Installing HiRDB Dataextractor

Type	Source system	Target system	File creation system
Host used to install HiRDB Dataextractor¹	For a HiRDB/Single Server: Host at which the single server is located	For a HiRDB/Single Server: Host at which the single server is located	Host used to create the file
For a HiRDB/Parallel Server, host at which one of the following is located: System manager Front-end server Back-end server Dictionary server	For a HiRDB/Parallel Server, host at which one of the following is located:² System manager Server at the data transmission destination³
For ORACLE: Host at which ORACLE is located

Table 2-3 lists the directories and files that are created by HiRDB Dataextractor. HiRDB Dataextractor creates directories and files at the host where it is installed. If this host already contains a directory or file with one of these names, you must save the directory or file under a different name.

Table 2-3 Directories and files created by HiRDB Dataextractor

Name of directory or file	Creation time	Description	Type
Name of directory or file	Creation time	Description	Source	Target	Creation
/opt/HIRDBXT/	Installation	Directory used to store HiRDB Dataextractor's various directories and files	C	C	C
/opt/HIRDBXT/bin		Directory used to store HiRDB Dataextractor's executable files (HiRDB Dataextractor's load modules and commands are stored in this directory)	C	C	C
/opt/HIRDBXT/bin/xtrep		Files for HiRDB Dataextractor's xtrep command	C	C	C
/opt/HIRDBXT/bin/xtstart		Files for HiRDB Dataextractor's xtstart command	C	C	C
/opt/HIRDBXT/bin/xtmsgtext		Text files used to output HiRDB Dataextractor messages	C	C	C
/opt/HIRDBXT/obj		Object storage directory	C	C	C
/opt/HIRDBXT/obj/pxto8160.o		Object for ORACLE extraction	C	C	C
/opt/HIRDBXT/lib		Libraries storage directory	C	C	C
/opt/HIRDBXT/lib/maptable/euc2jis.map		Mapping table for converting character codes (from EUC to SJIS)	C	C	C
/opt/HIRDBXT/lib/maptable/eucg2jis.map		Mapping table for converting character codes (from EUC to SJIS)	C	C	C
/opt/HIRDBXT/lib/maptable/euc2ucs2.map		Mapping table for converting character codes (from EUC to UCS2)	C	C	C
/opt/HIRDBXT/lib/maptable/eucg2u.map		Mapping table for converting character codes (from EUC to UCS2)	C	C	C
/opt/HIRDBXT/lib/maptable/jis2euc.map		Mapping table for converting character codes (from SJIS to EUC)	C	C	C
/opt/HIRDBXT/lib/maptable/jis2ucs2.map		Mapping table for converting character codes (from SJIS to UCS2)	C	C	C
/opt/HIRDBXT/lib/maptable/ucs22euc.map		Mapping table for converting character codes (from UCS2 to EUC)	C	C	C
/opt/HIRDBXT/lib/maptable/ucs22jis.map		Mapping table for converting character codes (from UCS2 to SJIS)	C	C	C
/opt/HIRDBXT/lib/usermap	When a mapping table is updated by the xtccnvedt command	Directory for mapping tables for converting user's character codes	--	C	C
/opt/HIRDBXT/lib/pxtora01.sl	At make in user environment	Library for ORACLE extraction	C	--	C
/opt/HIRDBXT/spool/¹	Installation	Directory used to store files output during HiRDB Dataextractor operation	C	C	C
/opt/HIRDBXT/spool/xter	HiRDB Dataextractor operation	HiRDB Dataextractor's error log file for TCP/IP	TCP²	TCP³	TCP³
/opt/HIRDBXT/spool/xtlog...		HiRDB Dataextractor's error log file for OSI	OSI³	OSI³	OSI³
/opt/HIRDBXT/spool/pder...		HiRDB database load utility's error files	--	C³	--
/opt/HIRDBXT/spool/.xtpc...		HiRDB database load utility's temporary error files	--	T³	--
/opt/HIRDBXT/spool/pdin... or /opt/HIRDBXT/spool/.pdin...		HiRDB database load utility's input data files	--	T³	--
/opt/HIRDBXT/spool/pdcf...	HiRDB Dataextractor operation	HiRDB database load utility's control information files	--	R³	--
/opt/HIRDBXT/spool/pdnf...		HiRDB database load utility's null value information files	--	R³	--
$XTTMPDIR/...		Output files For details about the filenames, see (3)(a) in 2.2.2 Installing HiRDB Dataextractor.	--	R⁴	C
/opt/HIRDBXT/spool/...			--	R⁴	C
/opt/HIRDBXT/spool/.pden...		EOF notification file for the HiRDB database load utility	--	T³	--
$XTTMPDIR/...		LOB input file storage directory For details about the filenames, see (3)(b) in 2.2.2 Installing HiRDB Dataextractor.	--	C⁵	C⁵
/opt/HIRDBXT/spool/...			--	C⁵	C⁵
$XTTMPDIR/.../xtlbf...		LOB input files For details about the filenames, see (3)(c) in 2.2.2 Installing HiRDB Dataextractor.	--	C⁵	C⁵
/opt/HIRDBXT/spool/.../xtlbf...			--	C⁵	C⁵
/opt/HIRDBXT/dump⁶		Directory for storing the module trace files used by HiRDB Dataextractor	C	C	C
/opt/HIRDBXT/dump/xtmt...		Module trace files that are output in the event of an error	C	C	C
/opt/HIRDBXT/dump/process-ID		Directory for storing core files for each load operation	C	C	C
/opt/HIRDBXT/dump/process-ID/core		Core files that are output in the event of an error	C	C	C

C: Created.

--: Not created.

T: File that is created temporarily when data is imported without creating an output file. Such a file is deleted when the processing is completed.

R: File that is retained even after the processing is completed because the -o option is specified in the xtrep command. If the -o option is omitted, such a file is deleted after processing is completed. However, if an error occurs during importing, the file is not deleted.

TCP: Created when the TCP/IP protocol is used.

OSI: Created when the OSI protocol is used.

Source: Source system

Target: Target system

Creation: File creation system

¹ Because /opt/HIRDBXT/spool contains files that are created dynamically by HiRDB Dataextractor, you must not delete /opt/HIRDBXT/spool. Also, you must not reference or delete any file under /opt/HIRDBXT/spool while HiRDB Dataextractor is active. You should delete these files periodically while HiRDB Dataextractor is stopped.

² This file is not created if an error occurs during parameter analysis. In such a case, error messages are output only to the standard output. You can output error messages to any file by specifying its filename in the -e option of the xtrep command. If the -e option is omitted, HiRDB Dataextractor assigns a unique filename. If you specify the XTTEMPNAM environment variable, you can use the date/time or the tempnam function to specify a name.

³ HiRDB Dataextractor assigns a unique filename. If you specify the XTTEMPNAM environment variable, you can use the date/time or the tempnam function to specify a name.

⁴ Such a file is created when the -o or -O option is specified in the xtrep command.

⁵ Such a file is created when the extracted data contains LOB data. If the -o option is specified in the xtrep command, this file is retained even after processing is completed; if the -o option is omitted, this file is deleted after processing is completed.

⁶ You should delete the files under /usr/HIRDBXT/dump periodically while HiRDB Dataextractor is stopped; use the XTTRCDIR environment variable to specify the desired directory name.

This section explains the conventions used by HiRDB Dataextractor to determine names for the following files, which are created by HiRDB Dataextractor:

Output files
LOB input file storage directory
LOB input files

(a) Output files

HiRDB Dataextractor determines names for output files on the basis of the following specifications:

-o or -O option in the xtrep command
XTTMPDIR environment variable
XTTEMPNAM environment variable

You can use the -o or -O option in the xtrep command or the XTFILESIZE environment variable to divide data and output it to multiple files. Table 2-4 lists the names of the output files. To specify the file mode for output files, use the XTFPERMIT environment variable.

Table 2-4 Names of output files

Specification				Name of output file
-o or -O option specified	Value of -o or -O option	XTFILESIZE environment variable specified	XTTMPDIR environment variable specified	Name of output file
Y	One file	Y	Y	$XTTMPDIR/specified-value-sequence-number
		Y	N	.../spool/specified-value-sequence-number
		N	Y	$XTTMPDIR/specified-value
		N	N	.../spool/specified-value
	Multiple files	N/A	Y	$XTTMPDIR/specified-value-x-files-count
	Multiple files	N/A	N	.../spool/specified-value-x-files-count
Y	One directory	Y	Y	$XTTMPDIR/specified-value/file-prefix¹-year.month.date-hour.minute.second-sequence-number
		Y	N	.../spool/specified-value/file-prefix¹-year.month.date-hour.minute.second-sequence-number
		N	Y	$XTTMPDIR/specified-value/file-prefix¹-year.month.date-hour.minute.second
		N	N	.../spool/specified-value/file-prefix¹-year.month.date-hour.minute.second
	Multiple directories	N/A	Y	$XTTMPDIR/specified-value/file-prefix¹-year.month.date-hour.minute.second-sequence-number-x-directories-count
	Multiple directories	N/A	N	.../spool/specified-value/file-prefix¹-year.month.date-hour.minute.second-sequence-number-x-directories-count
	Not specified	Y	Y	$XTTMPDIR/file-prefix¹-year.month.date-hour.minute.second-sequence-number
		Y	N	.../spool/file-prefix¹-year.month.date-hour.minute.second-sequence-number
		N	Y	$XTTMPDIR/file-prefix¹-year.month.date-hour.minute.second
		N	N	.../spool/file-prefix¹-year.month.date-hour.minute.second
N	Not specified	N/A	N/A	.../spool/.pdin-year.month.date-hour.minute.second²

...: Depends on the operating system being used:

HP-UX, Solaris, Linux, or AIX: /opt/HIRDBXT
Windows: installation-directory-name

Y: Specified.

N: Not specified.

N/A: Not applicable.

specified-value: Name specified in the -o or -O option.

$XTTMPDIR: Value specified in the XTTMPDIR environment variable.

¹ When the XTSQL environment variable is set to 0 or omitted: extracted-table-name

When the XTSQL environment variable is set to 1: xtof-process-ID

² If the value 0 is specified in the XTTEMPNAM environment variable, HiRDB Dataextractor uses the value created by the tempnam function for -year.month.date-hour.minute.second.

Output file creation example 1

To only create a file, assigning a unique name to the output file:

When you specify the -o option but do not specify a value, HiRDB Dataextractor assigns a unique filename. The file is created under the directory specified in the XTTMPDIR environment variable.

Environment variables

XTFILESIZE: None
XTTMPDIR: /tmp

xtrep command

xtrep -R -o TBL_NAME

Filename

/tmp/TBL_NAME-99.06.11-10.00.00

Output file creation example 2

To import data into a HiRDB table, dividing the data and creating multiple files in the same partition:

Specify the XTFILESIZE environment variable in order to divide the file. Because the XTTMPDIR environment variable is not specified in this example, the file is created under .../spool.

Environment variables

XTFILESIZE: 100000
XTTMPDIR: None

xtrep command

xtrep -o TBL_NAME

Filenames

.../spool/TBL_NAME-99.06.11-10.00.00-1
.../spool/TBL_NAME-99.06.11-10.00.00-2
                  :
                  :
.../spool/TBL_NAME-99.06.11-10.00.00-n

A file is divided in units of 100,000 bytes, and the maximum file size is as follows because no record spans files:

[Figure] 100K/L x L

L: Record length

Output file creation example 3

To only create files, dividing the data and creating multiple DAT-format files in separate partitions:

If you specify multiple partitions with the -o option, HiRDB Dataextractor creates files in separate partitions. Because the XTFILESIZE environment variable is not specified in this example, HiRDB Dataextractor fills one partition, and then creates a file in another partition.

HP-UX

Environment variables

XTFILESIZE: None
XTTMPDIR: /

xtrep command

xtrep -R dat -o HD000/, HD001/ TBL_NAME

Filenames

/HD000/TBL_NAME-99.06.11-10.00.00
/HD001/TBL_NAME-99.06.11-10.00.00

When /HD000 becomes full, HiRDB Dataextractor creates a file in /HD001.

Windows

Environment variables

XTFILESIZE: None
XTTMPDIR: \

xtrep command

xtrep -R dat -o C:\DATA\,D:\DATA\ TBL_NAME

Filenames

C:\DATA\TBL_NAME-99.06.11-10.00.00
D:\DATA\TBL_NAME-99.06.11-10.00.00

When C:\DATA becomes full, HiRDB Dataextractor creates a file in D:\DATA.

(b) LOB input file storage directory

If extracted data contains BLOB columns, HiRDB Dataextractor creates a LOB input file for each LOB data item. The directory used to store these LOB input files is the LOB input file storage directory.

HiRDB Dataextractor determines the name of the LOB input file storage directory on the basis of the following specifications:

-b option in the xtrep command
XTTMPDIR environment variable
XTTEMPNAM environment variable

Table 2-5 shows the name of the LOB input file storage directory.

Instead of creating LOB input files, you can store data for BLOB columns in the same output file for non-BLOB data by specifying the XTLOBKIND environment variable. Additionally, you can use the XTDPERMIT environment variable to specify a file mode for the LOB input file storage directory.

Table 2-5 Name of the LOB input file storage directory

Specification		Name of LOB input file storage directory
-b option	XTTMPDIR environment variable	Name of LOB input file storage directory
Y	Y	$XTTMPDIR/-b-option-value
Y	N	.../spool/-b-option-value
N	Y	$XTTMPDIR/xtlb-year.month.date-hour.minute.second-process-ID*
N	N	.../spool/xtlb-year.month.date-hour.minute.second-process-ID*

...: Depends on the operating system being used:

HP-UX, Solaris, or Linux, AIX: /opt/HIRDBXT
Windows: installation-directory-name

Y: Specified.

N: Not specified.

$XTTMPDIR: Value specified in the XTTMPDIR environment variable.

* If the value 0 is specified in the XTTEMPNAM environment variable, HiRDB Dataextractor uses the value created by the tempnam function for -year.month.date-hour.minute.second-process-ID.

(c) LOB input files

HiRDB Dataextractor creates LOB input files with a unique filename for each LOB data item in the LOB input file storage directory.

The following shows the convention for naming LOB input files:

xtlbf-data-item-number-BLOB-column-number

data-item-number: Sequence number beginning with 1 (up to 2³²)

BLOB-column-number: Sequence number beginning with 1 (up to 2³²)

Use the XTFPERMIT environment variable to specify the file mode for LOB input files.

2.2.2 Installing HiRDB Dataextractor

(1) Host used to install HiRDB Dataextractor

(2) Directories and files created by HiRDB Dataextractor

(3) File naming conventions

(a) Output files

(b) LOB input file storage directory

(c) LOB input files