2.2.25 Operands related to RDAREAs

117) pd_max_rdarea_no = maximum-number-of-RDAREAs
~<unsigned integer>((5-8388592)) <<1000>>
  • 0904 compatibility mode: <<200>>
Specifies the maximum number of RDAREAs allowed. If the total number of RDAREAs exceeds the value specified for this operand, HiRDB cannot be started normally. Here, RDAREAs also include master directory RDAREAs, data directory RDAREAs, and data dictionary RDAREAs.
Specification guidelines
  • Specify a value that is equal to or greater than the total number of RDAREAs (leave some extra room). You can use the pddbls command to check the total number of RDAREAs. If you plan to add RDAREAs, specify a value by taking the additional RDAREAs into account.
  • For a HiRDB parallel server configuration, the value of this operand is applied to each back-end server. For example, if 100 is specified for this operand, a maximum of 100 RDAREAs can be created in each back-end server. Therefore, use the largest number of RDAREAs among the back-end servers as the guideline when specifying a value for this operand.
  • If you are using the inner replica facility, include the number of replica RDAREAs as well.
  • If you are using shared RDAREAs, also add the number of shared RDAREAs used by reference-only back-end servers.
Notes
  • Do not specify an unnecessarily large value for this operand. Increasing the specification value of this operand increases the size of the shared memory used by HiRDB. If a shortage occurs in the shared memory, HiRDB might not be able to start.
  • Do not specify this operand for a utility special unit.
Effects on individual estimation formulas
If the value of the pd_max_rdarea_no operand is changed, the following estimation formulas are affected:
HiRDB Version 9 Installation and Design Guide:
  • Calculation of required memory under Estimating the memory size required for a HiRDB single server configuration
  • Formulas for shared memory used by a unit controller under Estimating the memory size required for a HiRDB single server configuration
  • Calculation of required memory under Estimating the memory size required for a HiRDB parallel server configuration
  • Formulas for shared memory used by a unit controller under Estimating the memory size required for a HiRDB parallel server configuration
  • Determining the value of S under Determining the size of status files
  • Formula 1, Formula 2, and Formula 3 under Formulas for shared memory used by a single server
  • Formula 1 and Formula 2 under Formulas for the size of the shared memory used by a dictionary server
  • Memory size required for the execution of the database structure modification utility (pdmod)
  • Formula 1, Formula 2, and Formula 3 under Formulas for the size of the shared memory used by a back-end server
  • Formula for the shared memory used by a front-end server
118) pd_max_file_no = maximum-number-of-HiRDB-files-comprising-an-RDAREA
~<unsigned integer>((5-134217728))<<2000>>
  • 0904 compatibility mode: <<400>>
Specifies the maximum number of HiRDB files that comprise an RDAREA. If the total number of HiRDB files exceeds the value specified for this operand, HiRDB cannot be started normally. Here, HiRDB files comprising an RDAREA also include the HiRDB files of master directory RDAREAs, data directory RDAREAs, and data dictionary RDAREAs.
Specification guidelines
  • Specify a value that is equal to or greater than the total number of HiRDB files comprising an RDAREA (leave some extra room). You can use the pdfstatfs command to check the number of HiRDB files inside each HiRDB file system area. If you plan to add HiRDB files, specify a value by taking them into account. HiRDB files are added when RDAREAs are added, reinitialized, or extended.
  • For a HiRDB parallel server configuration, the value of this operand is applied to each back-end server. For example, if 100 is specified for this operand, a maximum of 100 HiRDB files can be created in each back-end server. Therefore, use the largest number of HiRDB files among the back-end servers as the guideline when specifying a value for this operand.
  • If you are using the inner replica facility, include the number of the HiRDB files comprising replica RDAREAs as well.
  • If you are using shared RDAREAs, also add the number of HiRDB files comprising the shared RDAREAs used by reference-only back-end servers.
Notes
  • Do not specify an unnecessarily large value for this operand. Increasing the specification value of this operand increases the size of the shared memory used by HiRDB. If a shortage occurs in the shared memory, HiRDB might not be able to start.
  • Do not specify this operand for a utility special unit.
Effects on individual estimation formulas
If the value of the pd_max_file_no operand is changed, the following estimation formulas are affected:
HiRDB Version 9 Installation and Design Guide:
  • Formulas for shared memory used by a unit controller under Estimating the memory size required for a HiRDB single server configuration
  • Formulas for shared memory used by a unit controller under Estimating the memory size required for a HiRDB parallel server configuration
  • Determining the value of S under Determining the size of status files
  • Formula 2 under Formulas for shared memory used by a single server
  • Formula 2 under Formulas for the size of the shared memory used by a dictionary server
  • Formula 2 under Formulas for the size of the shared memory used by a back-end server
119) pd_rdarea_warning_point = segment-usage-ratio-1[,segment-usage-ratio-2[,segment-usage-ratio-3]]
~<unsigned integer>((0-100) ) (%)
Specifies triggers for issuing warning messages concerning RDAREA segment usage and area usage by HiRDB files that can be extended. Each trigger is specified as a percentage. The warning messages are KFPH00211-I or KFPA12300-I for RDAREA segment area usage, and KFPH22037-W for area usage by HiRDB files that can be extended.
When auto-extension of RDAREAs is not used
The KFPH00211-I or KFPA12300-I warning message about segment usage is issued.
The following table shows the relationship between specification of this operand and the segment usage ratio warning message for the different types of RDAREAs.
RDAREA typepd_rdarea_warning_point not specifiedpd_rdarea_warning_point specified
  • Master directory RDAREA
  • Data directory RDAREA
No message is output.Segment use has started that makes the segment usage ratio equal to the value specified for all segments of the applicable RDAREA.
  • Data dictionary RDAREA
  • User RDAREA
Segment use has started that makes the segment usage ratio 80%, 90%, or 100% in terms of the relative position of the segment in the last file of the applicable RDAREA.Segment use has started that makes the segment usage ratio equal to the value specified for all segments of the applicable RDAREA.
  • LOB RDAREA
Segment use has started that makes the segment usage ratio 80%, 90%, or 100% in terms of the relative position of the segment in the last file of the applicable RDAREA.Use of the segment corresponding to the specified relative position has started.
When auto-extension of RDAREAs is used
The KFPH22037-W warning message about area usage by HiRDB files that will be extended automatically is issued.
When RDAREA auto-extension is being used, the message output depends on whether this operand is specified:
Operand specificationpd_rdarea_warning_point omittedpd_rdarea_warning_point specified
Message output conditionThe extended HiRDB file area allocated exceeds a usage ratio of 80%, 90%, or 100%.The extended HiRDB file area allocated exceeds a specified usage ratio.
The HiRDB file area usage that will be extended automatically is the percentage used of the maximum area that can be added. It is determined as follows.

Area usage by file that will be extended automatically (%) = A[Figure] B[Figure] 100

  • For a HiRDB file system area specified by the pdfmkfs -e command or a HiRDB file system area in a character special file that is specified by pdfmkfs -a:
    A: Capacity of HiRDB file that will be extended automatically
    B: Maximum capacity that can be extended automatically = min {(A + C), 64 GB}
    C: Free space in the HiRDB file system that contains the HiRDB file that will be extended automatically
  • For a HiRDB file system area in a regular file that is specified by the pdfmkfs -a command:
    A: Capacity of HiRDB file that will be extended automatically
    B: Maximum capacity that can be extended automatically = min {(A + C), 64 GB}
    C: Free space in the HiRDB file system that contains the HiRDB file that will be extended automatically + Free space on the disk where the HiRDB file system that contains the HiRDB file that will be extended automatically is located
If the number of used extents of the automatically extended HiRDB file is 20, 22, 24, or higher, a warning message (KFPH22038-W) for the number of used extents of the HiRDB file that will be extended automatically will be displayed.
The following table shows specification examples of this operand:
Message output conditionOperand specification value
To output a message when segment use has started that makes the segment usage ratio 80%, 90%, or 100% of all the segments in the applicable RDAREApd_rdarea_warning_point = 80,90,100
To output a message when segment use has started that makes the segment usage ratio 50% or 90% of all the segments in the applicable RDAREApd_rdarea_warning_point = 50,90
To not output a segment usage warning messagepd_rdarea_warning_point = 0
Operand rules
  • Up to three values can be specified.
  • When the same value is specified more than once, only one message will be output for that value.
  • When 0 is specified for all three values, no segment usage warning message is output.
  • When 0 and a non-zero numeric value are both specified, the non-zero numeric value is used as the trigger for message output.
  • The KFPA12300-I segment usage warning message is issued only if YES is specified in the PDEXWARN client environment definition.
Note
Specification of this operand is not applicable to the database initialization utility (pdinit) or the database recovery utility (pdrstr).The message content is the same regardless of whether this operand is specified.
120) pd_rdarea_warning_point_msgout = Y | N
Specifies whether an RDAREA segment usage warning message (KFPH00211-I or KFPA12300-I) is to be issued.
Y: Issues an RDAREA segment usage warning message according to the specification of pd_rdarea_warning_point.
N: Does not issue an RDAREA segment usage warning message.
Specification guidelines
For guidance about the RDAREA segment usage warning message specifications, see Environment settings under Free space reusage facility in the HiRDB Version 9 Installation and Design Guide.
Notes
In the following cases, a segment usage warning message can be issued, even though N is specified in this operand.
  • When data is being stored by the database creation utility (pdload) or the database organizing utility (pdrorg)
  • When data is being stored in tables that do not use the free space reusage facility
  • When data is being stored in an index
  • When configuration columns of tables that use the free space reusage facility include variable-length columns, and updates that increase the data size occur frequently
  • When the free space reusage facility can no longer be used because the number of table definitions has exceeded the value specified in the pd_assurance_table_no operand
Relationship to other operands
This operand is related to the following operands:
  • pd_rdarea_warning_point
  • pd_assurance_table_no
121) pd_rdarea_extension_timing = use | nouse
  • 0904 compatibility mode: <<nouse>>
Specifies the timing for performing RDAREA automatic extension when the RDAREA auto-extension facility is being used.
use:
Extend automatically the number of free segments in an RDAREA when the number of available free segments becomes equal to or less than the number of segments to be extended in a single auto-extension. For example, when the number of segments extended in a single auto-extension is 50, auto-extension will occur whenever the number of free segments drops to 50 or fewer.
nouse:
Extend automatically only when there are no free segments left in an RDAREA and new segments cannot be secured.
Specification guidelines
Advantages and disadvantages of the specification options are described below.
ItemOperand specification
usenouse
AdvantageEmpty segments (segments in the extended portion) can be used until addition of data causes an insufficient-pages error, even if auto-extension cannot be performed.# During this period, the RDAREA's status can be analyzed and reorganized to reflect its use status, the RDAREA can also be expanded, thereby minimizing the impact on operations.Because auto-extension is performed only after there are no more empty segments, storage efficiency is improved when multiple HiRDB files are being created in the HiRDB file system area.
DisadvantageStorage efficiency might be degraded when multiple HiRDB files are being created in the HiRDB file system area.If auto-extension cannot be performed,# there are no empty segments, and operations might stop if the addition of data causes an insufficient-pages error.

#: Auto-extension cannot be used in the following cases:

  • The mode that expands HiRDB files in the HiRDB file system area (pdfmkfs -e command) is being used, but there is no free space in the HiRDB file system area.
  • The mode that expands the HiRDB file system area automatically (pdfmkfs -a command) is being used, but there is no free disk space or the HiRDB files have reached their maximum size (64 GB).
  • Auto-extension has stopped because RDAREA backup has shut down.

We recommend that you specify use in this operand to minimize the impact on operations.
122) pd_rdarea_expand_format = Y | N
Specifies whether to initialize the extended area when an RDAREA is extended automatically. With normal files, this operand can be specified for HiRDB file system areas that are extended through specification of the -a option in the pdfmkfs command.
Y: Initializes the extended area.
N: Does not initialize the extended area
Specification guidelines
When Y is specified in this operand, disk I/Os occur for initializing the extended area. Because such disk I/O processing requires time, the following transactions can be impacted by delays:
  • Transactions that update segments being initialized
  • Transactions that allocate new segments in the extended area
We recommend that you specify N for this operand in most environments.
However, this recommendation depends on the specifications for allowing monitoring of capacity in RDAREAs and other locations. For details, see Applying RDAREA automatic extension in the HiRDB Version 9 System Operation Guide.
Note
When Y is specified in this operand, updating transactions that access an RDAREA being auto-extended might take longer to execute, which means that transactions might also be canceled by one of the following factors:
  • Lock release wait time is exceeded
  • HiRDB client maximum wait time or maximum utility execution time is exceeded
To avoid such types of transaction cancellation, estimate in seconds (as shown in the guide below) the processing time that will be needed for initialization of extended areas, considering the RDAREA configuration information, the I/O performance of the disk where the HiRDB file system area is allocated, and the operands listed in Relationship to other operands.
  • Page size
  • Segment size
  • Extended segment count
    Guide to processing time requirement for initialization of extended areas (sec) =
    page-size (bytes) [Figure] (segment-size[Figure] extended-segment-count + [Figure]) [Figure] I/O-performance-of-disk-
    where-HiRDB-file-system-area-is-allocated (bytes/sec)
    [Figure]: Page count of the creating directory at time of extension
    • For data dictionary, user, or registry RDAREAs
      [Figure]d[Figure] b[Figure] + [Figure]d[Figure] f[Figure]
    • For LOB RDAREAs
      [Figure]S[Figure] 64,000[Figure][Figure] 96
    d: Extended segment count
    b: [Figure](P - 20) [Figure] ([Figure]S[Figure] 32[Figure][Figure] 8 + 56)[Figure]
    f: [Figure](125 [Figure] P) [Figure] (16 [Figure] b)[Figure][Figure] b
    P: Page size
    S: Segment size
Relationship to other operands
This operand is related to the following operands:
  • PDCWAITTIME
  • pd_lck_wait_timeout
  • pd_cmd_exec_time
  • pd_utl_exec_time
123) pd_rdarea_open_attribute_use = Y | N
Specifies whether to use the DEFER or SCHEDULE attribute as the RDAREA opening trigger.
Y: Uses the DEFER or SCHEDULE attribute.
N: Does not use the DEFER or SCHEDULE attribute.
When this operand is omitted or when N is specified, the RDAREA opening trigger attribute is always INITIAL. Therefore, even if the DEFER or SCHEDULE attribute is specified as the RDAREA opening trigger in the operand or utility described as follows, the specification is invalid.
  • pd_rdarea_open_attribute
  • Database initialization utility
  • Database structure modification utility
Notes
  • When Y is specified, HiRDB requires a larger shared memory. Consequently, a shared memory shortage might occur, preventing the HiRDB system from starting.
  • If the rapid system switchover facility, standby-less system switchover (1:1) facility or standby-less system switchover (effects distributed) facility is used, Y is assumed for this operand. Because the size of the shared memory used by the server increases as a result, re-estimate the shared memory size. For the formula for estimating the size of the shared memory used by a server, see the HiRDB Version 9 Installation and Design Guide.
Effects on individual estimation formulas
If the value of the pd_rdarea_open_attribute_use operand is changed, the following estimation formulas are affected:
HiRDB Version 9 Installation and Design Guide:
  • Processes started by a HiRDB single server configuration
  • Processes started by a HiRDB parallel server configuration
  • Formulas for shared memory used by a single server
  • Formulas for the size of the shared memory used by a back-end server
124) pd_rdarea_open_attribute = INITIAL | DEFER | SCHEDULE
Specifies the standard value for the RDAREA opening trigger attribute.
The attribute specified by this operand is assumed for RDAREAs for which open attribute is not specified by the database initialization utility or the database structure modification utility.
For System RDAREAs, INITIAL is always assumed.
Specification guidelines
  • Specify the same attribute for the RDAREAs in the same HiRDB file system area. If different attributes are specified, the expected result might not be obtained.
  • The following table lists the opening/closing triggers for each attribute and their advantages and disadvantages.
    AttributeInitial statusOpening triggerClosing triggerAdvantageDisadvantage
    INITIALOpen
    • HiRDB startup
    • pdopen command execution
    pdclose command executionFast execution from the first SQL.System startup takes time.
    DEFERClosed
    • Initial access to RDAREA
    • pdopen command execution
    pdclose command execution
    • Fast system startup.
    • Regular SQLs are executed at high-speed, as well as after the initial access.
    First access to each RDAREA takes time.
    SCHEDULEClosed
    • Initial RDAREA access inside transaction
    • pdopen command execution
    • Transaction termination
    • pdclose command execution
    • Fast system startup.
    • Avoids concentrated file opening.
    • Initial access to RDAREA results in high workload for a transaction.
    • If a hybrid method is used for processing Real Time SAN Replication, an overhead of at least 2 seconds occurs for each transaction during its termination.
  • The following table shows the operation mode appropriate to each attribute:
    AttributeAppropriate operation mode
    INITIALHiRDB file system area is opened at system startup and keeps the RDAREA information resident in the memory. The HiRDB file system area is opened also during the initial RDAREA access. Because the RDAREA information is not re-created in this case, high-speed operation can be performed from the first SQL.
    The initial status of the RDAREA at system startup is open, and this RDAREA status will not change unless an operation command is entered subsequently, except during shift to an error shutdown.
    This attribute is recommended unless an unusual operation mode is used.
    When this attribute is used, a closed RDAREA cannot be accessed.
    DEFERHiRDB file system area is not opened at system startup. Instead, the HiRDB file system area is opened during the initial RDAREA access and the RDAREA information is kept resident in the memory. In the second and subsequent accesses, the processing beyond opening of the HiRDB file system area is not performed, and thus high-speed operations can be achieved.
    The initial status of the RDAREA at system startup is closed, and each RDAREA is opened during the initial access to that RDAREA. The RDAREA status will not change unless an operation command is entered subsequently, except during a shift to an error shutdown.
    Specify this attribute if you want to avoid cases in which a large number of HiRDB file system areas are opened concurrently or if you want to shorten the time required for starting HiRDB.
    When HiRDB is restarted, the RDAREAs to be recovered are opened during recovery processing.
    When this attribute is used, a closed RDAREA can also be accessed.
    SCHEDULEHiRDB file system area is not opened at system startup. Instead, after HiRDB startup, the HiRDB file system area is opened during the initial RDAREA access in each transaction and the RDAREA information is kept resident in the memory. When a transaction terminates, the HiRDB file system area opened in that transaction is closed. Thereafter, the processing beyond opening will also be performed during the initial access to an RDAREA whenever the transaction is changed, and thus the workload required for transactions will increase.
    The initial status of the RDAREA at system startup is closed, and the RDAREA is kept open only during the transaction for the accessed RDAREA. When a transaction terminates, all RDAREAs that were opened in the transaction are closed.
    If the pdopen command is entered, these RDAREAs can be kept open until they are closed by the next shutdown. It is also possible to use other operation commands to change the status of the RDAREAs. If an error is detected, an error shutdown occurs.
    This attribute is recommended when it is necessary to compensate for many HiRDB file system areas being opened simultaneously, or when it is necessary to reduce the HiRDB system startup time.
    When HiRDB is restarted, the RDAREAs to be recovered are opened during recovery processing and closed after the completion of the recovery processing.
    When this attribute is used, a closed RDAREA can also be accessed.
    If a hybrid method is used for processing Real Time SAN Replication, the system waits for synchronization with a database at the remote site. During transaction termination, an overhead of at least number-of-RDAREAs-accessed[Figure] 2 (seconds) might occur.
Note
The following table describes the notes related to the use of the rapid system switchover facility, standby-less system switchover (1:1) facility, and standby-less system switchover (effects distributed) facility.
Facility usedNotes
Rapid system switchover facilityA standby unit that is targeted by the rapid system switchover facility has not opened RDAREAs while it is in a standby state. Furthermore, to minimize the time required for system switchover, the standby unit opens only those RDAREAs that are necessary for full recovery when system switchover occurs, and does not open other RDAREAs. Therefore, the RDAREA opening trigger for the standby system cannot be INITIAL. The INITIAL attribute of RDAREAs is changed to DEFER.
Standby-less system switchover (1:1) facilityTo minimize the time required for system switchover, the standby-less system switchover (1:1) facility opens only those RDAREAs that are necessary for full recovery when system switchover occurs, and does not open other RDAREAs. Therefore, the opening trigger for the RDAREAs in the normal BES or alternate portion is as follows:
  • When system switchover occurs, the opening trigger for the RDAREAs in the alternate portion is SCHEDULE.
  • When the error is corrected and the system switches back to the normal BES, the opening trigger for RDAREAs with the INITIAL or DEFER attribute under the normal BES is changed to DEFER. RDAREAs with the SCHEDULE attribute retain this attribute.
Standby-less system switchover (effects distributed) facilityTo minimize the time required for system switchover, the standby-less system switchover (effects distributed) facility opens only those RDAREAs that are necessary for full recovery when system switchover occurs, and does not open other RDAREAs. Therefore, when system switchover occurs, the opening trigger for RDAREAs with the INITIAL attribute under the guest BES is changed to DEFER.
The following table shows the relationships between the opening trigger for RDAREAs and the rapid system switchover facility, standby-less system switchover (1:1) facility, and standby-less system switchover (effects distributed) facility.
Conditionpd_rdarea_open_attribute_use specification value
NY
RDAREA opening trigger
INITIALDEFERSCHEDULE
System switchover facility not used#1INITIALINITIALDEFERSCHEDULE
Standby system switchover facility#1Rapid system switchover facility not used
Rapid system switchover facilityRunning system
Standby systemDEFER#2DEFERDEFERSCHEDULE
Standby-less system switchover (1:1) facility#1Accepting portion (other than the alternate portion of the alternate BES unit)Running systemINITIAL#2, #3INITIAL#3DEFERSCHEDULE
Standby systemDEFER#2DEFERDEFERSCHEDULE
Alternate portionRunning systemSCHEDULE#2SCHEDULESCHEDULESCHEDULE
Standby system
Standby-less system switchover (effects distributed) facility#1Normal start or restartINITIALINITIALDEFERSCHEDULE
Restart due to system switchoverDEFER#2DEFERDEFERSCHEDULE
#1: The opening trigger for the System RDAREA is INITIAL.
#2: It is assumed that pd_rdarea_open_attribute_use = Y is specified.
#3: Changed to DEFER after a restart.
125) pd_shared_rdarea_use = Y | N
This operand is applicable only to a HiRDB parallel server configuration.
Specifies whether to use a shared RDAREA.
Y: A shared RDAREA is used.
N: A shared RDAREA is not used.
Notes
  • If you omit this operand (or specify N for it) even though a shared RDAREA has been defined, HiRDB cannot be normally started.
  • Specifying Y for this operand allocates the same shared memory block as that allocated when commit is specified for the pd_dbsync_point operand.
Effects on individual estimation formulas
If the value of the pd_shared_rdarea_use operand is changed, the following estimation formula is affected:
HiRDB Version 9 Installation and Design Guide:
  • Formulas for the size of the shared memory used by a back-end server
126) pd_db_access_error_action = dbhold | unitdown
Specifies the action to be taken when a file access error occurs while accessing an RDAREA.
If an RDAREA is being accessed, and a process other than HiRDB has opened with an exclusive lock a file in the same HiRDB file system area that constitutes the RDAREA being accessed, a file access error results (error code -1540#). This operand specifies the action to be taken in response to such a file access error.
For the master directory RDAREA, the unit is shut down without resulting in the RDAREA in error shutdown status, even if the facility for taking a unit down when a file access error is detected is not used. When the unit is shut down, the KFPH23040-I message is issued.
#: This error code is also returned when a mistake is made in setting the HiRDB file system area access authority.
dbhold:
Shut down the RDAREA being accessed when a file access error occurs.
unitdown:
Issue the KFPH23040-I message when a file access error occurs, and then apply the facility for taking a unit down when a file access error is detected. For details about the facility for taking a unit down when a file access error is detected, see the HiRDB Version 9 System Operation Guide.
Application criterion
Normally, this operand need not be specified.
Consider specifying this operand if a process other than HiRDB might perform an operation that accesses a HiRDB file system area made up of RDAREAs while HiRDB is operating, and that process is able to obtain an exclusive lock on the HiRDB file system area.
Notes
  • When a file access error occurs and unitdown has been specified, the RDAREA being processed might be subject to error shutdown in the following cases:
    [Figure]A UAP or utility is being executed in the pre-update log acquisition mode or the no-log mode.
    [Figure]A UAP or utility is being run in a user LOB RDAREA that has been placed in the no-log mode by specification of NO in the RECOVERY operand of CREATE TABLE.
    If you use the facility for taking a unit down when a physical error is detected, avoid running these operations, if possible. If you need to run these operations, make a backup prior to running the UAP or utility so that recovery from an RDAREA error shutdown can be performed. For details about making back-ups, see the HiRDB Version 9 System Operation Guide.
  • When the log-only synchronous method Real Time SAN Replication facility is being used, this operand is ignored even if unitdown is specified when HiRDB at the log application site was starting.
  • During recovery processing by the database recovery utility (pdrstr), the unit will not be shut down, even if unitdown is specified. In this case, re-execute pdrstr to recover.
Relationship to other operands
  • pd_mode_conf operand
  • pd_db_io_error_action operand
  • pd_db_hold_action operand
If unitdown is specified in more than one of the pd_db_io_error_action, pd_db_access_error_action, and pd_db_hold_action operands, the operand value that takes effect is determined in the following order:
  1. pd_db_io_error_action operand
  2. pd_db_access_error_action operand
  3. pd_db_hold_action operand
If more than one RDAREA input/output, file access, or physical error has occurred, determine the error that caused unitdown based on the above priority. In addition, see the message that is issued.
127) pd_db_hold_action = dbhold | unitdown
Specifies the action to be taken when a physical error (that results in shutdown with I/O error occurred or open error occurred displayed as the reason in the KFPH00306-E message) occurs while accessing an RDAREA. If this is the master directory RDAREA, a unitdown, not error shutdown, occurs, even if the facility for taking a unit down when a physical error is detected is not used. For details about the facility for taking a unit down when a physical error is detected, see the HiRDB Version 9 System Operation Guide.
dbhold:
Shut down the RDAREA being accessed when a physical error occurs.
unitdown:
Issue the KFPH23047-I message when a physical error occurs, and then use the facility for taking a unit down when a physical error is detected.
Application criterion
Normally, this operand does not need to be specified.
Consider specifying this operand if your system has a small number of RDAREAs and the entire application is shut down if one of the RDAREAs is placed in error shutdown status.
If you specify unitdown, also specify MANUAL2 in the pd_mode_conf operand.
Notes
  • If a physical error occurs while accessing an RDAREA and the KFPH00307-E is output, resulting in the RDAREA being placed in command hold status, the unit will not be shut down event if unitdown is specified in the pd_db_hold_action operand.
  • If a physical error occurs and unitdown has been specified, the RDAREA being processed might be subject to error shutdown in the following cases:
    [Figure]A UAP or utility is being executed in the pre-update log acquisition mode or the no-log mode.
    [Figure]A UAP or utility is being run in a user LOB RDAREA that has been placed in the no-log mode by specification of NO in the RECOVERY operand of CREATE TABLE.
    If you use the facility for taking a unit down when a physical error is detected, avoid running these operations, if possible. If you need to run these operations, make a backup prior to running the UAP or utility so that recovery from an RDAREA error shutdown can be performed. For details about making back-ups, see the HiRDB Version 9 System Operation Guide.
  • When the log-only synchronous method Real Time SAN Replication facility is being used, this operand is ignored even if unitdown is specified when HiRDB at the log application site was starting.
  • During recovery processing by the database recovery utility (pdrstr), the unit will not be shut down, even if unitdown is specified. In this case, re-execute pdrstr to recover.
Relationship to other operands
This operand is related to the following operands:
  • pd_mode_conf operand
  • pd_db_io_error_action operand
  • pd_db_access_error_action operand
If unitdown is specified in more than one of the pd_db_io_error_action, pd_db_access_error_action, and pd_db_hold_action operands, the operand value that takes effect is determined in the following order:
  1. pd_db_io_error_action operand
  2. pd_db_access_error_action operand
  3. pd_db_hold_action operand
If more than one RDAREA input/output, file access, or physical error has occurred, determine the error that caused unitdown based on the above priority. In addition, see the message that is issued.