Is every container-managed DataSource configured in TC (e.g. via server.xml
/ context.xml) automatically made pooled using commons pooling?
Yes. No additional code is needed (i.e. TC takes care of the pool.)
Or can
No, as it would not be compliant. It might be possible to create a pool of one connection, but it would still be handled as a pool.ordinary non-pooled connections be created too using this mechanism?
To answer your original questions:
To meet the J2EE requirement the DataSource method must be made avalible in a pooled configuration.Now, I'm *not* criticising the docs, I'm seeking clarification. I'm not sure whether this is saying that a JDBC DataSource has to be pooled? Or is instead saying that J2EE requires pooled JDBC connections, or maybe that J2EE requires connections via a DataSource?
I didn't think that DataSources had to be pooled, based on what is says here: http://java.sun.com/j2se/1.4.2/docs/api/javax/sql/DataSource.html
This is only the API for the language. The language may be able to do many things that are not compliant with the J2EE standard reguarding the server. Think about if you were writing a stand alone desktop app and had it talking to a datasource. The J2EE requirements for the server are of no interest and don't apply.
This might sound like nit picking, but the answer to this will help me help someone else on another thread. Thanks.
Hope they get the picture.
Doug www.parsonstechnical.com
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
