Re: Totally weird behaviour in org.postgresql.Driver

On Thursday 12 March 2009 08:22:53 am Peter wrote:
> Obiously I do not need to synchronize the method. the synchronized block
> inside it should work just fine. Was just grasping at straws there.
>
>
>
> I checked out dbcp and c3po (thanks for the pointers!) but none fits my app
> as each user needs to connect with his own credentials.

This was a problem for me as well.  So, I made a factory for returning
connections from dbcp and made some adjustments to the connection returned
from the pool, so that our auditing triggers knew the application user  that
was using the connection.

That way we could get the auditing and use a pool that has hopefully had the
bugs beaten out of if.  But, that would not work in every situation.


C.


--
When I'm on my own I'm in bad company.
                               -- David Bowie