Skip to main content

​
Prerequisites

Before you add and set up the source connector, you need to set up the database for change data capture (CDC), as explained in the following sections. Limitations: The following capabilities are not supported in :
  • remote journals or failover functionality
  • large objects such as CLOB, XML, TEXT, BLOB

​
Set Up for CDC

Journaling records database changes in journal receivers, which the change data capture (CDC) process uses to capture and replicate those changes to other systems in real-time or near real-time synchronization. A journal is a database object that serves as a tracking mechanism, capturing metadata about changes made to specific database objects. Each journal is linked to one or more journal receivers, which physically store the detailed change records, including before-and-after images of the data as well as metadata and transaction details. Because CDC depends on journaling to supply detailed change records, creating a journal and journal receiver and setting user permissions are the first steps in enabling CDC for , as explained in the next sections.

​
Create the Journal Receiver and Journal and Enable Journaling on Tables

Run the following command-line (CL) commands to create the journal receiver, journal, and enable journaling on tables: 
   -- 1. Create a journal receiver.
CRTJRNRCV <RECEIVER_LIBRARY>/<RECEIVER_NAME>

   -- 2. Create a journal.
CRTJRN <JOURNAL_LIBRARY>/<JOURNAL_NAME> <RECEIVER_LIBRARY>/<RECEIVER_NAME>

   -- 3. Enable journaling on a file (table) by using the STRJRNPF command. As a best practice, capture both before-and-after images of the records to support features like auditing and recovery.
STRJRNPF <FILE_LIBRARY>/<FILE_NAME> <JOURNAL_LIBRARY>/<JOURNAL_NAME> IMAGES(*BOTH)

   -- If you already started journaling and need to change the image settings, submit the CHGJRNOBJ command.
CHGJRNOBJ OBJ((<FILE_LIBRARY>/<FILE_NAME> *FILE)) ATR(*IMAGES) IMAGES(*BOTH)

   -- To stop journaling for a specific physical file (table), use the ENDJRNPF command.
ENDJRNPF FILE(<FILE_LIBRARY>/<FILE_NAME>) JRN(<JOURNAL_LIBRARY>/<JOURNAL_NAME>)
To verify that journaling is enabled for a table, or for details about the journals and their journal receivers, execute the following queries: 
SELECT * FROM QSYS2.JOURNAL_RECEIVER_INFO WHERE JOURNAL_RECEIVER_LIBRARY = <RECEIVER_LIBRARY>;
SELECT * FROM QSYS2.JOURNAL_INFO WHERE JOURNAL_LIBRARY = '<JOURNAL_LIBRARY>';
SELECT * FROM QSYS2.JOURNALED_OBJECTS WHERE JOURNAL_LIBRARY = '<JOURNAL_LIBRARY>';
The following example creates the RCV01 journal receiver and the JRN1 journal. Then, it adds the PRODUCTS table for journaling by using QSYS2.QCMDEXC to call any CL command through the JDBC tool. All of the objects are created in the TEST2 schema/library. 
   -- 1.Create a NEW journal receiver.
CALL QSYS2.QCMDEXC('CRTJRNRCV TEST2/RCV01');

   -- 2. Create a journal.
CALL QSYS2.QCMDEXC('CRTJRN TEST2/JRN1 TEST2/RCV01');

   -- 3. Start journaling tables/files.
CALL QSYS2.QCMDEXC('STRJRNPF FILE(TEST2/PRODUCTS) JRN(TEST2/JRN1) IMAGES(*BOTH)');

   -- 4. Stop journaling tables/files.
CALL QSYS2.QCMDEXC('ENDJRNPF FILE(TEST2/PRODUCTS) JRN(TEST2/JRN1)');

VALUES CURRENT_SCHEMA; -- TEST2
SELECT * FROM QSYS2.JOURNAL_RECEIVER_INFO WHERE JOURNAL_RECEIVER_LIBRARY = CURRENT_SCHEMA;
SELECT * FROM QSYS2.JOURNAL_INFO WHERE JOURNAL_LIBRARY = CURRENT_SCHEMA;
SELECT * FROM QSYS2.JOURNALED_OBJECTS WHERE JOURNAL_LIBRARY = CURRENT_SCHEMA;
When you create a CDC job in , you select the journal and schema to determine which tables are available for capture.

​
Set Required User Permissions

To verify all permissions for the user to which you are connected, submit the following query: 
SELECT * FROM QSYS2.OBJECT_PRIVILEGES WHERE AUTHORIZATION_NAME = '<CDC_USER>'
If the user lacks privileges to the journal, journal receiver, or the schema (library) from which you want to capture changes, run the following CL commands: 
GRTOBJAUT OBJ(<JOURNAL_LIBRARY>) OBJTYPE(*LIB) USER(<CDC_USER>) AUT(*EXECUTE)
GRTOBJAUT OBJ(<JOURNAL_LIBRARY>/*ALL) OBJTYPE(*JRNRCV) USER(<CDC_USER>) AUT(*USE)
GRTOBJAUT OBJ(<JOURNAL_LIBRARY>/<JOURNAL_NAME>) OBJTYPE(*JRN) USER(<CDC_USER>) AUT(*USE *OBJEXIST)

GRTOBJAUT OBJ(<RECEIVER_LIBRARY>) OBJTYPE(*LIB) USER(<CDC_USER>) AUT(*EXECUTE)
GRTOBJAUT OBJ(<RECEIVER_LIBRARY>/*ALL) OBJTYPE(*FILE) USER(<CDC_USER>) AUT(*USE)

To execute the commands through a Java Database Connectivity (JDBC) tool, use this command:

CALL QSYS2.QCMDEXC(‘CL COMMAND’);

​
Add the DB2 for i Connector

​
Authenticate to DB2 for i

After you add the connector, you need to set the required properties.
  • Server: Enter the address or host name of the server. The default server is localhost.
  • Database: Enter the name of your database.
  • Port: Enter the port number for your server. The default port is 446.
  • User: Enter the username that you use to authenticate to the database.
  • Password: Enter the password that you use to authenticate to the database.

​
Complete Your Connection

To complete your connection:
  1. Define advanced connection settings on the Advanced tab. (In most cases, though, you should not need these settings.)
  2. Click Create & Test to create your connection.

​
Create a Change Data Capture (CDC) Job with Journal and Schema Selections

This section explains how to create a change data capture (CDC) job for and how to configure row identification for tables without primary keys.

​
Create the CDC Job

Follow these steps to create a CDC job for :
  1. Enter a name for your job.
  2. Select your source.
  3. Select your destination.
  4. Specify values for the following fields:
    • Journal Library
    • Journal Name
    • Source Schema
    You can select the values for each of these properties from their drop-down lists. validates these values against the system to ensure that the journal exists and is properly configured. DB2 for i CDC job creation form
  5. Click Add Job. After the job is created, the journal settings are displayed in the CDC engine settings and cannot be modified.
  6. Continue to the job’s Task tab for table selection. Only tables that meet the following criteria are displayed:
    • The table is journaled.
    • The table is associated with the selected journal.
    • The table belongs to the selected schema.
If you change the journal or schema, previously selected incompatible tables are automatically deselected.
When you create DB2 for i CDC jobs through the API, you must provide the journal library and journal name.

​
Configure Row Identification for Tables without Primary Keys

When you configure a change data capture (CDC) job for , you can use the relative record number (RRN) as the value for the _cdatasync_id column. This approach provides a stable identifier for tables that do not have a primary key. To use the RRN as the identifier, set the following option: SyncIdSourceColumn=Db2Rrn When you set this option, uses the RRN instead of the default MD5-based row hash to generate the _cdatasync_id value.
For jobs that you create in v26.2, the RRN is set as the identifier by default.
When you use the RRN-based identifier:
  • The _cdatasync_id column is represented as a BigInt data type.
  • The identifier is derived directly from the physical row location in the source table.
  • The _cdatasync_id column can be used as a partition key for destinations that support partitioning.
This approach is recommended for new CDC jobs that involve tables that do not have primary keys. Existing CDC jobs continue to use the hash-based identifier by default to maintain backward compatibility.
Reorganizing the source table resets the RRN values for all rows. This change invalidates previously generated _cdatasync_id values because they no longer correspond to the same physical rows. To correct this issue, you must perform a full re-sync of the affected tables. For details, see Re-Syncing Job Task Execution.