Use these steps to define a data source on multiple nodes that
comprise a cluster.
Before you begin
The first part of this task defines a JDBC provider with a cluster
scope setting. Be aware that all members of your cluster must run at least
Version 6 of WebSphere Application Server to use this scope setting for the
cluster. (See
Administrative console scope settings for
more information about scope settings in general.)
Why and when to perform this task
The cluster scope has precedence over the node and cell scopes. Create
a data source for a cluster if you want the data source to:
- Be available for all the members of this cluster to use
- Override any resource factories that have the same JNDI name that is defined
within the cell scope
Steps for this task
- Open the administrative console.
- Click Resources > JDBC Providers. In
the Scope section, note that the default scope setting is at the node level.
- Click Browse Clusters. The JDBC providers
> Select a Cluster Scope panel is displayed.
- Select the cluster for which you want to define a data source,
and click OK. The JDBC providers panel is displayed
again.
- Click Apply.
- Click New to create a new JDBC provider at the cluster level.
The class path for your new JDBC provider is already filled in; part
of that class path is specified using a symbolic variable, for example: ${DB2390_JDBC_DRIVER_PATH}/classes/db2j2classes.zip .
Leave it at the default.
- Finish creating the JDBC provider.
- Click Environment.
- Click WebSphere Variables.
- For each node, select the symbolic variable used in the class path
of your JDBC provider, and provide a value that is appropriate for the selected
node. For example, if the class path of your JDBC provider uses
the symbolic variable ${DB2390_JDBC_DRIVER_PATH}, you might supply
the value /usr/lpp/db2 on one node and /usr/lpp/db2710 on
another node, depending on where your DB2 390 installation is located.
- Click DB2_JDBC_DRIVER_PATH (this already exists by default).
Here provide the path (in the value field) where db2java.zip exists
on the selected node.
- Click Apply and save the changes.
Note: This
variable must be defined on each node within the cluster.