used only when there is more backup data than can be accommodated in the first file. Therefore, you need to specify in such a manner that each file is allocated to a different partition.
    -k iYou can specify multiple sets of "HiRDB-file-system-area-name/HiRDB-file-name". The second and subsequent files will be used only when there is backup data that can be accommodated in the first file. You must create the HiRDB file system area as a utility-dedicated area in advance with the pdfmkfs command.
    -k eYou can specify an MT special file name.
    -k mYou can specify a maximum of two device symbolic names or one device group name managed by MTguide. The system does not check whether a device symbolic name or a device group name is specified.
    -k oYou can specify any name with a maximum length of 512 bytes. The specified name becomes the object name of JP1/OmniBack II.
    For the host name, specify the name of the host that has both a disk agent of JP1/OmniBack II and HiRDB.
    You can check the object names with the omnidb command (with the -stream option specified) of JP1/OmniBack II.
    -k nYou can specify the policy name specified with NetBackup. Express a policy name as 1-128 bytes of alphanumeric characters.
    For the host name, specify the name of a host that has both NetBackup client and HiRDB.
    You can check the policy names with the bpimagelist command of NetBackup.
Notes
  1. When you create multiple backup files, note the following:
    • NFS files cannot be used.
    • A tape device cannot be used directly without EasyMT.
    • When you specify multiple backup files, use regular files or HiRDB files.
  2. If you specify the same policy name to execute pdcopy more than once, you can specify a backup file using the -U option's time specification when the database is restored (during execution of pdrstr). For the time of the backup file, specify the date and time of the KFPR26071-I message that is displayed during execution of pdcopy.

(13) -a

Specifies that RDAREAs are to be backed up in units of systems (all RDAREAs in the system are to be backed up).

Rules
  1. If you specify the -a option, the control statement file can contain only one line of control statement.
  2. If you specify the -a option when a replica RDAREA has been defined, the replica RDAREA will also be backed up.
  3. If an open error occurs on a HiRDB file in a replica RDAREA, the utility skips backup of that RDAREA and backs up another RDAREA. If the differential backup facility is used and an open error occurs in a HiRDB file in a replica RDAREA, the utility terminates with an error.
  4. If an input/output error occurs while a HiRDB file in a replica RDAREA is being read, the utility terminates with an error.

(14) -u unit-identifier[,unit-identifier]...

[Figure]<identifier> ((4 characters))

Specifies unit identifiers when RDAREAs are to be backed up in units of units (all RDAREAs in the specified units are to be backed up).

Rules
  1. If specifying multiple unit identifiers, make sure none of them is duplicated.
  2. When using a control statement file, make sure that each specified unit identifier is unique among all the control statements.
  3. If you specify the -u option when a replica RDAREA has been defined, the replica RDAREA will also be backed up.
  4. If an open error occurs in a HiRDB file in a replica RDAREA, the utility skips backup of that RDAREA and backs up another RDAREA. If the differential backup facility is used and an open error occurs in a HiRDB file in a replica RDAREA, the utility terminates with an error.
  5. If an input/output error occurs while a HiRDB file in a replica RDAREA is being read, the utility terminates with an error.

(15) -s server-name[,server-name]...

[Figure]<identifier> ((1-8 characters))

Specifies the names of servers when RDAREAs are to be backed up in units of servers (all RDAREAs in the specified servers are to be backed up).

Rules
  1. If specifying multiple server names, make sure none of them is duplicated.
  2. When using a control statement file, make sure that each specified server name is unique among all the control statements.
  3. If you specify the -s option when a replica RDAREA has been defined, the replica RDAREA will also be backed up.
  4. If an open error occurs in a HiRDB file in a replica RDAREA, the utility skips backup of that RDAREA and backs up another RDAREA. If the differential backup facility is used and an open error occurs in a HiRDB file in a replica RDAREA, the utility terminates with an error.
  5. If an input/output error occurs while a HiRDB file in a replica RDAREA is being read, the utility terminates with an error.

(16) -r RDAREA-name[,RDAREA-name]...

[Figure]<identifier> ((1-30 characters))

Specifies the names of the RDAREAs to be backed up.

Rules
  1. If specifying multiple RDAREA names, make sure none of them is duplicated.
  2. When using a control statement file, make sure that each specified RDAREA name is unique among all the control statements (except for a regular expression).
  3. You can specify RDAREA names in three different regular expressions, as follows. They enable a group of RDAREAs to be backed up collectively. When using a regular expression, be sure to use only one type of expression.
    Regular expressionExplanationExample
    character-string*Any RDAREA name beginning with the specified character stringPDUSER*
    All RDAREAs with names beginning with PDUSER are processed.
    *character-string*Any RDAREA name containing the specified character string*PDUSER*
    All RDAREAs with names containing the character string PDUSER are processed.
    -Any RDAREA name containing the specified character string at the specified location..USER
    All RDAREAs with names containing the character string USER beginning in character 3 are processed.
Note
You can combine a regular expression of RDAREA names with individual RDAREA names.
Example

-r RDUSER*,RDMASTER

  1. The following shows the rules for specifying RDAREA names:
    TypeCoding method without using regular expressionCoding method using regular expression
    Command line specification (without using a control statement file)When specifying an RDAREA name in lowercase alphabetic characters or when the RDAREA name contains a space, enclose the entire name in double quotation marks ("). If you use the Bourne shell (sh), C shell (csh), or Korn shell (ksh), you also need to enclose the RDAREA name in single quotation marks (').Enclose the entire regular expression with single quotation marks ('). Regular expression is case sensitive.
    Specifying in a control statement fileWhen specifying an RDAREA name in lowercase alphabetic characters or when the RDAREA name contains a space, enclose the entire name in double quotation marks (").Regular expression is case sensitive.
  2. If the -q and -r options are both specified, only system RDAREAs and original RDAREAs can be specified. In this case, in the inner replica group to which an original RDAREA belongs, the replica RDAREA with the generation specified in the -q option is subject to processing.
  3. If the replica RDAREA with the generation specified in the -q option is undefined, the utility ignores this RDAREA and backs up another RDAREA.
  4. If an open error occurs in a HiRDB file in a replica RDAREA, the utility skips backup of that RDAREA and backs up another RDAREA. If the differential backup facility is used and an open error occurs in a HiRDB file in a replica RDAREA, the utility terminates with an error.
  5. If an input/output error occurs while a HiRDB file in a replica RDAREA is being read, the utility terminates with an error.
  6. If the -q option is specified, the regular expression is applicable to the original RDAREA. If the -q option is omitted, the regular expression is applicable to all RDAREAs including the replica RDAREAs.

(17) -k {u|i|e|m|o|n}

Specifies the type of backup file.

u
Specifies that the backup is to be made to a regular file or to MT without using EasyMT, JP1/OmniBack II, or NetBackup. You cannot specify this option if the backup file consists of multiple volumes (specify -k m instead).
i
Specifies that the backup is to be made to a HiRDB file system area. This option is applicable when the backup file is shared by the running system and a standby system, and the standby system is to be recovered using the backup made by the running system.
e
Specifies that the backup is to be made to magnetic tape using EasyMT. You cannot specify this option if the backup file consists of multiple volumes. This specification requires the use of EasyMT.
m
Specifies that the backup is to be made to magnetic tape using EasyMT and MTguide. You must specify this option if the backup file consists of multiple volumes. This specification requires the use of EasyMT and MTguide.
o
Specifies that the backup is to be made using JP1/OmniBack II. Specify in the-b option the object name of JP1/OmniBack II.
n
Specifies that the backup is to be made using NetBackup. Specify in the -b option the policy name set by NetBackup.

(18) -v volume-name[,volume-name]...

[Figure]<alphanumerics> ((1-6 characters))

Specifies the volume names of the magnetic tapes to which the backup will be made. When this option is specified, an error results if a volume mounted at a tape deck does not match any of the specified values. If the number of volumes actually needed to make the backup is larger than the number of volumes specified here, the utility does not check any excess volume names. When this specification is omitted, the utility does not perform volume name checking.

This option is applicable to the backup file for which e or m is specified in the -k option. To specify multiple volume names, you need to specify m in the -k option.

Make sure that each specified volume name is unique among all volumes used.

(19) -N EasyMT-filename

[Figure]<alphanumerics> ((1-17 characters))

Specifies the file name to be assigned when the backup file is created. This specification takes effect only on the backup file made with e or m specified in the -k option. The utility always creates a backup file as the first file on the mounted magnetic tape (beginning at file number 1).

(20) -G barlist-filename

[Figure]<alphanumerics (excluding an underline (_))> <<1-64 characters>>

Specifies a barlist filename for JP1/OmniBack II.

A barlist file must be created at the server machine where JP1/OmniBack II's cell server is located.

Create the barlist file in the following directory:

/etc/opt/omni/barlists/stream

The following shows an example of a barlist file format:

Example

BARLIST "barlist-filename"   ...................1
DEFAULTS
{
}
DEVICE "logical-device-name"   .................2
{
}
CLIENT HiRDB host-name   .......................3
{
 -public   ....................................4
} -protect protection-specification   ..........5

Explanation
  1. Barlist file name (e.g., BARLIST "DLT01FILE")
  2. Logical device name (e.g., DEVICE"DLT01")
    For details about the logical device, see the applicable JP1/OmniBack II manual.
  3. Name of the host where the disk agent is located (e.g., CLIENT HiRDB h9000vr3)
  4. Required operand
  5. Protection specification
    protection-specification={none|days n|weeks n|until Date|permanent}
    none: Not protected.
    days n: For n, specify the number of days to be protected.
    weeks n: For n, specify the number of weeks to be protected.
    until Date: For Date, specify the date at which protection expires in the format YYYY/MM/DD, enclosed in double quotation marks (").
    permanent: Permanent protection.

(21) -g differential-backup-group-name

[Figure]<alphanumerics> ((1-30))

Specifies the name of a group as a unit subject to the differential backup facility.

If the specified differential backup group name ends with S, the utility immediately starts differential backup management, at which time the differential backup management file specified in the -k option is created. If an existing differential backup management file has the specified time, the utility deletes it. Therefore, you need to delete the backup files in the previous differential backup group.

For details about the differential backup facility, see the HiRDB Version 8 System Operation Guide.

Rules
  1. The utility always obtains a full backup for user LOB RDAREAs, data dictionary LOB RDAREAs, and registry LOB RDAREAs.
  2. For the same differential backup group, you cannot change the RDAREAs subject to backup processing.
  3. Store all backup files belonging to the same differential backup group at the same host (specify only one host name in the -b option).
  4. Create a differential backup file and a differential backup management file in separate HiRDB file systems.
  5. When the differential backup facility is used, the method for specifying the differential backup file name in the -b option depends on the specification of the -k option.
  • If -k u, -k i, or -k o is specified, you must specify a different name for each differential backup file.
  • This is not applicable when -k e or -k m is specified (the differential backup facility cannot be used).
  • If -k n is specified, there is no need to specify a different policy name for each differential backup file (different names may be specified, in which case the operation becomes complex because a policy needs to be created for each differential backup file).

(22) -d backup-type

If you are using the differential backup facility, use this option to specify the type of backup to be made:

a: Makes a full backup.

b: Makes an accumulation-differential backup from the most recent full backup.

c: Makes an accumulation-differential backup from one of the following backups, whichever is most recent:

d: Makes a differential backup from the previous backup.

(23) -K HiRDB-file-system-area-name-storing-differential-backup-management-file

[Figure]<pathname>

If you are using the differential backup facility, use this option to specify the name of the HiRDB file system area where the differential backup management file is to be stored. The utility uses the differential backup group name specified in the -g option as the name of the differential backup management file.

After executing pdcopy, you need to back up the differential backup management file using the pdfbkup command. If an error occurs in the differential backup management file, you must restore the differential backup management file from this backup copy using the pdfrstr command.

If you have executed pdmod while using the differential backup facility, make a full backup. In this case, also create a differential backup management file by specifying the -K option.

Make sure that "name-of-HiRDB-file-system-area-storing-differential-backup-management-file/differential-backup-group-name" does not exceed 167 characters.

(24) -o differential-backups-history-filename

[Figure]<pathname>

If you are using the differential backup facility, use this option to specify the name of the file to which historical information about differential backups is to be output.

For details about historical information, see Section 18.6 Database copy utility process results listing.

(25) -L differential-backup-management-file-size

[Figure]<unsigned integer> ((1-2,046)) <<1>>

If you are using the differential backup facility, use this option to specify the size of the differential backup file in MB.

This size must be less than the value used when the HiRDB file system area was created (value of the -n option in the pdfmkfs command minus 5).

You can increase the size of this file a maximum of 23 times. The utility uses this value when increasing the size of the file.

(26) -q generation-number

[Figure]<unsigned integer> ((0-10))

Specifies the generation number of the RDAREA to be backed up when the inner replica facility is used. When 0 is specified, the utility backs up the original RDAREA; when a value in the range 1-10 is specified, the utility backs up the specified replica RDAREA generation.

Rules
  1. Specify this option together with the -r option. This option cannot be specified together with the -a, -u, or -s option.
  2. Specify this option to back up the replica RDAREA with the indicated generation number in the inner replica group to which the original RDAREA specified in the -r option belongs.

(27) -S {backup-file-initial-size,increase-value | '(backup-file-initial-size, increase-value) [,(backup-file-initial-size,increase-value)]...'}

[Figure]<unsigned integer> ((1-1048574)) <<100,10>>

This option is applicable to the backup file for which i is specified in the -k option.

Specify in MB an initial size and a size increase value for the backup file (the specified increase value is allocated when the amount of backup data exceeds the initial size).

Rules
  1. The initial size must be less than the value specified in the -n option of the pdfmkfs command when the HiRDB file system area was created. This is because an area for system management information is required.
  2. When this option is omitted, the utility assumes 100 MB as the backup file initial size and 10 MB as the increase size.
  3. You can increase the size of a backup file a maximum of 23 times, provided that the number of increases does not exceed the permitted maximum number of increases specified when the HiRDB file system area was created.
  4. If specifying multiple files, you can specify an initial allocation size and increase value for each file. In this case, enclose each set of initial allocation sizes and increase values in parentheses, then enclose the entire specification in single quotation marks. If you are specifying this information in the control statement file, do not use single quotation marks.
  5. If the number of initial allocation sizes paired with increase values is less than the number of specified files, the utility assumes the last pair of values for the remaining files. If the number of initial allocation sizes paired with increase values is more than the number of specified files, the utility ignores the excess pairs of values.
Example

-b file1,file2,file3,file4
-k i
-s '(10,1),(1,1)'

FileInitial sizeIncrease valueRemarks
file1101Set by the (10, 1) specification.
file211Set by the (1, 1) specification.
file311Because there is no specification, the last specified values (1, 1) are applied.
file411