Now that we're using Oracle's CLOB hack instead of our own, we no longer need to require that Oracle connections go through a server-side connection pool.

What we do need to do, is make sure that we set the SetBigStringTryClob=true parameter on the connection. Unlike most drivers, this can't be done via the connection URL, we need to save it as a hibernate property in confluence.cfg.xml: i.e. hibernate.connection.SetBigStringTryClob=true

(I did this manually down in Melbourne by shutting down Confluence, editing the config file and starting up again, and it worked fine.)

We should probably do this for other required connection properties too, like the MySQL ones.