Configuring JDBC Replication
In order to get started with JDBC replication, you need at first to create the table and fields where the Http Session will be stored. Download and execute the sql script from http://bit.ly/14jF5Ya:
SQL> @jdbc_replication_oracle.sql
Now verify the presence of a new table named WL_SERVLET_SESSIONS:
SQL> select count(*) from WL_SERVLET_SESSIONS;
COUNT(*)
– – – – – – – – – – 0
As last step, you will need to add to your weblogic.xml the specific Data Source information:
Modify the existing <session-descriptor> element as follows:
<session-descriptor>
<persistent-store-type>jdbc</persistent-store-type>
<persistent-store-pool>jdbc/OracleDS</persistent-store-pool>
<persistent-store-table>WL_SERVLET_SESSIONS</persistent-store-table>
</session-descriptor>
Note: In order to work, you need to target the jdbc/OracleDS data source to the cluster:
Choosing between JDBC Session replication and In-Memory replication
When using JDBC Session replication it’s not necessary to define a primary-secondary memory storage for your sessions because all members of the cluster have full access to the session data. This comes however at a price, that is, slower applications if your JDBC based Session replication stores large sets of data in the Session.
