HiRDB Dataextractor Version 8 Description, User's Guide and Operator's Guide
![[Contents]](FIGURE/CONTENT.GIF)
![[Index]](FIGURE/INDEX.GIF)
![[Back]](FIGURE/FRONT.GIF)
![[Next]](FIGURE/AFTER.GIF)
This section describes the environment variables that are required in order to use HiRDB Dataextractor. These environment variables are specified as components of the following information, depending on the facility to be used:
- xtrep command executor's environment
- HiRDB Dataextractor's environment variable information setup file
Table 2-6 shows where to specify the HiRDB Dataextractor environment variables. Table 2-7 lists the environment variables required in order to use HiRDB Dataextractor.
Table 2-6 Location of HiRDB Dataextractor environment variables
| Item |
Source system |
Target system |
File creation system |
| Importing extracted data into HiRDB or creating a file |
Using XDM/XT to import data into XDM/RD or create a sequential dataset |
| Location of environment variables |
xtrep command executor's environment |
HiRDB Dataextractor's environment variable information setup file |
Table 2-7 Environment variables required in order to use HiRDB Dataextractor
| Environment variable |
Type |
| XDM/XT linkage |
HiRDB Dataextractor linkage |
| Source system |
Target system |
File creation system |
Source system |
Target system |
File creation system |
| EXT1 |
EXT2 |
EXT1 |
EXT2 |
| XTHOST |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| XTXHOST |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| XTPORTNO |
-- |
-- |
O |
-- |
O |
O |
-- |
-- |
| XTFESHOST |
-- |
-- |
O |
-- |
-- |
-- |
-- |
-- |
| XTOPNWTIME |
-- |
-- |
O |
-- |
-- |
-- |
O |
-- |
| XTWRTWTIME |
-- |
-- |
O |
-- |
-- |
-- |
O |
-- |
| XTTCPWTIME |
O |
O |
O |
O |
O |
O |
O |
O |
| XTOSIWTIME |
O |
O |
O |
O |
-- |
-- |
-- |
-- |
| XTTMPDIR |
-- |
-- |
O |
O |
-- |
-- |
O |
O |
| XTDPERMIT |
-- |
-- |
O |
O |
-- |
-- |
O |
O |
| XTFPERMIT |
-- |
-- |
O |
O |
-- |
-- |
O |
O |
| XTSTRETRY |
O2 |
O2 |
O2 |
O2 |
-- |
-- |
-- |
-- |
| XTLOGNAME |
O2 |
O2 |
O2 |
O2 |
O |
O |
O2 |
O2 |
| XTLOGRETRY |
O2 |
O2 |
O2 |
O2 |
O |
O |
O2 |
O2 |
| XTTEMPNAM |
O2 |
O2 |
O2 |
O2 |
O |
O |
O2 |
O2 |
| XTLOBBUFSIZE |
O |
O |
-- |
-- |
O |
O |
-- |
-- |
| XTFILESIZE |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| XTLOBKIND |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| XTERRLEVEL |
O |
O |
-- |
-- |
O |
O |
-- |
-- |
| XTPDCFPATH |
-- |
-- |
O |
-- |
-- |
-- |
O |
-- |
| XTPDCFxxxx |
-- |
-- |
O |
-- |
-- |
-- |
O |
-- |
| XTARRAY |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| XTTRCDIR |
O |
O |
O |
O |
O |
O |
O |
O |
| XTCNDSKP |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| XTEXTRACTDB |
O |
R |
-- |
-- |
-- |
-- |
-- |
-- |
| XTNLDFLT |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| XTORAUSER |
-- |
R |
-- |
-- |
-- |
R |
-- |
-- |
| XTSQL |
-- |
-- |
-- |
-- |
O |
-- |
-- |
-- |
| XTLOCALE |
-- |
-- |
-- |
-- |
O |
O |
O |
O |
| XTUNDEF |
-- |
-- |
-- |
-- |
-- |
-- |
O |
O |
| XTCLMxxxx |
O |
O |
-- |
-- |
-- |
-- |
-- |
-- |
| XTWHExxxx |
O |
O |
-- |
-- |
-- |
-- |
-- |
-- |
| XTLCKxxxx |
O |
O |
-- |
-- |
-- |
-- |
-- |
-- |
| XTTBLxxxx |
O |
O |
-- |
-- |
-- |
-- |
-- |
-- |
| XTLPRMxxxx |
-- |
-- |
O |
-- |
-- |
-- |
O |
-- |
| XTPDSRxxxx |
-- |
-- |
O |
-- |
-- |
-- |
O |
-- |
| TZ |
O2 |
O2 |
O2 |
O2 |
O |
O |
O2 |
O2 |
| XTPCHKTIME |
O2 |
O2 |
O2 |
O2 |
O |
O |
O2 |
O2 |
| PDDIR |
R |
-- |
R |
-- |
R |
-- |
R |
-- |
| PDCONFPATH |
R |
-- |
R |
-- |
R |
-- |
R |
-- |
| PDUSER1 |
R |
-- |
R |
-- |
R |
-- |
R |
-- |
| PDHOST1 |
O3 |
-- |
-- |
-- |
O |
-- |
-- |
-- |
| PDNAMEPORT1 |
O3 |
-- |
-- |
-- |
R |
-- |
-- |
-- |
| PDBLKF1 |
O |
-- |
-- |
-- |
O |
-- |
-- |
-- |
| PDFESHOST1 |
O |
-- |
-- |
-- |
O |
-- |
-- |
-- |
| PDSERVICEGRP1 |
O |
-- |
-- |
-- |
O |
-- |
-- |
-- |
| PDCLTPATH1 |
O |
-- |
-- |
-- |
O |
-- |
-- |
-- |
| PDSQLTRACE1 |
O |
-- |
-- |
-- |
O |
-- |
-- |
-- |
| ORACLE environment variables |
-- |
R |
-- |
-- |
-- |
R |
-- |
-- |
| PATH |
-- |
-- |
-- |
-- |
R |
R |
-- |
-- |
| SHLIB_PATH |
-- |
-- |
-- |
-- |
O |
O |
-- |
-- |
| LD_LIBRARY_PATH |
O |
-- |
O |
-- |
O |
-- |
O |
-- |
| LIBPATH |
O |
-- |
O |
-- |
O |
-- |
O |
-- |
| LANG |
O |
O |
O |
O |
O |
O |
O |
O |
R: Required.
O: Optional.
XDM/XT linkage: XDM/XT is linked for data extraction or import.
HiRDB Dataextractor linkage: HiRDB Dataextractor is linked to another HiRDB Dataextractor for data extraction or import.
EXT1: HiRDB source system.
EXT2: ORACLE source system.
1 HiRDB client's environment variable. For details about the specification, see the HiRDB UAP Development Guide.
2 The specification of this environment variable depends on the HiRDB Dataextractor startup method; for details, see (4) in 2.2.3 Specifying environment variables.
3 Always specify when the RDBDEF statement, which is a JXUMCTL control statement of XDM/XT, is specified.
The following explains how to specify each of the environment variables.
- XTHOST
- Specify on the source system the name of the host to which data is to be sent. You can specify the FQDN instead of the host name. This environment variable is assumed when the -h option is omitted from the xtrep command.
- To import extracted data into a table, specify the name of the host that contains the following server:
- HiRDB/Single Server
Single server (SDS)
- HiRDB/Parallel Server
System manager (MGR)
- To create a file from extracted data, specify the name of the host at which the file is to be created.
- Note
- If the hosts on the source and target systems use different LANs, you must specify the XTXHOST environment variable. The XTHOST and XTXHOST environment variables are mutually exclusive.
- If the target HiRDB uses the system switchover facility, specify the name of the host that inherits the IP address.
- XTXHOST
- Specify on the source system the name of the host on the data target system. You can specify the FQDN instead of the host name. If the hosts on the source and target systems use different LANs, you must specify this environment variable.
- This environment variable is assumed when the -H option is omitted from the xtrep command.
- Specify the name of the host on the data target system that contains the following server(s):
- HiRDB/Single Server
Single server (SDS)
- HiRDB/Parallel Server
System manager (MGR) and server to which data is to be sent (front-end server (FES) or back-end server (BES))
- For details about the specification, see the -H option in 4.2.2 xtrep command.
- Note
- The XTXHOST and XTHOST environment variables are mutually exclusive.
- If the target HiRDB uses the system switchover facility, specify the name of the host that inherits the IP address.
- XTPORTNO
- ~ <unsigned integer> ((1025-65535))
- Specify on the source system the port number to be used by the HiRDB Dataextractor on the data import or file creation system.
- This environment variable is assumed when the -p option is omitted from the xtrep command.
- On the target system, to transfer data from XDM/XT to a specified server using the OSI protocol, use this environment variable to specify the port number of the HiRDB Dataextractor running at the host where the target HiRDB's system manager is located. This port number must have been assigned (by the user) for HiRDB Dataextractor with /etc/services.
- XTFESHOST {STANDARD|HIRDB}
- Specify on the target system the type of the HiRDB server's host name (as specified in the pdunit -x option in the HiRDB system common definitions). This specification is required only when the OSI protocol is used, the target server (front-end server or back-end server) and XDM/XT are connected with the OSI protocol, and the system manager and XDM/XT are not connected with the OSI protocol.
- STANDARD: Use the default host name for the server.
- HIRDB: Use another host name for the server.
- XTOPNWTIME
- ~ <unsigned integer> ((60-65535)) <<60>>
- Specify on the target system the input file data write monitoring time (in seconds). This value is used in monitoring the amount of time used for the following processing:
- Time required for pdload to start loading data after HiRDB Dataextractor has issued an internal call to pdload
- The monitoring time must be greater than HiRDB's lock-release wait time; for details about the lock-release wait time, see the HiRDB System Definition manual. HiRDB Dataextractor does not monitor the time when writing data into an output file (-o or -O option is specified in the xtrep command).
- XTWRTWTIME
- ~ <unsigned integer> ((600-65535)) <<600>>
- Specify on the target system the input file data write monitoring time (in seconds). This environment variable is specified in case HiRDB Dataextractor terminates with an error while importing data into HiRDB in the addition mode due to a timeout because too many data items have been registered. This value is used in monitoring the amount of time used for the following processing:
- Time required for pdload to load data after HiRDB Dataextractor has written data into an input file
- HiRDB Dataextractor does not monitor the time when writing data into an output file (-o or -O option is specified in the xtrep command).
- XTTCPWTIME
- ~ <unsigned integer> ((60-65535))
- When the TCP/IP protocol is being used, specify the response monitoring time during data transfer (in seconds). If you omit this environment variable, HiRDB Dataextractor will not monitor for responses.
- Notes
- If the specified value is too small, HiRDB Dataextractor may erroneously assume a communication error and terminate processing.
- In the case of Windows, HiRDB Dataextractor monitors the time only during a data receive operation, not during a data send operation.
- XTOSIWTIME
- ~ <unsigned integer> ((60-65535)) <<3600>>
- When the OSI protocol is being used, specify the response monitoring time during data transfer (in seconds).
- Note
- If the specified value is too small, HiRDB Dataextractor may erroneously assume a communication error and terminate processing.
- XTTMPDIR
- Specify on the import or file creation system the name of the directory in which output files are to be created and the name of the LOB input file storage directory, expressed as an absolute pathname. If you omit this environment variable, HiRDB Dataextractor assumes /opt/HIRDBXT/spool.
- XTDPERMIT
- ~ <unsigned integer> ((1-777)) <<700>>
- Specify on the import or file creation system the file mode for the LOB input file storage directory, expressed as an octal number.
- Specify this environment variable when the user executing HiRDB Dataextractor is not the HiRDB administrator and file creation is to be followed by data import into a HiRDB database.
- If this environment variable is omitted, pdload called by HiRDB Dataextractor will terminate with an error because it does not have the privilege to read the created file.
- Note
- When umask(2) is specified, the actual file mode depends on the umask(2) specification.
- XTFPERMIT
- ~ <unsigned integer> ((1-777)) <<600>>
- Specify on the import or file creation system the file mode for the output file and LOB input files, expressed as an octal number.
- Specify this environment variable when the user executing HiRDB Dataextractor is not the HiRDB administrator and file creation is to be followed by data import into a HiRDB database.
- If this environment variable is omitted, pdload called by HiRDB Dataextractor will terminate with an error because it does not have the privilege to read the created file.
- Note
- If umask(2) is specified, the actual file mode depends on the umask(2) specification.
- XTSTRETRY
- ~ <unsigned integer> ((1-360))
- Specify the maximum number of retries for executing the xtstart command if the program required for OSI communication (XNF) is not active. If XNF is not active during execution of the xtstart command and this environment variable is specified, HiRDB Dataextractor retries the xtstart command until XNF starts, up to the number of times set in this environment variable. The retries are executed at 10-second intervals. If you omit this environment variable, the xtstart command will not be retried.
- XTLOGNAME
- Specify the absolute pathname of the file to which log messages are to be output. For the file naming conventions, see the applicable OS manual. Do not enclose the filename in quotation marks (").
- This environment variable is effective only if the value set in the XTTEMPNAM environment variable is 1.
- XTLOGRETRY
- ~ <unsigned integer> ((0-100)) <<20>>
- Specify the number of retries that may be attempted when message output to syslog fails.
- XTTEMPNAM {0|1}
- Specify the method for determining the names of files created by HiRDB Dataextractor:
- 0: Use the tempnam function to determine filenames
- 1: Use the date and time to determine filenames
- If you omit this environment variable, 1 is assumed.
- For details about the tempnam function, see the applicable OS manual.
- XTLOBBUFSIZE
- ~ <unsigned integer> ((1-2147483647)) <<BLOB column definition length>>
- Specify the buffer size for extracting columns with the BLOB or BINARY attribute at the source or for importing ORACLE data with the BLOB or BINARY attribute to the target.
- If the definition length does not match the actual LOB data length, you can save buffer space by specifying the actual data length in this environment variable, because HiRDB Dataextractor uses a buffer of the size specified here.
- If the specified value is greater than the BLOB or BINARY column definition length, HiRDB Dataextractor ignores this specification and uses the BLOB column definition length.
- The value specified in this environment variable is effective for all BLOB and BINARY columns in a table subject to extraction.
- Notes
- To extract a column of the SGMLTEXT type, you must specify this environment variable. If you omit this environment variable in such a case, HiRDB Dataextractor will issue the JXU7002E message and terminate with return code 12. The value of this environment variable must be the maximum length of the original text data subject to extraction.
- To extract ORACLE's LONG, LONG RAW, BLOB, CLOB, NCLOB, or BFILE columns, you must specify this environment variable. If you omit this environment variable in this case, HiRDB Dataextractor may issue the JXU7002E message and terminate with return code 12. The value of this environment variable must be the maximum length of the real data subject to extraction.
- If the extracted data exceeds the specified value, HiRDB Dataextractor executes postprocessing according to the specification of the XTERRLEVEL environment variable.
- The following is an example of determining the buffer size for real data (for HiRDB):
- This example outputs the maximum length of real data to the size output file.
- Column name specification file for BLOB type (clm):
max(length(column-name))
- Column name specification file for SGMLTEXT type (clm):
max(length(extracts(column-name)))
- xtrep command:
xtrep -R dat -o size -s clm table-name
- Example of the XTLOBBUFSIZE environment variable
- This example specifies the real data length (2,048 bytes) as the buffer size for the following data definition:
- Table definition
| Table A |
| Column A1 |
Column A2 |
Column A3 |
| BLOB (1K) |
SGMLTEXT |
BLOB (3K) |
- Definition lengths
- Column A1: 1,024 bytes
- Column A2: 2,147,483,647 bytes
- Column A3: 3,072 bytes
- Specification of environment variable
setenv XTLOBBUFSIZE=2048
- Buffer size
- When the above specification is made, the buffer sizes that are actually allocated are as follows:
- Column A1: 1,024 bytes
- Column A2: 2,048 bytes
- Column A3: 2,048 bytes
- Note
- If real data in column A2 or A3 exceeds 2,048 bytes with these settings, the data will be truncated.
- XTFILESIZE
- ~ <unsigned integer> ((8192-2147483647))
- Specify on the source system the maximum output file size.
- If you are creating a file during data import or you are creating a file during file creation, and there is too much data to be stored in a single file, you can divide the data into multiple output files by specifying a maximum output file size in this environment variable. HiRDB Dataextractor also saves data in multiple output files when multiple file names are specified in the -o or -O option of the xtrep command.
- HiRDB Dataextractor creates each output file so that each line fits within the specified size. For details about the names of these output files when this environment variable is specified, see (3) File naming conventions in 2.2.2 Installing HiRDB Dataextractor.
- Notes
- If you omit this environment variable and the size of an output file exceeds 2 GB (gigabytes) or the partition used to create the file becomes full, an error results and HiRDB Dataextractor issues the JXU7002E message (file write error).
- You must specify in this environment variable a value that is at least the size of a line of extracted data. If the actual data is larger than the value specified here, an error results and HiRDB Dataextractor issues the JXU7209E message (file write error).
- You can use this environment variable only if the HiRDB Dataextractor on the import or file creation system also supports this function. If HiRDB Dataextractor does not support this function, an error results and HiRDB Dataextractor issues the JXU7001E message on the import or file creation system.
- Note that the size of a row of extracted data depends on the XTNLDFLT environment value. For details about the data format, see (4) Output file in 4.2.4 Contents of files specified with the xtrep command, and A. Memory and File Requirements.
- XTLOBKIND {0|1}
- At the source system, specify the format of the data for BLOB-type columns (LOB data) to import it into a HiRDB table. This specification can improve performance because it enables BLOB-type LOB data to be stored in the same file as non-LOB data.
- If 2 is specified in the XTNLDFLT environment variable, this environment variable serves no purpose, because the input data format specified in the -W option of the database load utility (pdload) is used when data is imported into a HiRDB table or a binary-format file is created.
- 0: Output BLOB-type LOB data to separate files (LOB input files)
- 1: Output BLOB-type LOB data to the same files as non-LOB data
- Table 2-8 shows the format of the data that is created if 0 is specified in the XTLOBKIND environment variable, and Table 2-9 shows the format of the data that is created if 1 is specified in the XTLOBKIND environment variable.
Table 2-8 Format of the data to be created when 0 is specified in the XTLOBKIND environment variable
| Source system |
Import or file creation system |
Column attribute |
Format of data with BLOB attribute |
Value of -k option in pdload |
Remarks |
| ADT |
BLOB |
ADT |
BLOB |
| HiRDB or ORACLE |
Import to HiRDB table |
No |
No |
N/A |
N/A |
N/A |
N/A |
| Yes |
N/A |
1. |
f |
| Yes |
No |
2. |
N/A |
v |
| Yes |
2. |
1. |
| Creation of binary file |
No |
No |
N/A |
N/A |
N/A |
N/A |
| Yes |
N/A |
1. |
| Yes |
No |
1.#1 |
N/A |
#2 |
| Yes |
1.#1 |
1. |
| Creation of DAT file |
No |
No |
N/A |
N/A |
N/A |
N/A |
| Yes |
N/A |
1. |
| Yes |
No |
1. |
N/A |
| Yes |
1. |
1. |
- ADT: Abstract Data Type
- Format of BLOB-attribute data:
- Create a BLOB-attribute column in a separate file (input file or LOB input file).
- Create in one file (input file only).
- N/A: Not applicable
- #1
- This becomes 2. if SGMLTEXT is specified in the import information file.
- #2
- If you are extracting data from a table containing SGMLTEXT in order to create a file, verify that SGMLTEXT is specified with respect to the import information file. If this specification is omitted and the created file is used with pdload to import data into a HiRDB table, the results may be unpredictable.
Table 2-9 Format of the data to be created when 1 is specified in the XTLOBKIND environment variable
| Source system |
Import or file creation system |
Column attribute |
Format of data with BLOB attribute |
Value of -k option in pdload |
Remarks |
| ADT |
BLOB |
ADT |
BLOB |
| HiRDB or ORACLE |
Import to HiRDB table |
No |
No |
N/A |
N/A |
N/A |
N/A |
| Yes |
N/A |
2. |
d |
| Yes |
No |
2. |
N/A |
| Yes |
2. |
2. |
| Creation of binary file |
No |
No |
N/A |
N/A |
N/A |
N/A |
| Yes |
N/A |
2. |
| Yes |
No |
2. |
N/A |
N/A |
* |
| Yes |
2. |
2. |
| Creation of DAT file |
No |
No |
N/A |
N/A |
N/A |
Specification not permitted |
| Yes |
| Yes |
No |
| Yes |
- ADT: Abstract Data Type
- Format of data with BLOB attribute:
- Create a BLOB-attribute column in a separate file (input file or LOB input file)
- Create in one file (input file only)
- N/A: Not applicable.
- * If you are extracting data from a table containing SGMLTEXT in order to create a file, verify that SGMLTEXT is specified with respect to the import information file. If this specification is omitted and the created file is used with pdload to import data into a HiRDB table, the results may be unpredictable.
- Relationship between the value of the XTLOBKIND environment variable and the number of files to be created
- Table 2-10 shows the relationship between the value of the XTLOBKIND environment variable and the number of files to be created.
Table 2-10 Relationship between the value of the XTLOBKIND environment variable and the number of files to be created
| Specified value |
File split1 |
Number of output files (pdload input files) |
Example of data status2 |
| Input files |
LOB input files |
| 0 |
No |
1 |
Number of data items |
1. |
| Yes |
Number of split files |
Number of data items |
2. |
| 1 |
No |
1 |
0 |
3. |
| Yes |
Number of split files |
0 |
4. |
- 1 The output file can be divided if specified in the -o or -O option in the xtrep command or in the XTFILESIZE environment variable.
- 2 The following are examples of data status:
![[Figure]](FIGURE/ZU020203.GIF)
- Data format for BLOB columns when XTLOBKIND 1 is specified
- The following explains the data format for columns of the BLOB type (hereafter called BLOB) and abstract data type with the BLOB attribute (hereafter called ADT (BLOB)) when 1 is specified in the XTLOBKIND environment variable:
- The data size fields for BLOB and ADT (BLOB) columns are output with other columns in the order of their definition.
- The data fields of ADT (BLOB) columns are output at the end of the row data in the order of their definition.
- The BLOB data fields are output at the end of the row data in the order of their definition.
If the table to be imported into HiRDB contains both BLOB and ADT (BLOB), the BLOB data fields are output following the ADT (BLOB) data fields.
- If data is the null value, -1 is set in the size field for BLOB and ADT (BLOB). In this case, the data fields are not output.
- In the case of zero-bytes data, 0 is set in the size field for BLOB and ADT (BLOB). In this case, the data fields are not output.
- Example of data creation when XTLOBKIND is specified
- The following is an example of data creation when the XTLOBKIND environment variable is specified:
![[Figure]](FIGURE/ZU020204.GIF)
- XTERRLEVEL {0|1}
- Specify on the source system the HiRDB Dataextractor processing to be performed in the event of an error during data extraction.
- 0: If an error occurs, output the JXU7208I message, resume processing, and terminate with return code 4.
- 1: If an error occurs, output the JXU7208E message and terminate with return code 12.
- Notes
- The following errors are subject to the setting in this environment variable:
- Data truncation
- Overflow
For overflow to be detected, you must have specified Y in the pd_overflow_suppress operand in the HiRDB system common definition.
- Specify 1 in this environment variable in the following cases:
- When a column that has the SGMLTEXT data type is extracted
- When data is imported into a table in which truncation may occur
- XTPDCFPATH
- ~ <pathname>
- Specify on the target system the absolute pathname of the control information file that contains the idxwork, index, and lobmid statements for HiRDB's database load utility (pdload). This environment variable enables you to specify the destination of index information files and LOB middle files that are created by HiRDB's database load utility.
- Notes
- If this environment variable and the XTPDCFxxxx environment variable are omitted, pdload creates an index information file and LOB intermediate file under /tmp. If the target table contains indexes or columns of the BLOB type, temporary files are output under /tmp according to the amount of data. If there is not enough space in /tmp, the entire system may be affected adversely.
- If the target table contains no indexes but the specified control information file contains the idxwork or index statement, or if the target table contains no BLOB column but the specified control information file contains the lobmid statement, pdload results in an error. In such a case, use the XTPDCFxxxx environment variable. Create a control information file for each target table, and specify the XTPDCFxxxx environment variable. During execution of HiRDB Dataextractor, specify the XTPDCFxxxx environment variable in the xtrep command that corresponds to the table on the target system in order to select the control information file to be used.
- Specification example
- Environment variable information setup file
:
:
set XTPDCFPATH=/HIRDBXT/conf/xtpdcf
:
:idxwork /hd001/idxwork
lobmid /hd002/tmp
- XTPDCFxxxx
- ~ <pathname>
- Specify on the target system the absolute pathname of the control information file that contains the idxwork, index, or lobmid statement for HiRDB's database load utility (pdload). This environment variable enables you to specify the destination of index information files and LOB middle files that are created by HiRDB's database load utility.
- For xxxx, specify any character string (0 to 4 alphanumeric characters).
- When you specify the name of this environment variable in the xtrep command's -I option, pdload stores data using the control information file specified in this environment variable.
- Notes
- If you omit this environment variable or the XTPDCFPATH environment variable, pdload creates the index information file and LOB middle file under /tmp. If the target table contains indexes or columns of the BLOB type, temporary files are output under /tmp according to the amount of data. If there is not enough space in /tmp, the entire system may be affected adversely.
- If the target table contains no indexes but the specified control information file contains the idxwork or index statement, or if the target table contains no BLOB column but the specified control information file contains the lobmid statement, pdload results in an error. In such a case, use this environment variable to specify the control information file to be used by pdload. Create a control information file for each target table, and specify this environment variable. During execution of HiRDB Dataextractor, specify the XTPDCFxxxx environment variable in the xtrep command that corresponds to the table on the target system in order to select the control information file to be used.
- XTARRAY {FF|FV|VV} [,{C|E}]
- Specify the data format (FF, FV, or VV) and null value handling method (C or E) for output of repetition columns to a file.
- Specify this environment variable to specify the data format or null value handling method for each table. To specify this information for each column, use the null value information file that is specified in the -v option of the xtrep command. If you specify both this environment variable and a null value information file, the null value information file takes precedence.
- If you omit this environment variable, HiRDB Dataextractor assumes VV and C.
- If 2 is specified in the XTNLDFLT environment variable, this environment variable is ignored and VV and C are assumed when data is imported into a HiRDB table or when a binary-format file is created.
- For details about the specification of data format and null value handling method for repetition columns, see (6) in 3.1.1 Extracting selected data.
- Data format
- FF
- The number of elements is not at the beginning of a column and there is data for all column elements.
- FV
- The number of elements is at the beginning of a column and there is data for all column elements.
- VV
- The number of elements is at the beginning of a column and there are as many data items as there are elements.
- Null value handling method
- C
- When there is no data for any element, handle the entire column as the null value.
- E
- When there is no data for any element, handle each element as a null value.
- XTTRCDIR
- ~ <pathname>
- Specify the absolute pathname of the directory for the module trace file and core file that are created by HiRDB Dataextractor. The following file and directory are subject to this environment variable setting:
- Module trace file
- Directory for storing core
- If you omit this environment variable, HiRDB Dataextractor assumes /opt/HIRDBXT/dump.
- Note
- You should delete the files under /opt/HIRDBXT/dump periodically while HiRDB Dataextractor is stopped.
- XTCNDSKP~ {0|1} <<0>>
- Specify whether or not transfer data is to be compressed. If either of the following conditions is satisfied, you should not compress transfer data (in order to reduce processing time):
- Communication performance is better than your machine performance
- Transfer data's compression rate is low (there are not many consecutive identical codes)
- 0: Compress transfer data.
- 1: Do not compress transfer data.
- XTEXTRACTDB
- Specify the type of database that is to be subject to extraction.
- oracle: ORACLE database
- hirdb: HiRDB table
- ORACLE environment variables
- Define the ORACLE environment variables required in order to extract from ORACLE databases; for the names of the applicable environment variables and their values, see the ORACLE manual.
- XTNLDFLT ~{0|1|2}
- Specify the method to be used to import the null value when data is imported into a HiRDB table or file.
- 0: When a null value information file is omitted (-v option is omitted) or an assumed value is used as the default null value for a column that is not specified in the null value information file.
- Specify this value if the value to be assumed is not used as real data.
- For details about default values that are assumed to be null values, see Table 4-13 Null-value information file specification rules (applicable to data extraction from a HiRDB table).
- 1: When a null value information file is omitted (-v option is omitted) or an assumed value is used as the default null value for a column that is not specified in the null value information file.
- Specify this value for the following data types if real data uses the default value that is assumed when the value of this environment variable is 0:
- DECIMAL
- DATE
- INTERVAL YEAR TO DAY
- TIME
- INTERVAL HOUR TO SECOND
- TIMESTAMP
- For details about default values that are assumed to be null values, see Table 4-13 Null-value information file specification rules (applicable to data extraction from a HiRDB table).
- 2: The null value is to be imported as is without using the default null value.
- Specify this value if real data uses the default value that is assumed when the value of this environment variable is 0 or 1. Note that when this value is specified, the XTLOBKIND environment variable is ignored.
- The following table shows the combinations of this environment variable and the NOT NULL constraint for columns in the source and target tables:
| NOT NULL constraint |
XTNDFLT environment variable |
| Source table |
Target table |
0 |
1 |
25 |
| NULL |
NULL |
Y1 |
Y1 |
Y |
| NOT NULL |
Y2 |
N3 |
N4 |
| NOT NULL |
NULL |
-- |
-- |
-- |
| NOT NULL |
-- |
-- |
-- |
- Y: Can be specified.
- --: Specification not required (because the source table does not contain the null value, specification of this environment variable serves no purposes).
- N: Cannot be specified (if pdload is used to import data, the input data will be regarded as invalid).
- 1 Real data that is the same as the default null value is treated as the null value.
- 2 For the null value, the default null value is imported as real data.
- 3 If the following data types contain the null value, the input data is regarded as invalid; for any other data types, the default null value is imported as real data:
- DECIMAL, DATE, INTERVAL YEAR TO DAY, TIME, INTERVAL HOUR TO SECOND, TIMESTAMP
- 4 If the source table contains the null value, the input data is regarded as invalid.
- 5 When this value is specified, the amount of data is greater than when any of the other values is specified because offset information is added to check the null values. Therefore, if the source data contains no real data that is the same as the default null value, specifying this value provides no benefits.
- For details about the data format, see (4) Output file in 4.2.4 Contents of files specified with the xtrep command and A. Memory and File Requirements.
- The format of output data during data import into a HiRDB table or file creation depends on this specification.
- The following table shows the relationship between the value of this environment variable and the format of output data:
XTNDFLT
environment variable |
Format of output data |
| Import to HiRDB |
File format |
| Binary format |
DAT format |
| 0 |
Binary input file format supported by HiRDB's database load utility. |
Same as at left |
DAT input file format supported by HiRDB's database load utility. |
| 1 |
| 2 |
Input file format used when the -W option is specified in HiRDB's database load utility1. |
Same as at left |
- 1 BLOB-type LOB data is output to the same file as non-BLOB-type data, regardless of the specification of XTLOBKIND environment variable.
- Notes on specification value 1
- There are HiRDB version requirements (07-00/J or later) for this specification. If the requirements are not satisfied, HiRDB's database load utility (pdload) results in an error due to invalid contents in the null value information file.
- If this specification is used for the file creation (binary-format) facility and the source table contains the null value, the default null value is stored as data in the file. Note that when a file created by this facility is used, the format of the default null value is different from that of the data type (X'FF ... FF').
- Notes on specification value 2
- This specification is applicable only when the target or the HiRDB Dataextractor where the file is created also supports this facility. If the target or the HiRDB Dataextractor where the file is created does not support this facility, it issues the JXU7017E message, resulting in an error.
- For data import into a FIX table, specify a value other than 2 because specification of 2 serves no purpose. If 2 is specified to import data into a FIX table, HiRDB Dataextractor issues the JXU7223E message, resulting in an error.
- XTORAUSER
- Specify the user name and password of the user who will connect to ORACLE on the source system, expressed as "user-name"/"password".
- The specified user must have ORACLE's SELECT or SELECT ANY TABLE privilege for the table to be extracted. If LOCK is issued to the table to be extracted, the user must also have the LOCK ANY TABLE system privilege or object privilege.
- XTSQL
- Specify whether or not user-specified SQL statements may be specified when multiple HiRDB Dataextractors are linked and data is extracted from an HiRDB table or ORACLE database. For details about specifying the command, see 4.2.3 Additional data extraction and import functions.
- 0: Extract data from the table specified on the command line of xtrep.
- 1: Extract data from the table specified in the table expression specification file.
- Notes
- When you specify 1 in this environment variable, you must specify the following parameters; if you omit these parameters, HiRDB Dataextractor issues the JXU7307E message, resulting in an error:
-g option
-r or -R option
- When you specify 1 in this environment variable, you must not specify the following parameter; if you specify this parameter, HiRDB Dataextractor issues the JXU7306E message, resulting in an error:
-w option
When you specify 1 in this environment variable, specify on the command line the name of the table expression specification file, not the name of the extraction table. If the specification is omitted, HiRDB Dataextractor issues the JXU7300E message, resulting in an error.
- XTLOCALE
- ~ {sjis|euc|utf-8|unknown} <<unknown>>
- To use the character code conversion function, specify a character code system used by both the locale and the target system:
- sjis: JIS8 or Shift JIS system
- euc: EUC system
- utf-8: UTF-8 system
- unknown: Do not convert character codes
- Notes
- HiRDB Dataextractor does not check the value of this environment variable against the locales supported by the OS. If the specified value is invalid, operations cannot be guaranteed.
- If the character code system is UTF-8, specify utf-8 regardless of whether code conversion is used.
- XTUNDEF
- ~ {space|sharp|hexadecimal-character-string}[,cancellation-count] <<space,100>>
- To use the character code conversion function, specify the handling of undefined codes at the target system:
- space: Replace with the single-byte space (0x2020).
- sharp: Replace with the double-byte hash mark (#).
- hexadecimal-character-string: Replace with the specified double-byte code. Specify a value that begins with 0x and consists of 4 characters in the range 0 to f.
- cancellation-count ~ <unsigned integer> ((0-10000))
- Specify the number of records before processing is to be cancelled after a conversion error is detected. This means that a record is counted as 1 even if multiple conversion errors are detected within that record. When 0 is specified, processing will not be cancelled.
- Notes
- When sharp is specified in this environment variable, the code for the double-byte hash mark (#) differs depending on the specification value for the XTLOCALE environment variable at the target system:
- sjis: 0x8194
- euc: 0xa1f4
- utf-8: 0xefbc83
- If you used the XTLOCALE environment variable to specify the UTF-8 code set at the target system or the file creation system, specify the UCS2 code value corresponding to the value to be output in UTF-8 in hexadecimal-character-string.
- XTCLMxxxx (xxxx: any character string)
- ~ <pathname> ((1-256 characters))
- Specify on the source system the absolute pathname of the column name specification file.
- XTWHExxxx (xxxx: any character string)
- ~ <pathname> ((1-256 characters))
- Specify on the source system the absolute pathname of the table expression specification file.
- XTLCKxxxx (xxxx: any character string)
- ~ <pathname> ((1-256 characters))
- Specify on the source system the absolute pathname of the lock information file.
- XTTBLxxxx (xxxx: any character string)
- ~ <path name> ((1-255 characters))
- Specify at the source system the absolute path name of the table name information file. For details about the table name information file, see (7) Setting the table name information file in 2.2.4 Files required in order to use HiRDB Dataextractor.
- XTLPRMxxxx (xxxx: any character string)
- ~ <path name> ((1-255 characters))
- Specify at the target system the absolute path name of the file that contains the pdload command line information. For details about this file, see (8) Setting the pdload command line information file in 2.2.4 Files required in order to use HiRDB Dataextractor.
- XTPDSRxxxx (xxxx: any character string)
- ~ <path name> ((1-255 characters))
- Specify at the target system the absolute path name of the file that contains the pdload control information file's source statement information. For details about this file, see (9) Setting the pdload control information file's source statement information file in 2.2.4 Files required in order to use HiRDB Dataextractor.
- TZ
- ~ <character string>
- Specify the time zone for date and time information that is to be output to the error log file. Verify that this environment variable's setting is the same in both HiRDB and HiRDB Dataextractor. If the settings are different, HiRDB's and HiRDB Dataextractor's message times in the error log file and syslog may not match.
- If this environment variable is omitted, the OS settings are assumed.
- XTPCHKTIME
- ~ <unsigned integer> ((30~600)) <<180>>
- Specify one of the following as the monitoring interval to be used, as appropriate to the facility being used. Specify a multiple of 30. If the specified value is not a multiple of 30, HiRDB Dataextractor rounds it up to the nearest multiple of 30.
- When using a HiRDB Dataextractor-linked extraction facility:
Interval from issuance to the target system of a confirmation request to reception from the target system of a confirmation message
- When using any other facility:
Interval from the start of target confirmation to reception of a target system confirmation message
- PDDIR
- Specify on the source system the absolute pathname of the HiRDB directory for the HiRDB running on the source system.
- Specify on the target system the absolute pathname of the HiRDB directory for the HiRDB running on the target system.
- PDCONFPATH
- Specify on the source system the absolute pathname of the HiRDB system definition files for the HiRDB running on the source system.
- Specify on the target system the absolute pathname of the HiRDB system definition files for the HiRDB running on the target system.
- PDUSER
- Specify on the source system the user name and password of the user who will connect to the HiRDB running on the source system, expressed as "user-name"/"password". The specified user must have HiRDB's SELECT privilege for the table to be extracted.
- Specify on the target system the user name and password of the user who will connect to the HiRDB running on the target system, expressed as "user-name"/"password". The specified user must have HiRDB's INSERT privilege for the table to be imported. If existing table data is to be deleted before importing data, the user must have HiRDB's INSERT and DELETE privileges.
- PDHOST
- Specify on the source system the host name of the system manager for the HiRDB running on the source system.
- PDNAMEPORT
- Specify on the source system the port number of the HiRDB system located on the source system.
- PDBLKF
- Specify on the source system the number of rows for block transfer.
- This environment variable may be able to reduce HiRDB's extraction time. For details, see the HiRDB UAP Development Guide.
- PDFESHOST
- Specify on the source system the host name of the front-end server.
- PDSERVICEGRP
- Specify on the source system the name of the front-end server.
- PDCLTPATH
- Specify on the source system a directory for storing the SQL trace file that is used to collect HiRDB's SQL trace information. If PDCLTPATH is omitted, the SQL trace file is stored in the current directory.
- PDSQLTRACE
- Specify this environment variable on the source system in order to collect HiRDB's SQL trace information. In this environment variable specify the size of the SQL trace file. When you extract data from HiRDB, HiRDB Dataextractor issues HiRDB SQL statements in order to perform the data extraction. If you collect SQL trace information, you can use this information in the event of a data extraction error. For details about the SQL trace file, see the HiRDB System Operation Guide.
- PATH
- Specify this environment variable on the source system in order to add /opt/HIRDBXT/bin.
- SHLIB_PATH
- In a user environment where the xtrep command is entered, or in the case of using an environment variable information setup file, specify the absolute path to the directory that contains the source HiRDB or ORACLE shared library.
- If you are extracting data from an ORACLE database, specify the path to a shared library that contains 32-bit client libraries.
- If the OS being used is Solaris, Linux, or AIX, this environment variable is ignored when specified.
- LD_LIBRARY_PATH
- ~ <pathname>
- In a user environment where the xtrep command is entered, or in the case of using an environment variable information setup file, specify the absolute path to the directory that contains the source HiRDB or ORACLE shared library.
- If you are extracting data from an ORACLE database, specify the path to a shared library that contains 32-bit client libraries.
- This environment variable is applicable only when the OS being used is Solaris or Linux.
- LIBPATH
- <path name>
- In a user environment where the xtrep command is entered or in the case of using an environment variable information setup file, specify the absolute path to the directory that contains the source HiRDB or ORACLE shared library.
- If you are extracting data from an ORACLE database, specify the path to a shared library that contains 32-bit client libraries.
- This environment variable is applicable only when the OS being used is AIX.
- LANG
- Specify on the source system the character code set being used when HiRDB on the source system uses national character strings.
- Specify on the target system the character code set being used when HiRDB on the target system uses national character strings. To import data from a mainframe database into a HiRDB database using XDM/XT, this character code set must be of the character type after conversion that is specified in XDM/XT's CONVERT JXUMCTL control statement. If a different character type is specified, the results may not be correct.
- Specify on the file creation system the character code set being used when the system on the file creation system uses national character strings.
- The default value is ja_JP.SJIS.
- Note
- When the LC_ALL or LC_* environment variable is specified, the LANG environment variable is ignored if specified, because the former takes effect. The LANG environment variable settings take effect only when the LC_ALL or LC_* environment variable is omitted.
Note that with some systems, a default value may be set for the LC_ALL or LC_* environment variable.
If you need to change the value of the LC_ALL or LC_* environment variable set by the system, specify the LC_ALL or LC_* environment variable instead of the LANG environment variable.
Take this into account when you are executing a command that requires the LANG environment variable.
- When the OS being used is Windows, the LANG environment variable is ignored even if specified.
(3) Specifying environment variables in HiRDB Dataextractor's environment variable information setup file
The following shows how to specify environment variables in HiRDB Dataextractor's environment variable information setup file.
- Format
set environment-variable=environment-variable-value
- Rules
- Specify one set statement per line.
- Do not place any spaces before or after the equals sign (=).
- A comment begins with the number sign (#) and ends at the end of the line.
- Example
set PDDIR=/HiRDB
set PDCONFPATH=/HiRDB/conf
set PDUSER="user01"/"user01"
set LANG=ja_JP.SJIS
set XTPORTNO=20052
(4) Exception methods for specifying environment variables
This subsection describes the environment variable specification methods that are different from the normal specification method.
Table 2-11 shows the exception methods for specifying environment variables.
Table 2-11 Exception methods for specifying environment variables
| Protocol |
OS being used |
HiRDB Dataextractor's startup method |
Environment variable specification method |
Remarks |
| TCP/IP |
HP-UX
Solaris
Linux
AIX |
Automatic startup by inetd |
Specify in inetd.conf. |
See specification example 1. |
| Windows |
Automatic startup by service |
Specify in hirdbxt.ini. |
N/A |
| OSI |
HP-UX |
Automatic startup by localrc or rc |
Specify in localrc or rc |
See specification example 2. |
| Manual startup using the xtstart command |
Specify setenv in the executing user environment. |
See specification example 3. |
Windows
Solaris
Linux
AIX |
N/A |
N/A |
N/A |
- Specification example 1
Specifying /etc/inetd.conf:
hirdbxt stream tcp nowait user01 /bin/env env XTTEMPNAM=0
/opt/HIRDBXT/bin/pxtcntl /usr/home/user01/xtenv1
- Note
- Specify the entire coding in one line. If you specify an environment variable, do not specify pxtcntl.
- Specification example 2
Specifying /etc/localrc:
/bin/env XTTEMPNAM=0 /opt/HIRDBXT/bin/xtstart
/opt/HIRDBXT/HIRDB_conf /opt/HIRDBXT/OSI_conf
- Note
- Specify the entire coding in one line.
- Specification example 3
Specifying the xtstart command:
setenv XTTEMPNAM=0
/opt/HIRDBXT/bin/xtstart /opt/HIRDBXT/HIRDB_conf /opt/HIRDBXT/OSI_conf
- Note
- Send one line at a time.
- Note
The environment variable specified by the exception method is underlined.
All rights reserved. Copyright (C) 2007, Hitachi, Ltd.