Because of the three-day break, my response is late.
> Valgrind is a great tool, but you must learn how to identify false
> positives and tell the difference between a leak that matters (say 1kb
> allocated and not freed in a loop that runs once per second) and a leak
> that doesn't.
I get the memory leak scenario not only from Valgrind, but also from the
output of top command.
At first I think the memory leak occur when I disconnect database by
Valgrind, then I write a test sample that just connect and disconnect
database in a infinite loop to check it. after 1 hour performation, 7M
memory is leaked.