Skip to content

Frameworks.md

Latest commit

 

History

History
61 lines (47 loc) · 3.79 KB

Frameworks.md

File metadata and controls

61 lines (47 loc) · 3.79 KB
Raw

The AWS Advanced JDBC Wrapper Integration with 3rd Party Frameworks

The AWS Advanced JDBC Wrapper works with multiple 3rd party frameworks. Here are some examples of configuring the driver to work with these frameworks:

Open Liberty

Open Liberty is an open-source Jakarta EE and MicroProfile application server from IBM. The AWS Advanced JDBC Wrapper can be used as the JDBC driver within Open Liberty's dataSource configuration to provide failover support, authentication, and other AWS database features to applications deployed on Liberty.

Configuring Liberty with Failover

The AWS Advanced JDBC Wrapper uses the following SQL states for failover exceptions:

  • 08S02 — Failover succeeded, the active connection has changed.
  • 08007 — Failover occurred during an active transaction, transaction outcome is unknown.
  • 08001 — Failover was attempted but failed, no valid connection is available.
  • 08003 — Operation attempted on a connection that is no longer open.

Open Liberty natively recognizes 08001, 08003, 08006, and 08S01 as stale connection codes. The SQL states 08S02 and 08007 are not recognized by default and require further configuration.

To ensure the AWS Advanced JDBC Wrapper's failover behavior is handled correctly by the Connection Manager, configure identifyException to map the wrapper's failover SQL states as stale connections. This allows Liberty to discard pooled connections and cached statements after a failover event, enabling the application to recover automatically.

Add the following to your dataSource configuration in server.xml when using the failover, failover2, or gdbFailover plugins:

<dataSource id="default" jndiName="jdbc/myDS" type="javax.sql.DataSource">
  <identifyException as="StaleConnection" sqlState="08S02"/>
  <identifyException as="StaleConnection" sqlState="08007"/>
  ...
</dataSource>

As an alternative, to discard cached statements but not pooled connections, setting statementCacheSize="0" on the dataSource element disables statement caching entirely and avoids the need for identifyException configuration at the cost of re-preparing statements on every use.

Example DataSource Configuration

Configure the AWS Advanced JDBC Wrapper as a javax.sql.DataSource using AwsWrapperDataSource. Wrapper connection properties such as wrapperPlugins, wrapperDialect, and clusterId are passed through targetDataSourceProperties. See Connecting with a DataSource for more details.

<library id="aws-advanced-jdbc-wrapper">
    <fileset dir="/config/aws-advanced-jdbc-wrapper" includes="*.jar"/>
</library>

<dataSource id="default" jndiName="jdbc/myDS" type="javax.sql.DataSource">
  <identifyException as="StaleConnection" sqlState="08S02"/>
  <identifyException as="StaleConnection" sqlState="08007"/>
  <jdbcDriver libraryRef="aws-advanced-jdbc-wrapper"
              javax.sql.DataSource="software.amazon.jdbc.ds.AwsWrapperDataSource"/>
  <properties
    serverName="db-identifier.cluster-XYZ.us-east-2.rds.amazonaws.com"
    serverPort="5432"
    user="username"
    password="password"
    database="employees"
    jdbcProtocol="jdbc:postgresql:"
    targetDataSourceClassName="org.postgresql.ds.PGSimpleDataSource"
    targetDataSourceProperties="wrapperPlugins: auroraConnectionTracker,efm2,failover2; wrapperDialect: aurora-pg; clusterId: my-cluster" />
</dataSource>