FULL
. This option is available if you use PowerExchange Condense on
i5/OS
or z/OS. PowerExchange accumulates change data in keyed condense files. Because later changes supersede earlier changes, this condense option does not maintain transactional consistency. Also, the following limitations apply:
Adabas and IDMS log-based CDC data sources are not supported.
On z/OS, data sources must have key columns. The total length of all key columns for a source cannot exceed 250 bytes.
On i5/OS, source tables must have primary keys, or DDS files must be defined with a unique key.
PART
. This option is available if you use PowerExchange Condense on
i5/OS
or z/OS or the PowerExchange Logger for Linux, UNIX, or Windows. Changes in successfully committed UOWs are written to condense files or PowerExchange Logger log files in chronological order based on UOW end time. PowerExchange writes all changes for the columns of interest, not just the latest changes. This condense type maintains transactional consistency.
NONE
. Capture registrations are not eligible for full or partial condense processing.

CREATEBICI=(COLUMNS={PKFK|ALL|FILE}, MAXCOLS={
number_of_columns|
500}, EXTINFO={BI|CI|
BICI}, INPUT_FN=
input_file_name, OUTPUT_FN=
output_file_name, ERROR_FN=
error_file_name)

COLUMNS
. The columns in the extraction map for which the utility generates before-image columns, change-indicator columns, or both types of columns, as indicated by the EXTINFO keyword. Options are:
PKFK
. Generate before-image and change-indicator columns for primary-key and foreign-key columns.
ALL
. Generate before-image and change-indicator columns for all selected columns in the extraction map.
FILE
. Generate before-image and change-indicator columns for columns that are defined in the table schema in the utility input file.
No default is provided.
If you are generating before-image or change-indicator columns and the COLUMNS keyword identifies user-defined columns with expressions that invoke PowerExchange functions, ensure that the functions support BI buffering. If a function does not support BI buffering, PowerExchange does not generate BI or CI columns for the field. For more information, see Expression-Field Functions That Support CREATEBICI Processing.
MAXCOLS
. Optional. The maximum number of columns for which the utility generates before-image and change-indicator columns in the extraction map. Valid values are 10 through 32000. Default is 500.
How you set this keyword depends on the COLUMNS setting. Use the following guidelines:
If COLUMNS=FILE, the MAXCOLS value must be equal to or greater than the number of entries in the input file.
If COLUMNS=ALL, the MAXCOLS value must be equal to or greater than the largest number of columns in a registered table.
If COLUMNS=PKFK, the MAXCOLS value must be equal to or greater than the number of primary and foreign keys in the largest registered table.
EXTINFO
. Optional. The type or types of metadata columns to generate in the extraction maps. Options are:
BI
. Before-image columns.
CI
. Change-indicator columns.
BICI
. Both before-image and change-indicator columns.
Default is BICI.
In some cases, the before-image or change-indicator column cannot be created. For example, the utility cannot create a before-image column for DB2 z/OS LOB columns in extraction maps. If you try to create a before-image column for DB2 for z/OS LOB columns, the request is ignored and a warning message is written to the output file.
INPUT_FN
. On i5/OS, Linux, Unix, and Windows, the name of the input file that contains the column names for the before-image and change-indicator columns. On z/OS, enter
Y
or
N
. If you enter
Y
, the utility uses a default partitioned data set extended (PDSE) that has been pre-allocated with a DSNTYPE of LIBRARY and a record length of 255 bytes. You can use this keyword only if the COLUMNS keyword is set to
FILE
.
OUTPUT_FN
. On i5/OS, Linux, Unix, and Windows, the name of the output file that contains log information about the before-image and change-indicator columns created in the extraction maps. On z/OS, enter
Y
or
N
. If you enter
Y
, the utility uses a default partitioned data set extended (PDSE) that has been pre-allocated with a DSNTYPE of LIBRARY and a record length of 255 bytes.
ERROR_FN
. On i5/OS, Linux, Unix, and Windows, the name of the file that lists the columns for which before-image and change-indicator columns failed to be generated in the extraction maps. On z/OS, enter
Y
or
N
. If you enter
Y
, the utility uses a default partitioned data set extended (PDSE) that has been pre-allocated with a DSNTYPE of LIBRARY and a record length of 255 bytes.

The TABLE parameter specifies a mask that contains the asterisk (*) wildcard. To use CRGNAME, the table name must be explicitly specified in the same manner as when registering the table.
The REUSECRGNAME parameter is set to
Y
.
The CRGPREFIX parameter is specified. You must specify CRGPREFIX or CRGNAME, but do not specify both of them.

xnnnn
xxnnnn
xxxnnnn
xxxxnnnn

x...
is the prefix value that you enter. The first
x
value must be a letter from a to z. Subsequent
x
values can include the letters a to z or the integers 0 to 9.
nnnn
is a sequential number that the utility generates, starting from 0001.

ADA
for Adabas
AS4
for DB2 for i (i5/OS)
DB2
for DB2 for z/OS
DCM
for Datacom
IDL
for IDMS log-based
IMS
for IMS
MSS
for Microsoft SQL Server
MYS
for MySQL
ORA
for Oracle
PGS
for PostgreSQL
UDB
for DB2 for Linux, UNIX, and Windows
VSM
for VSAM

For
ADA
, enter an Adabas nucleus name.
For
AS4
, enter a DB2 for i (i5/OS) instance value that matches the INST parameter value in the AS4J CAPI CONNECTION statement in the DBMOVER member of the
dtllib
/CFG file. If you use PowerExchange Condense, this instance value must also match the DBID parameter value in the CAPTPARM member.
For
DB2
, enter a DB2 subsystem ID (SSID).
For
DCM
, enter a Datacom Multi-User Facility (MUF) name.
For
IDL
, enter an IDMS log-based CDC instance value that matches the
registration_logsid
parameter in the LOGSID statement in the DBMOVER configuration member.
For
IMS
, enter an IMS subsystem ID that matches the
ims_ssid
parameter value in the IMSID statement in the DBMOVER configuration member.
For
MSS
, optionally enter a unique user-defined instance identifier for the SQL Server database server and database name combination defined in the MSSOPTS DBSERVER and DBNAME parameters. Maximum length is seven characters. This identifier is incorporated into the names of the extraction maps that the utility creates. If you use the PowerExchange Logger for Linux, UNIX, and Windows, ensure that the instance identifier matches the DBID parameter value in the Logger configuration file. If you do not enter this instance value, PowerExchange generates a unique instance identifier that is composed of all or part of the publication database name followed by a 3-digit number if a number is required to make the identifier unique.
This INSTANCE parameter is useful in migration scenarios. If you need to deploy change capture from one environment to another, such as from test to production, and you do
not
define an instance identifier, PowerExchange uses the generated instance identifier in the new environment. The generated instance identifier might be different than the one in the original source environment. To avoid having to update the extraction map names in PowerCenter workflows and edit the DBID parameter value for the PowerExchange Logger, enter an instance identifier in the INSTANCE parameter that matches the instance identifier in the original environment when creating registrations for the new environment.
In this migration scenario, ensure that the dbmover.cfg configuration files in the original environment and new environment specify unique paths in CAP_PATH and CAPT_XTRA statements.
For
MYS
, enter a unique, user-defined instance identifier for the MySQL database server. Maximum length is seven characters. This identifier is used to identify a set of registrations for tables in the database on the MySQL server. This identifier is also incorporated into the names of the extraction maps that the utility creates. If you use the PowerExchange Logger for Linux, UNIX, and Windows, ensure that this instance identifier matches the DBID parameter value in the Logger configuration file.
For
ORA
, enter the user-defined collection ID for the Oracle instance that matches the
collection_id
parameter in the ORACLEID statement in the PowerExchange DBMOVER configuration member.
For
PGS
, enter a unique, user-defined instance identifier for the PostgreSQL database server and database name combination defined in the PGSOPTS DBSERVER and DBNAME parameters. Maximum length is seven characters. This identifier is incorporated into the names of the extraction maps that the utility creates. If you use the PowerExchange Logger for Linux, UNIX, and Windows, ensure that the instance identifier matches the DBID parameter value in the Logger configuration file. If you do not enter this instance value, PowerExchange generates a unique instance identifier that is composed of all or part of the database name followed by a three-digit number.
For
UDB
, enter a DB2 for Linux, UNIX, and Windows database name.
For
VSM
, enter a VSAM collection identifier.

OUTPUT=c:\pwx\outfile.txt

Uppercase and lowercase letters
The numbers 0 to 9
Spaces
The following special characters:
’ - ; # \ , . / ! % & * ( ) _ + { } : @ | < > ? 
The first character is an apostrophe.

Y
. Replace existing inactive registrations that match the mask specified in the TABLE parameter.
N
. Do not replace any existing registrations. Add registrations for tables that match the mask and do not yet exist.

Y
. Replace existing active registrations that match the mask specified in the TABLE parameter. For this replacement to occur, the REPLACE parameter must also be set to Y.
N
. Do not replace active registrations.

Y
. Retain the current names of existing registrations.
N
. Rename existing registrations by using the CRGPREFIX and sequential number format.

Y
. Retain the current names of existing registrations.
N
. Rename the existing registrations by using the CRGPREFIX and sequential number format.

A
. Create registrations with the status of Active.
I
. Create registrations with the status of Inactive. Later, you will need to activate the registrations to make them eligible for change capture use.

For relational database tables, the mask can be specified in the following format:

owner_or_schema.
tablename

Within this mask, the schema name and table name can each be up to 128 characters in length.

For nonrelational sources, the mask includes the data map name and can be specified in the following format:

schema.
mapname

Optionally, you can define a TABLE parameter value in NRDB or NRDB2 format that includes a table name to narrow the selection of tables when multiple tables are defined in a data map but not all of the tables need to be registered.

The following example shows the three-tier NRDB format, which uses period (.) separators between the schema name, map name, and table name:

SCHEMA.
mapname.
tablename

The following example shows the two-tier NRDB2 format, which is uses an underscore (_) separator between the map name and table name:

SCHEMA.
mapname_
tablename

OWNAB.T*

SCHEMA1.M*_T*

SCHEMA1.M*.T*

Y
. Perform a test run to determine the capture registrations to be updated or added. The test run does not actually update or add any registration.
N
. Do not perform a test run. When you run the utility, it will update or add capture registrations.