JDBC 2.0 with standard extensions provides two mechanisms for connecting to the database:
-
The
javax.sql.DataSourceinterface provides unmanaged physical connections. -
The
javax.sql.ConnectionPoolDataSourceinterface provides managed physical connections. To connect to a connection pooling data source, you need a managedConnectionPoolDataSourceconnection.
Using a connection pool with a JDBC resource
By default, a JDBC resource does not use connection pooling. A connection pool may, however, be attached to a JDBC resource to improve efficiency. When a JDBC resource uses connection pooling, the way Rhino manages connections depends on what interface the datasource class of the persistence instance used by the JDBC resource is an implementation of, as follows:
| Interface | How Rhino manages connections |
|---|---|
javax.sql.DataSource |
uses an internal implementation of |
javax.sql.ConnectionPoolDataSource |
uses managed connections from the |
Connection pool configurable parameters
A connection pool has the following configurable parameters:
| Parameter | What it specifies |
|---|---|
max-connections |
Maximum number of active connections a Rhino process can use at any one time. |
max-idle-connections |
Maximum number of inactive connections that should be maintained in the connection pool. This value must be less than or equal to |
min-connections |
Minimum number of connections that should be maintained in the connection pool. |
max-idle-time |
Time in seconds after which an inactive connection may become eligible for discard.
An idle connection will not be discarded if doing so would reduce the number of idle connections below the If this parameter has the value |
idle-check-interval |
Time in seconds between idle connection discard checks. |
connection-pool-timeout |
Maximum time in milliseconds an SBB will wait for a free connection before a timeout error occurs. |
