On Mon, 3 Jan 2011 08:13:20 -0600, "Caldarale, Charles R"
<chuck.caldar...@unisys.com> wrote:
I suspect you're not using the word "instantly" appropriately here...
You're right, should be constantly rather.
PermGen can steadily increase if there's something in your
environment that creates new, typically anonymous, classes as part of
its operation. PermGen will not decrease until a full GC occurs,
which won't happen unless either the old gen or PermGen fills up (or
a
full GC is forced).
Simple increased usage of PermGen is *not* necessarily indicative of
a memory leak, but not returning to a stable value *after* a full GC
would be.
I thought, that PermGen data cannot be garbage collected...
--
Mikolaj Rydzewski <m...@ceti.pl>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
