Sorry for the noise, just after posting I finally tracked it down to a
problem in the application. In one code path a prepared statement was
being double-closed, which apparently triggers a bug in the version of
DBCP being used.
I'm still not sure why it only started misbehaving after upgrading the
JDBC driver.
-- Mark
On Mon, 2005-04-25 at 16:38, Mark Lewis wrote:
> After upgrading from pg74.215.jdbc3.jar to postgresql-8.0-310.jdbc3.jar,
> I've encountered problems getting a "This statement has been closed"
> exception when using a PreparedStatement.
>
> The application is using Jakarta's DBCP 1 connection pooling with
> prepared statement caching, and is configured to execute "select 1" each
> time a connection is checked out from the pool to ensure that the
> connection is still alive, so I'm pretty sure the server connection is
> still good.
>
> Are there any known gotchas migrating to the new server prepared
> statement implementation? Something that could be causing statements to
> close when they otherwise would not be?
>
> Thanks,
> Mark Lewis
>
> Thanks,
> Mark Lewis
>
>
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Don't 'kill -9' the postmaster