The CDC splitter splits a single flow of change rows from a CDC source data flow into different data flows for Insert, Update and Delete operations. The data flow is split based on the required column
__$operation and its standard values in SQL Server 2016 change tables.
|Value of Operation||Output||Description|
|2||Insert||Inserted row (not available when using Net with Merge CDC mode)|
|3||Update||Before-update row (available only when using All with Old Values CDC mode)|
|4||Update||After-update row (follows the Before-update)|
|5||Update||Merge row (available only when using Net with Merge CDC mode)|
You can use the splitter to connect to pre-defined INSERT, DELETE, and UPDATE outputs for further processing.
The CDC Splitter transformation has one regular input and one error output.
The CDC Splitter transformation has an error output. Input rows with an invalid value of the $operation column are considered erroneous and are handled according to the ErrorRowDisposition property of the input.
The component error output includes the following output columns:
Error Code: Set to 1.
Error Column: The source column causing the error (for conversion errors).
Error Row Columns: The input columns of the row that caused the error.
Configuring the CDC Splitter
There are no configurable properties for the CDC splitter.
For more information about using the CDC splitter, see CDC Components for Microsoft SQL Server Integration Services.
The Advanced Editor dialog box contains the properties that can be set programmatically.
To open the Advanced Editor dialog box:
- In the Data Flow screen of your SQL Server 2016 Integration Services (SSIS) project, right click the CDC splitter and select Show Advanced Editor.