PlantsByWebSphere on WebSphere Application Server 7.0
Cluster Considerations for PlantsByWebSphere
The embedded Derby database is not intended to run in a clustered environment. To
work around this issues use the steps below, but be aware that in doing the following,
each application is running it's own database. Apply the steps below when creating the JDBC Providers and
Data Sources. Create a separate instance of JDBC Providers and Data Sources for each server
in the cluster.
- In the ${APP_INSTALL_ROOT}/${CELL}/PlantsByWebSphereAjax.ear/Database/ directory create
a uniquely named deep copy of PLANTSDB for each server in the cluster.
For example for server2 create a copy of PLANTSDB and all its contents named PLANTSDB_2.
- When defining the Derby XA Provider and specifying the name, specify a server specific name.
For example:
- For server1 enter "server1 Derby Provider for PlantsByWebSphereAjax (XA)"
- For server2 enter "server2 Derby Provider for PlantsByWebSphereAjax (XA)"
- When Defining the Derby XA Provider do the following:
- For the data source name enter a server specific name which corresponds to the
directory names you created above. For example for server1 use PLANTSDB
and for server2 use PLANTSDB_2.
- When selecting an existing JDBC Provider select the server specific name which you defined earlier, such as
- server1 Derby Provider for PlantsByWebSphereAjax (XA) and
- server2 Derby Provider for PlantsByWebSphereAjax (XA).
- When specifying the database name enter the server specific name corresponding to the
original or new directories you created. For example:
- For server1 use ${APP_INSTALL_ROOT}/${CELL}/PlantsByWebSphereAjax.ear/Database/PLANTSDB
- For server2 use ${APP_INSTALL_ROOT}/${CELL}/PlantsByWebSphereAjax.ear/Database/PLANTSDB_2