• hdeid=source-Datareplicator-identifier
  ~ <hexadecimal> ((00-FF))
  Specify the identifier to be used to identify this source system. If the source HiRDB is a parallel server, one source system includes the source Datareplicator under the system manager and the source Datareplicators under all back-end servers. Thus, if the source HiRDB is a parallel server, you need one source Datareplicator identifier per source system.
  
• extdef="extraction-environment-definition-filename"
  ~ <[pathname/]filename of 1-64 bytes> <<extraction environment definition used when all operands are omitted>>
  Specify the filename of the extraction environment definition that defines the operating environment for extraction processing, as an absolute or relative pathname. If you specify a relative pathname, Datareplicator assumes $HDEPATH/relative-pathname as the absolute pathname. In this case, specify the pathname so that the final absolute pathname will not exceed 125 characters. If you omit the extdef operand, Datareplicator assumes the extraction environment definition that is used when all operands are omitted.
  
• sendcontrol=nodemst|sendmst
  Specify the processing method to be used to send extracted data.
  

  nodemst

  Start a transmission process for each destination and execute transmission processing for all destinations in parallel. Use the sendidxx and senddefxx operands to specify a maximum of 64 destinations.

  We recommend that you specify nodemst for an application that has only a few destinations with a high volume of traffic.

  sendmst

  Use the transmission master process to control the number of transmission processes. The transmission master process schedules transmission processing for the number of transmission processes specified in the sendprocnum operand. Use the sendidxxxx and senddefxxxx operands to specify a maximum of 4,096 destinations.

  We recommend that you specify sendmst for an application that has many destinations with a low volume of traffic.

• sendid01=target-identifier[...[sendid64=target-identifier]]
  ~ <symbolic name of 1-8 characters or **>
  Specify the names that identify the destinations of update information (target identifiers). These target identifiers are used to identify the target systems subject to processing when transmission statements or commands are specified in the extraction definition. These target identifiers are also displayed in messages.
  When you specify nodemst in the sendcontrol operand or you omit the sendcontrol operand, you must specify the sendidxx operand. This operand is effective only when nodemst is specified in the sendcontrol operand or the sendcontrol operand is omitted.
  You must specify the sendid01 to sendid64 operands consecutively in ascending order beginning with sendid01. If they are not consecutive or in ascending order, Datareplicator uses only those operands that are specified consecutively in ascending order from the beginning.
  When you specify ** in place of a target identifier, Datareplicator assumes an absent number, and the corresponding senddefxx operand is ignored (transmission to a target will not occur). You can specify ** in place of any number of target identifiers; if you specify ** for all target identifiers, Datareplicator assumes that no target identifiers at all are specified.
  In the case of Windows Datareplicator, you can specify a maximum of 63 target identifiers corresponding to transmission environment definition files; an error will result if sendid64 is specified.
  
• senddef01="transmission-environment-definition-filename"[... [ senddef64="transmission-environment-definition-filename"]]
  ~ <[pathname/]filename of 1-64 bytes> <<transmission environment definition used when all operands are omitted>>
  Specify the filenames of the transmission environment definitions that define operating environments for transmission processing, as absolute or relative pathnames. If you specify a relative pathname, Datareplicator assumes $HDEPATH/relative-pathname as the absolute pathname. In this case, specify the pathname so that the final absolute pathname will not exceed 125 characters.
  This operand is effective only when nodemst is specified in the sendcontrol operand or the sendcontrol operand is omitted. Each specified filename must be unique in the source system.
  You must specify the senddef01 to senddef64 operands so that they correspond to the sendid01 to sendid64 operands. If there is no sendidnn operand corresponding to a senddefxx operand, that senddefxx operand will be ignored. On the other hand, if there is a sendidnn operand but no corresponding senddefxx operand, Datareplicator assumes the transmission environment definition that is used when all operands are omitted.
  In the case of Windows Datareplicator, you can specify a maximum of 63 transmission environment definition files corresponding to target identifiers; an error will result if senddef64 is specified.
  
• sendid0001=target-identifier[... [sendid4096=target-identifier]]
  ~ <symbolic name of 1-8 characters>
  Specify the names that identify the destinations of update information (target identifiers). These target identifiers are used to identify the target systems subject to processing when transmission statements or commands are specified in the extraction definition. These target identifiers are also displayed in messages.
  When you specify sendmst in the sendcontrol operand, you must specify the sendidxxxx operand. This operand is effective only when sendmst is specified in the sendcontrol operand.
  You must specify the sendid0001 to sendid4096 operands consecutively in ascending order beginning with sendid0001. If they are not consecutive or in ascending order, Datareplicator uses only those operands that are specified consecutively in ascending order from the beginning.
  When you specify ** in place of a target identifier, Datareplicator assumes an absent number, and the corresponding senddefxxxx operand is ignored (transmission to a target will not occur). You can specify ** in place of any number of target identifiers; if you specify ** for all target identifiers, Datareplicator assumes that no target identifiers at all are specified.
  
• senddef0001="transmission-environment-definition-filename"[... [ senddef4096="transmission-environment-definition-filename"]]
  ~ <[pathname/]filename of 1-125 bytes> <<transmission environment definition used when all operands are omitted>>
  Specify the filenames of the transmission environment definitions that define operating environments for transmission processing, as absolute or relative pathnames. If you specify a relative pathname, Datareplicator assumes $HDEPATH/relative-pathname as the absolute pathname. In this case, specify the pathname so that the final absolute pathname will not exceed 125 characters.
  This operand is effective only when sendmst is specified in the sendcontrol operand. Each specified filename must be unique in the source system.
  You must specify the senddef0001 to senddef4096 operands so that they correspond to the senddid0001 to sendid4096 operands. If there is no sendidxxxx operand corresponding to a senddefxxxx operand, that sendidxxxx operand will be ignored. On the other hand, if there is a sendidxxxx operand but no corresponding senddefxxxx operand, Datareplicator assumes the transmission environment definition that is used when all operands are omitted.
  
• errfile_unique=true|false
  In the case of a HiRDB/Parallel Server system that employs the mutual system switchover configuration, specify whether the names of the extraction node master error information files are to be expanded so that they will be unique throughout the source Datareplicator. This specification also applies to the names of the extraction node master trace files.
  

  true

  Datareplicator is to add the applicable node's host name to the error information filenames and activity trace filenames so that the names are unique in the source Datareplicator.

  Extraction node master error information filenames: errfile1_host-name, errfile2_host-name

  Extraction node master trace filenames: exttrc_host-name.trc1, exttrc_host-name.trc2

  false

  Datareplicator is to use the same names for the error information files (errfile1 and errfile2) and the extraction node master trace files (exttrc.trc1 and exttrc.trc2) at all nodes of the HiRDB/Parallel Server.

  Specify true in the errfile_unique operand only when Datareplicator is used in a HiRDB/Parallel Server system that employs the mutual system switchover configuration. Otherwise, specify false or omit the errfile_unique operand.

• errfilesz=maximum-error-information-file-size
  ~ <unsigned integer> ((1-32767)) <<16>> (KB)
  Specify a maximum size for the extraction master error information files and extraction node master error information files. For details about the sizes for the extraction master error information files and extraction node master error information files, see 4.6.8 Designing the source Datareplicator's resources.
  
• syslogout=true|false
  Specify whether the information in the extraction master error information files is to be output also to the syslog file. If you are using JP1, you must output this information to the syslog file to implement automatic operation.
  

  true

  Output the information in the extraction master error information files also to the syslog file.

  false

  Do not output the information in the extraction master error information files to the syslog file.

  Note that the KFRB00501-I and KFRB00505-I messages are still output to the syslog file even when false is specified.

• syslog_message_suppress=message-number[,message-number] ...
  Specify the numbers of the messages whose output to the syslog file (event log for Windows) is to be suppressed.
  
  • Specify a maximum of 64 message numbers.
  • If you specify the same message number more than once, only the first specification of the number is effective.
  • Datareplicator does not output any of the messages specified in this operand, regardless of a message's significance level (E, W, I, or Q).
  • Messages that are output only to the syslog file (or event log) are always output regardless of whether you specify this operand.
• dblocale=sjis|euc|utf-8
  Specify the character code system supported by the source Datareplicator. Datareplicator assumes that the extraction definitions are coded in the code system specified in this operand when it analyzes extraction definitions. Specify the character code system used at the source HiRDB.
  

  sjis

  Use the JIS8/Shift JIS code system.

  euc

  Uses the EUC code system.

  utf-8

  Uses the utf-8 code system.

  The default value for this operand depends on the OS being used, as shown in the following table:
  

  OS	Default value
  HP-UX	sjis
  AIX
  Windows
  Solaris	euc
  Linux

  In the case of Windows Datareplicator, you must specify either sjis or utf-8, because the source Datareplicator supports only the JIS8/Shift JIS and Unicode systems. Specification of any other value in the dblocate operand will result in an error.
  
• msglocale={ english|sjis-japanese|euc-japanese }
  Specify the character code set to be used for messages issued by the source Datareplicator.
  For Windows Datareplicator, specify either english or sjis-japanese; specifying euc-japanese will result in an error.
  

  english

  Output messages in English.

  sjis-japanese

  Output messages in Japanese using the JIS8/Shift JIS code system.

  euc-japanese

  Output messages in Japanese using the EUC code system.

• watchintvl=error-monitoring-interval
  ~ <unsigned integer> ((10-32767)) <<60>> (seconds)
  Specify the interval at which the extraction master process is to monitor extraction and transmission processing for errors, or the interval at which a delay between extraction and transmission processes is to be monitored.
  
• cmwaittime=communication-wait-time
  ~ <unsigned integer> ((10-32767)) <<60>> (seconds)
  Specify the wait time for communication to occur between the extraction master process and the extraction node master process.
  
• mstservice=master-to-node-master-communication-service-name
  ~ <identifier of 1-64 characters> <<hdemaster>>
  Specify the service name to be used to add to the services file a communication entry between the source Datareplicator's extraction master process and extraction node master process.
  For Windows Datareplicator, the master-to-node master communication service name is always hdenmserv, and this name is specified in the services file after Datareplicator installation is completed. Therefore, you need not specify the mstservice operand in this case; Datareplicator will ignore this operand if it is specified for Windows Datareplicator.
  
• extinfonum=maximum-update-information-names-count
  ~ <unsigned integer> <<50>> ((1-4096))
  Specify the number of update information names that can be specified in the extraction definition. If this value is less than the number of update information names actually specified in the extraction definition, an error occurs during execution of the hdeprep command.
  
• syncterm=true|false
  Specify whether the source Datareplicator is to terminate normally automatically when the source HiRDB terminates normally.
  In the case of a system switchover configuration, all update information might not always be sent to the target system even when true is specified in this operand, because the order of the system switchover timing and the source Datareplicator's termination timing cannot be predicted. Therefore, do not specify true in this operand in the case of a system switchover configuration. If you perform planned system switchover after all update information has been sent to the target system, do so after using the hdestate command to check the status.
  

  true

  Automatically terminate the source Datareplicator normally when the source HiRDB terminates normally. All the following conditions must be satisfied:

  The source HiRDB's normal termination is detected.

  No transactions have occurred at the source HiRDB since detection of the source HiRDB's normal termination.

  The extraction process has completed extracting all update information from the system log into the extraction information queue file.

  The transmission process has completed sending all update information from the extraction information queue file to the target system.

  false

  Do not terminate the source Datareplicator normally when the source HiRDB terminates normally.

  • If true is specified, the source Datareplicator terminates itself after extracting all system log information from the system log file and sending all the extracted update information subject to transmission to the applicable target systems.
    A transmission process must have already started before the extraction process detects the source HiRDB's normal termination. Otherwise, the source Datareplicator terminates normally when the extraction process detects the source HiRDB's normal termination, in which case some update information might remain in the extraction information queue file. If this happens, Datareplicator sends the remaining update information from the extraction information queue file to the target systems the next time the transmission process is started by the hdestart or hdestart -s command.
    
  • If you will be using the facility for recovering the extraction information queue file, specify false. When true is specified, the facility for recovering the extraction information queue file cannot be used.
• termlevel={ normal|plan|both }
  When the source Datareplicator is specified to terminate normally automatically when the source HiRDB terminates, specify the HiRDB termination method that is to be subject to this synchronized termination. This operand is applicable only when true is specified in the syncterm operand.
  

  normal

  Terminate the source Datareplicator normally only when the HiRDB terminates normally.

  plan

  Terminate the source Datareplicator normally only when planned termination of the HiRDB occurs.

  both

  Terminate the source Datareplicator normally both when the HiRDB terminates normally and when planned termination of the HiRDB occurs.

• info_message_out=nosuppress|suppress
  Specify whether output of information-only messages to the syslog file (event log for Windows) and error information files is to be suppressed.
  

  nosuppress

  Do not suppress output of information-only messages.

  suppress

  Suppress output of information-only messages. The following message numbers are subject to this operand's specification:

  00551, 00552, 00553, 00554, 00555, 00557, 00581, 00583, 02031, 02032, 02033, 02034, 02036, 02037, 05001, 05002, 05008, 05012, 05013, 05018

• except_suppress=message-number[,message-number] ...
  ~ <5-digit unsigned integer>
  From among the messages whose output is suppressed because info_message_out=suppress is specified, specify the numbers of the messages whose output suppression is to be cancelled and which are to be output to the syslog and error information files. You can specify a maximum of 63 message numbers.
  Datareplicator ignores specification of a message number if it is not subject to output suppression. If you specify the same message number more than once, only the first specification is effective.
  This operand is applicable only when suppress is specified in the info_message_out operand.
  
• int_trc_lvl=activity-trace-collection-level[,activity-trace-collection-range]
  To change the items to be collected in activity trace files (extraction master trace files and extraction node master trace files), specify the appropriate value shown in the table below. If you omit this operand, Datareplicator collects only the information common to all facilities (the minimum requirement).
  If you specify na for the activity trace collection level, Datareplicator will not collect any activity trace information and will ignore any value specified for the activity trace collection range (however, Datareplicator still checks for syntax and range errors).
  
  • Value for the activity trace collection level
    

    Value	Information to be collected
    	Common information	Overview of performance	Details about performance
    p1	Y	Y	--
    p2	Y	Y	Y
    na	--	--	--
    int_trc_lvl operand omitted	Y	--	--

    Y: Collected.
    --: Not collected.
    Common information: Collective name for the general checkpoint information that indicates a point of change in start/termination information, error information, or process level.
    

  Note:

  When you specify p2, you must provide sufficient space for the activity trace files. Otherwise, important information might be deleted by the wraparound feature; in addition, trace collection might result in a large overhead workload.

  • Value for the activity trace collection range
    

    Value	Information to be collected
    	MST (control facility)	CAP (extraction facility)	SND (transmission facility)
    c1	Y	Y	--
    c2	Y	--	Y
    c3	Y	--	--
    nc	Y	#	#
    int_trc_lvl operand or activity-trace-collection-range omitted	Y	Y	Y

    Y: Collected.
    --: Not collected.
    #: Even if you specify nc in the extraction system definition, you can still collect activity trace information individually by specifying the int_trc_getv operand in the extraction environment definition and the transmission environment definition.
    

Guidelines for specifying the int_trc_lvl operand:

We recommend that you specify the int_trc_lvl operand as follows:

1. Production run
   For actual operations, we recommend that you omit the int_trc_lvl operand. If you cannot obtain reasonable performance with specification of the operand omitted, specify collection of activity trace information temporarily so that you can evaluate performance. In this case, specify p1 or p2 in the first parameter of the int_trc_lvl operand, nc in the second parameter, and the int_trc_getl or int_trc_getv operand in the specific facility (the one exhibiting low performance). If you specify p2 in the first parameter of the int_trc_lvl operand, specify a sufficient value in the int_trc_filesz operand.
   
2. Test run
   At the test stage, we recommend that you specify p1 in the first parameter of the int_trc_lvl operand and omit the second parameter. This enables you to execute the hdstrcedit command to view activity trace information in the event of a problem with performance. Use the command's execution results to tune the HiRDB table definitions and the number of import groups.
   If you need more detailed information (such as the unit price of SQL executions), change the second parameter to p2 in the int_trc_lvl operand. When you specify p2, specify a sufficient value in the int_trc_filesz operand (at least 1MB).
   

• int_trc_filesz=activity-trace-file-size
  ~ <unsigned integer> ((32-1048576)) <<128>> (KB)
  Specify the maximum storage size for an activity trace file (extraction master trace file or extraction node master trace file). Datareplicator ignores this operand when na is specified in the int_trc_lvl operand.
  We recommend that you specify a value that is a multiple of 32. Otherwise, Datareplicator rounds up the specified value to the nearest multiple of 32KB and uses that value as the maximum size for an activity trace file.
  In the source system, Datareplicator creates the activity trace files under the following names in the $HDEPATH directory (Datareplicator uses dual files with swapping and wrapping):
  

  Extraction master trace files

  $HDEPATH/msttrc.trc1 and $HDEPATH/msttrc.trc2

  Extraction node master trace files

  $HDEPATH/exttrc.trc1 and $HDEPATH/exttrc.trc2

  If you specify true in the errfile_unique operand, _host-name is added to the filename (that is, the filenames become exttrc_host-name.trc1 and exttrc_host-name.trc2).

• int_trc_rintvl=activity-trace-information-collection-interval
  ~ <unsigned integer> ((5-30000)) <<50>> (milliseconds)
  Specify the interval at which activity trace information is to be collected. If na is specified in the int_trc_lvl operand, Datareplicator ignores this operand if it is specified.
  You can minimize the amount of missed activity trace information by specifying a short activity trace information collection interval (specify a small value), but the number of monitorings per second increases, resulting in a higher CPU utilization factor. Specify a small activity trace information collection interval in the following cases:
  
  • There are missing entries in the activity trace information.
  • The specified activity trace collection level indicates a high level of collection information.
  • The frequency of Datareplicator activity is high (there is a large amount of extraction data).
  On the other hand, specify a large activity trace information collection interval in the following case:
  
  • The frequency of Datareplicator activity is low (there is a small amount of extraction data) and a reduction of CPU utilization by Datareplicator is desired.
• sendprocnum=maximum-transmission-processes-to-be-started
  UNIX Datareplicator: ~ ((1-4096)) <<1>>
  Windows Datareplicator: ~ ((1-63)) <<1>>
  Specify the maximum number of transmission processes that can be started by the transmission master process. This operand is applicable when sendmst is specified in the sendcontrol operand.
  
• smt_sendintvl=transmission-master-process-transmission-interval
  ~ ((0-1440)) <<5>>
  Specify the transmission master process's transmission interval. Use the smt_sendintvl_scale operand to specify the units of the transmission interval. This operand is applicable when sendmst is specified in the sendcontrol operand. If you specify 0, the transmission master process will send update information in units of transactions. If you want to reduce the time leading up to import processing, start tuning from one second. Specifying 0 might slow down the transmission processing.
  When sendmst is specified in the sendcontrol operand, Datareplicator ignores the sendintvl operand in the transmission environment definition and uses the value of this operand to execute transmission processing.
  
• smt_sendintvl_scale=minute|second
  Specify the units of the smt_sendintvl operand's value.
  

  minute

  Use minutes for the smt_sendintvl operand's value.

  second

  Use seconds for the smt_sendintvl operand's value.

This operand is applicable when the smt_sendintvl operand is specified.

• smt_editbufsize=update-information-editing-buffer-size
  ~ ((1-2097151)) <<300>> (KB)
  Specify the buffer size needed by the transmission process to edit the format of update information so that the target system can receive it when the transmission process is controlled by the transmission master process.
  This operand is applicable when sendmst is specified in the sendcontrol operand, in which case Datareplicator ignores the editbufsize operand in the transmission environment definition and uses the value of this operand to execute transmission processing. You can reduce the number of communications by specifying a large value in this operand. We recommend that you specify a value that is greater than the amount of data (amount of update information) that is written to the extraction information queue file within the amount of time specified in the smt_sendintvl operand.
  To determine the amount of data that is written to the extraction information queue file, execute periodically the hdestate command and measure the amount of increase in the offset information indicated by the Queue write position.
  
• smt_readbufnum=I/O-buffers-count-for-reading-update-information
  ~ ((1-255)) <<1>>
  Specify the number of I/O buffers to be used by the transmission master process to read update information from the extraction information queue file when the transmission process is controlled by the transmission master process.
  This operand is applicable when sendmst is specified in the sendcontrol operand, in which case Datareplicator ignores the readbufnum operand in the transmission environment definition and uses the value of this operand to execute transmission processing.
  
• smt_queue_read_wait_interval=transmission-process's-extraction-information-queue-file-read-interval
  ~ <unsigned integer> ((100-60000)) <<2000>> (milliseconds)
  Specify the interval after which the next read operation is to be started once the end of the extraction information queue file is detected when the transmission is controlled by the transmission master process.
  By specifying a small value in this operand, you can improve the spontaneity of transmission processing, because there is only a brief wait after the end of the extraction information queue file is detected. However, this might have an adverse effect on performance, because the CPU utilization factor increases. You need to take into account the frequency of update processing and CPU performance when specifying this value.
  This operand is applicable when sendmst is specified in the sendcontrol operand, in which case Datareplicator ignores the queue_read_wait_interval operand in the transmission environment definition and uses the value of this operand to execute transmission processing.
  
• file_dupenv=duplexing-definition-file-name
  ~ <filename of 1 to 125 bytes>
  Specify the absolute or relative path name of the duplexing definition file. If a relative path is specified, the system assumes that the path is relative to the source Datareplicator directory. If this operand is omitted, Datareplicator assumes that the duplexing function is not used.
  
• nodecontrol= unit|server
  Specify how to control the extraction node master process.
  

  unit

  Start the extraction node master process for each unit of the HiRDB subject to extraction processing.

  server

  Start the extraction node master process for each back-end server of the HiRDB subject to extraction processing.

  You must specify server if the HiRDB subject to extraction processing uses a standby-less system switchover (effects distributed) configuration.

  If you use the hdestart command to start the extraction master process, the HiRDB subject to extraction processing must be running so that start host information for the extraction node master process can be acquired from the HiRDB subject to extraction processing (if you use the hdestart_n command to start the extraction master process, there is no need for the HiRDB subject to extraction processing to be running).

  The following table shows the resources whose name and creation unit depend on this operand's value:
  

  Resource type	nodecontrol=unit specified		nodecontrol=server specified#
  	Name	Creation unit	Name	Creation unit
  Extraction node master error information file	errfile1, errfile2	2 for each node	errfile1_server-name, errfile2_server-name	2 for each server
  Extraction node master activity trace file	exttrc.trc1, exttrc.trc2		exttrc_server-name.trc1, exttrc_server-name.trc2

  #: The system ignores the errfile_unique operand in the extraction system definition and creates resources using the names shown in this table.
  
• node_connection_accept= true|false
  Specify whether a connection is to be established from the extraction node master process to the extraction master process. The extraction node master process establishes connection with the extraction master process at the following times:
  
  • When the extraction node master process is started by the hdestart_n command
  • When the extraction node master process retries connection establishment after it was disconnected from the extraction master process
  Specification of the node_connection_accept operand enables you to control the operation of the extraction master process and extraction node master process at the times noted above.
  The following describes how the operation of the extraction master process and extraction node master process is controlled by the node_connection_accept operand:
  

  Event	When true is specified	When false is specified
  Extraction node master process was started by the hdestart_n command.	Operation of the extraction master process Accepts a connection requests from the extraction node master process. Operation of the extraction node master process Connects to the extraction master process, and then starts replication.	Operation of the extraction master process Does not accept a connection request from the extraction node master process. Operation of the extraction node master process Terminates the process due to a timeout when the connection wait time specified in the hdestart_n command expires.
  Extraction node master process was disconnected from the extraction master process.	Operation of the extraction master process Waits for a connection request from an extraction node master process if the extraction master process is disconnected from all extraction node master processes. Operation of the extraction node master process Reconnects the extraction master process.	Operation of the extraction master process Terminates the process if the extraction master process is disconnected from all extraction node master processes. Operation of the extraction node master process Terminates the process after detecting the disconnection.

  The node_connection_accept operand value is effective only when server is specified in the nodecontrol operand. If the HiRDB subject to extraction processing uses a standby-less system switchover (effects distributed) configuration, you must specify true in this operand.
  When you specify true in the node_connection_accept operand, you must specify the connection_accept_hostname and connection_accept_service operands.
  
• connection_accept_hostname=extraction-node-master-process-connection-request-accepting-host-name
  ~ <identifier of 1-32 characters>
  Specify the host name used to accept a connection request from the extraction node master process when connection is to be established from extraction node master process to extraction master process.
  If you have specified true in the node_connection_accept operand, you must specify this operand.
  In the case of a standby-less system switchover (effects distributed) configuration, you must specify in this operand a host name that inherits IP addresses, because the connection request issued by the extraction node master process to the extraction master process that is started in the standby system after system switchover uses the same host name as the primary system.
  If the extraction node master process is to be started by the hdestart_n command, the host name specified in this operand must also be specified in the -x option as the name of the host that runs the extraction master process.
  Note that the host name specified in this operand must be added to the OS's hosts file.
  
• connection_accept_service=extraction-node-master-process-connection-request-accepting-service-name
  ~ <identifier of 1-64 characters>
  Specify the service name used to accept a connection request from the extraction node master process when connection is to be established from extraction node master process to extraction master process.
  If you specified true in the node_connection_accept operand, you must specify this operand.
  This service name (and port number) must be different from the service name (and port number) specified in the mstservice operand of the extraction system definition and the hdeservice operand of the transmission environment definition.
  If the extraction node master process is to be started by the hdestart_n command, the service name specified in this operand must also be specified in the -n option.
  Note that the service name specified in this operand must be added to the OS's services file.
  
• connection_accept_waittime=extraction-node-master-process-connection-request-wait-time
  ~ ((1-3600)) <<300>> (seconds)
  Specify the amount of time to wait for the extraction master process to accept a reconnection request from an extraction node master process when the extraction master process is disconnected from all extraction node master processes.
  If the time specified in this operand elapses before a connection request is received from any of the extraction node master processes, the extraction master process is terminated.
  This operand is applicable only when true is specified in the node_connection_accept operand.
  
• connection_retry_time=extraction-node-master-process-reconnection-processing-time
  ~((1-3600)) <<300>> (seconds)
  Specify the amount of time during which the extraction node master process is to attempt to reestablish connection with the extraction master process when the extraction node master process is disconnected from the extraction master process.
  If the time specified in this operand elapses before the extraction node master process can reconnect to the extraction master process, the extraction node master process is terminated.
  This operand is applicable only when true is specified in the node_connection_accept operand.
  
• node_syslogout= true|false
  Specify whether the information that is output to the extraction node master error information file is to be output also to the corresponding machine's syslog file.
  

  true

  Output to the syslog file the information that is output to the extraction node master error information file.

  false

  Do not output to the syslog file the information that is output to the extraction node master error information file.

  The Windows edition does not support this operand because whether error information for each node is to be output is determined by the syslogout operand; if this operand is specified in the Windows edition, it is ignored.
  

  Note:

  When true is specified in this operand, the system outputs all error information for each node. Specifying true is not recommended, because a large amount of information might be output (depending on the operating environment), thereby consuming a large amount of syslog file resources.

• send_counter_reset = true|false
  Specify whether the data transmission count is to be reset when the source system starts.
  

  true

  Reset the data transmission count when the source system starts.

  false

  Do not reset data transmission count when the source system starts.

• hirdb_audit_trail = all|none
  Specify whether an audit trail is to be collected if an operation subject to audit trail collection (audit event) occurs.
  An audit event is generated when the hdeevent or hdeprep command or the extraction master process is executed.
  

  all

  Collect an audit trail.

  none

  Do not collect an audit trail.

  If you want to suppress the overhead of having source Datareplicator processing collect an audit trail, we recommend that you specify none.
  
• resource_chk_err = continue|stop
  Specify the action to be taken if an error is detected when the source Datareplicator checks file integrity and file sizes during initialization or start processing.
  If you always perform initialization or start processing when no change has been made to the data linkage environment or operands, we recommend that you specify stop.
  

  continue

  Continue with the initialization or start processing if it's possible for data linkage to be resumed or if the user changed operands intentionally.

  stop

  Stop the initialization or start processing regardless of the nature of the error.

• recover_info_send = true|false
  Specify whether recovery information is to be sent to the target Datareplicator.
  

  true

  Send recovery information.

  false

  Do not send recovery information.

  Note:

  We recommend that you specify true because if recovery information is not sent, data linkage recovery cannot be performed via the system log file. For details about data linkage recovery via the system log file, see 9.5 Data linkage recovery via the system log file.

• recover_info_send_interval = recovery-information-transmission-interval
  ~((1-32767))<<1>>
  Specify the interval at which recovery information is to be sent to the target Datareplicator, expressed as a number of transactions in the source system. This operand takes effect only when true is specified in the recover_info_send operand. However, during event transmission, recovery information is sent to the target Datareplicator regardless of this operand's setting.
  For guidelines for the operand value, see 9.5.2(3)(a) Creating a recovery information file.
  
• cm_errno_check = true|false
  Specify whether the nature of a communication error is to be checked, and then connection establishment is to be retried only if reconnection is possible, or if connection establishment is to be retried unconditionally in the event of a communication error.
  

  true

  Retry connection establishment only if reconnection is possible.

  Note that depending on the combination of network configuration and devices, connection establishment might not be retried even in response to a transient error. If the error prohibits reconnection, the transmission process terminates with an error.

  false

  Retry connection unconditionally, regardless of the nature of the error.

  Notes:

  If true is specified, connection establishment is retried even for an error that is checked by the source Datareplicator. Therefore, a communication error that can be detected at an early stage when true is specified is not detected until connection establishment is retried as many times as specified in the retrynum operand in the transmission environment definition. To detect communication errors at an early stage, we recommend that you do not set the connection retry count (retrynum operand in the transmission environment definition) to unlimited.

  If connection establishment is retried in the event of an error that cannot result in reconnection, the KFRB02053-W message is output (the reason for line connection is other). Also, the system call type and errno are output to the extraction node master error information file as additional information.

• node_pddir=source-HiRDB's-PDDIR-environment-variable-value
  ~ <pathname of 1 to 255 bytes>
  If the source HiRDB is a parallel server and different information is specified in the PDDIR environment variable for each server machine, specify the absolute pathname of the source HiRDB's directory that is specified in the PDDIR environment variable. Specify this operand only in an individual definition section. If you omit the node_pddir operand, the source Datareplicator at the corresponding server machine will use the value of the PDDIR environment variable that is specified for the user environment in which the source Datareplicator's commands are executed.
  
• node_pdconfpath=HiRDB's-PDCONFPATH-environment-variable-value
  ~ <pathname of 1 to 255 bytes>
  If the source HiRDB is a parallel server and different information is specified in the PDCONFPATH environment variable for each server machine, specify the absolute pathname of the directory used to store the source HiRDB's system definition file that is specified in the PDCONFPATH environment variable. Specify this operand only in an individual definition section. If you omit the node_pdconfpath operand, the source Datareplicator at the corresponding server machine will use the value of the PDCONFPATH environment variable that is specified for the user environment in which the source Datareplicator's commands are executed.
  
• node_shlibpath=HiRDB's-SHLIB_PATH-environment-variable-value
  ~ < pathname of 1 to 255 bytes>
  If the source HiRDB is a parallel server and different information is specified in the SHLIB_PATH environment variable for each server machine, specify the absolute pathname of $PDDIR/lib that is specified in the SHLIB_PATH environment variable. Specify this operand only in an individual definition section. If you omit the node_shlibpath operand, the source Datareplicator at the corresponding server machine will use the value of the SHLIB_PATH environment variable that is specified for the user environment in which the source Datareplicator's commands are executed.
  
• node_host=node-master-host-name
  ~ <1-32 alphanumeric characters>
  If a system switchover configuration is used and the conditions listed below are not satisfied, you must allocate for Datareplicator a host name that inherits IP addresses and specify that host name in this operand:
  
  • In the system common definition for the HiRDB subject to extraction processing, a host name that inherits IP addresses is specified in the -x option of the pdunit or pdstart statement.
  • The system is set up in such a manner that, pursuant to the HA monitor's resource server facility, the above IP address is switched over after the HiRDB subject to extraction processing and the source Datareplicator have been terminated.
  The following is an example of specifying this operand: