> I've done some more tests with Jaybird 1.5.2, but the problem with
> statement pooling still exists.

After Carsten provided me a test case I have identified the issue. There's
one reference left on the closed connection (so memory leak exists also when
statement pooling is disabled, but the size of the leak is very small). I do
not have fix yet, from one point of view it is trivial, but from the other,
one has to be careful with event notifications when fixing. I will release
JayBird 1.5.4 after the fix.

Roman