IBM DB2 Universal Database Administration Client

If the Replication Agent for UDB software is installed on a different host machine from the IBM DB2 Universal Database server, you must install the IBM DB2 Universal Database Administration Client on the same host machine as the Replication Agent.

If the Replication Agent for UDB software is installed on the same host machine as the IBM DB2 Universal Database server, a separate IBM DB2 Universal Database Administration Client is not required.

On a Windows system, you must configure an ODBC data source in the IBM DB2 Universal Database Administration Client, then use the database name and database alias specified for that ODBC data source when you configure Replication Agent for UDB connectivity.

On a UNIX system, instead of using ODBC, simply catalog the node and the primary database in UDB. Then, set the Replication Agent pds_datasource_name parameter to the database alias.

1. Log in as the DB2 instance owner.

2. Set up the DB2 environment.

   In Korn shell, source the db2profile file:

   . /home/db2inst1/sqllib/db2profile
   

   In C shell, source the db2cshrc file:

   source /home/db2inst1/sqllib/db2cshrc
   

3. Start the DB2 command-line processor by typing the db2 command.

4. Catalog the remote TCP/IP node using the following command at the DB2 prompt:

   catalog tcpip node MYNODE remote MYHOST server XXXX
   

   Here, MYNODE is the node name, MYHOST is the host name or IP address of the data server, and XXXX is the data server port number.

5. Verify the catalog entry:

   list node directory
   

   DB2 should return something similar to the following:

   Node 1 entry:
         Node name            = MYNODE
         Comment              = 
         Directory entry type = LOCAL
         Protocol             = TCPIP
         Hostname             = MYHOST
         Service name         = XXXX
   

1. Catalog the primary database using the following command at the DB2 prompt:

   catalog database MYDB as MYDB_ALIAS at node MYNODE
   

   Here, MYDB is the database name, MYDB_ALIAS is an alias for the database, and MYNODE is the node name used in the catalog command.

2. Verify the catalog entry:

   list database directory
   

   DB2 should return something similar to the following:

   System Database Directory
   
   Number of entries in the directory = 1
   
   Database 1 entry:
   
   Database alias         = MYDB_ALIAS
   Database name          = MYDB
   Node name              = MYNODE
   Database release level = b.00
   Comment                = 
   Directory entry type   = Remote
   

1. In Replication Agent, set the pds_datasource_name parameter to the database alias:

   ra_config pds_datasource_name, MYDB_ALIAS
   

   Here, MYDB_ALIAS is the database alias that was used when cataloging the primary database.

2. Also set the following Replication Agent parameters:

   • pds_database_name

   • pds_username

   • pds_password

   For detailed information on these parameters, see the Replication Agent Reference Manual.