Navigator User Guide

10.5.6
- 10.5.5
- 10.5.4
- 10.5.3
- 10.5.2
- 10.5.1
- 10.5
- 10.4.1
- 10.4.0 HotFix 1
- 10.4.0

Back Next

Adding a Capture Registration

A capture registration defines a source object such as an IMS segment, DB2 table, or VSAM data set for PowerExchange change capture processing.

PowerExchange stores capture registrations in the CCT file on the source system.

Before adding a capture registration, you must define the registration group. Within a registration group, you can add a capture registration for a specific source only once. The capture registration provides source object metadata and identifies the columns for which you want to capture change data.

You cannot register a source object that has column names greater than 128 characters in length.

Open a registration group.

On the

Registration Group

tab in the

Resource Explorer

, right-click the registration group and click

Add Capture Registration

In the

Add Capture Registration - Name and Table Filter

dialog box, enter a capture registration name and optional information as needed.

The following table describes the fields in this dialog box:

Field	Description
Name	Required. A user-defined name for the capture registration. The name must be composed of lowercase alphanumeric characters and begin with a letter. Valid lengths are: For z/OS and IBM i sources: 1 to 10 characters For Linux, UNIX, and Windows sources: 1 to 13 characters
Table Filter	Optional. To filter the tables, enter values in one or more of the following filter fields for your source type: For Adabas, Datacom, and IDMS, enter filter information in the Schema and Table fields. The maximum length of each field is 128 characters. For all DB2 types, enter filter information in the Creator/Schema , DBName/Definer , and Table fields. For DB2 for i5/OS, the maximum length of the Creator/Schema field is 10 characters and the maximum length of the Table field is 32 characters. For DB2 for z/OS, the maximum length of the each field is 128 characters. For IMS, enter filter information in the Schema , MapName , and Table fields. In the MapName field, enter a PowerExchange IMS data map name. Maximum lengths are 10 characters for the Schema field and MapName field and 128 characters for the Table field. For MSSql, enter filter information in the Owner and Table fields. The maximum length for each field is 128 characters. For MySQL, enter filter information in the Schema and Table fields. The maximum length of each field is 64 characters. In MySQL, the schema name is equivalent to the database name. For Oracle, enter filter information in the Schema and Table fields. The maximum length of each field is 128 characters. For SAP HANA, enter filter information in the Schema and Table fields. The maximum length for each field is 127 characters. For PostgreSQL, enter filter information in the Schema and Table fields. The maximum length of each field is 63 characters. For VSAM, enter filter information in the Schema , MapName , and Table fields. For MapName , enter the name of a PowerExchange VSAM data map. Maximum length is 10 characters for the Schema field and MapName field and is 128 characters for the Table field. In any filter field, you can include the following wildcards to create a mask: An asterisk (*) to represent one or more characters. A question mark (?) to represent a single character.
Respect Case	By default, PowerExchange converts text that you enter in the filter fields to the standard case for the data source. For example, on Oracle, uppercase is the standard. A value of scott or SCOTT would both produce SCOTT . Select the Respect Case option for the filter fields for which you do not want PowerExchange to use the standard case for the data source.
Escape Character	Enter the escape character to use in the filter boxes to delimit an asterisk () or question mark (?) that is used a literal value instead of a wildcard. For example, if you specify an escape character of ~ , a filter value of tab returns all tables that have table names beginning with tab . A filter value of tab~* returns the table that has the name tab* . Do not use an escape character with multibyte-enabled data sources.

Click

The

Add Capture Registration - Tables and Columns

dialog box appears.

If you entered filter criteria, the

Tables

list displays the tables that matched your filter criteria. Otherwise, the

Tables

list displays all tables.

The

Tables

list can include the one or more of the following fields, depending on the source type:

Column	Description
Creator Name	The creator name for the table
Schema	The schema name
Name or Table	The table name
Type	The object type, such as TABLE
DataSet Name	The data set name

Double-click a table to select it.

If the table contains column names that are longer than 128 characters, an error message is issued and you cannot register the table.

The

Columns

list displays the columns in the selected table. The list can include the following fields, depending on the source type:

Field	Description
Select Column	A check box next to each column for selecting columns for registration. Index columns are selected by default.
Name	The column name.
CCSID	For a DB2 source, the coded character set identifier (CCSID) related to the column.
PWXCP	For a DB2 source, the PowerExchange internal representation of the code page.
Codepage	For a DB2 source, the code page related to the column.

Select the columns for which you want PowerExchange to capture changes.

To select columns individually, select the

Select Column

check box next to each column. To select all columns, click

Select All Columns

Select all and notify changes

. The following table describes the options for selecting all columns:

Check Box	Description
Select All Columns	Available for source types that support selective column capture. Selects all columns in the Columns list for change capture. After you add the registration, any columns that are added to the source table are ignored until you edit the capture registration to select them.
Select all and notify changes	Available for DB2 and Oracle sources. If you select this option, PowerExchange captures changes for all columns. Any change to the schema for the table causes PowerExchange CDC to fail and log an error message. For a DB2 for z/OS source, the DB2 ECCR abnormally ends after it reads the first change record for the table after the schema change. For an Oracle source, Oracle CDC fails and logs an error message in the following situations: If a change record for a table that you registered for capture contains a column that you did not register for capture If a change record does not contain a column that you registered for capture For Oracle CDC, if a definition for a table changes in a way that is compatible with the PowerExchange capture registration, Oracle CDC continues to capture changes for that table. For example, if the length of a character column decreases but the capture registration does not reflect this change, Oracle CDC continues to capture changes for the table. Conversely, if the datatype of a column changes from numeric to character without a change in the capture registration, Oracle CDC continues to capture changes for the table until it encounters the first change record that contains nonnumeric data for the column. When Oracle CDC encounters a change record containing nonnumeric data for the column, it fails and logs an error message. If you clear the Select all and notify changes option, the check boxes appear next to the columns are cleared. To select the columns again, you must select their check boxes.

Check Box

Description

Select All Columns

Available for source types that support selective column capture. Selects all columns in the

Columns

list for change capture.

After you add the registration, any columns that are added to the source table are ignored until you edit the capture registration to select them.

Select all and notify changes

Available for DB2 and Oracle sources. If you select this option, PowerExchange captures changes for all columns. Any change to the schema for the table causes PowerExchange CDC to fail and log an error message.

For a DB2 for z/OS source, the DB2 ECCR abnormally ends after it reads the first change record for the table after the schema change.

For an Oracle source, Oracle CDC fails and logs an error message in the following situations:

If a change record for a table that you registered for capture contains a column that you did not register for capture

If a change record does not contain a column that you registered for capture

For Oracle CDC, if a definition for a table changes in a way that is compatible with the PowerExchange capture registration, Oracle CDC continues to capture changes for that table.

For example, if the length of a character column decreases but the capture registration does not reflect this change, Oracle CDC continues to capture changes for the table.

Conversely, if the datatype of a column changes from numeric to character without a change in the capture registration, Oracle CDC continues to capture changes for the table until it encounters the first change record that contains nonnumeric data for the column. When Oracle CDC encounters a change record containing nonnumeric data for the column, it fails and logs an error message.

If you clear the

Select all and notify changes

option, the check boxes appear next to the columns are cleared. To select the columns again, you must select their check boxes.

Informatica recommends that you select all columns because a registration can be defined only once in a registration group for specific source. You can edit the corresponding extraction map to deselect columns.

If you clear the check box for an index column, you might compromise the uniqueness of the row. A key symbol indicates that a column is an index column that Condense processing uses.

Click

In the

Add Capture Registration - Type

dialog box, enter additional information about the capture registration.

The following table describes the fields in this dialog box:

Field	Source Type	Description
Condense	All	Whether PowerExchange Condense or the PowerExchange Logger for Linux, UNIX, and Windows is used and the type of PowerExchange Condense processing on i5/OS or z/OS. Options are: None . PowerExchange does not use PowerExchange Condense or the PowerExchange Logger for Linux, UNIX, and Windows to process changes for the registered source. Part . PowerExchange uses PowerExchange Condense or the PowerExchange Logger to process changes for the registered source. PowerExchange Condense uses partial condense processing. Informatica recommends this option because it maintains transactional consistency. Full . PowerExchange uses PowerExchange Condense full condense processing for i5/OS or z/OS sources but some limitations apply. This option does not maintain transactional consistency. You can change the Condense option later without incrementing the registration version number. For more information, see Condense Options.
Status	All	The status of the capture registration with regard to availability for change capture use. Options are: Inactive . PowerExchange does not yet use the registration to capture changes for the source. This option is the initial status setting. You must manually set the option to Active to make the registration eligible for change capture use. Active . PowerExchange can use the registration to capture changes for the source. Default is Inactive . You can change the status from Inactive to Active later to make it eligible for change capture processing. For more information about status settings, see Status Options.
Comment	All	User-defined comments about the capture registration.
Database ID	Adabas	The DBID of the Adabas database that contains the file from which changes are captured.
File Number	Adabas	The file number of the Adabas file in the database identified by the Database ID field.
Primary Dataset Name	IMS synchronous	The name of the primary data set for the IMS database. This name can be up to 44 characters in length. Depending on the database type, the primary data set is one of the following: Full-function database: The primary data set is the data set that contains the primary (or only) data set group for the database. Fast Path database: The primary data set is the data set that contains the first (or only) area for the database. HALDB database: The primary data set is the data set prefix for the first partition, which is used for the data set name.
Supplement Log Group Name	Oracle	The name of the supplemental log group that you will create for the source. PowerExchange generates DDL for creating supplemental log groups when you complete the capture registration. You must define a supplemental log group for each table for which you want PowerExchange to capture Oracle change data. This supplemental log group must contain all table columns for which change data is to be captured.
Execute DDL now	Oracle	Select this option to run DDL to create an Oracle supplemental log group when you complete the capture registration. You must also enter a value in the Supplement Log Group Name box. Clear this option to run the supplemental log group DDL later. The GRANT statements in the ORACAPT.SQL file for the Oracle capture user do not include the authority that is required to run the DDL.
Full Audit	SAP HANA	Select this option if you want PowerExchange to capture before images and after images of Updates and all Insert and Delete data, along with some source metadata, in the shadow CDC tables. A shadow CDC table is generated for each source table.
Row Table	SAP HANA	Select this option to identify whether the table is a row table. Selecting this option is required for SAP HANA source tables that use row-based data storage.
Trigger Prefix	SAP HANA	An optional prefix that is added to the names of the AFTER DELETE, AFTER INSERT, and AFTER UPDATE triggers that are generated for each source table. The triggers write metadata to the PKLOG table, and for Full Audit mode, they write before images and after images to the shadow CDC tables. Enter a prefix up to 16 characters in length. An underscore (_) follows the prefix in the trigger name, for example, TX_ SAP_DEMO_TABLE_DBMI_ASCHEMA_t_d. You might want to use the prefix to comply with your site's trigger naming conventions.
Trigger	SAP HANA	Click this button to generate a script file that contains the SQL for creating the triggers. If Full Audit is selected, the script also contains SQL for generating the shadow CDC tables and a sequence for each table.

Click

Finish

For an Oracle source, in the

Save Supplement Log Group SQL to File

dialog box, enter a file name in the

File Name

field for the file that contains the generated DDL and click

Save

PowerExchange saves the DDL to the specified file. By default, the file is located in the PowerExchange root installation directory.

The PowerExchange Listener writes the capture registration to the CCT file on the source system. In the PowerExchange Navigator, the capture registration appears in the list of capture registrations on the

Registration Group

tab in the

Resource Explorer

. Also, the registration appears at the following locations in the interface:

The

Capture Registration

window displays the columns in the capture registration.

The

Capture Registration

tab in the

Resource Inspector

displays the properties for the capture registration. Use the arrows to toggle between the