Debugging Resident Set Size growth

[Chris Burroughs]

On 04/15/2011 12:04 PM, Coleen Phillimore wrote:
> Hi,
> We had another unverifiable sighting of this sort, so I'd like to follow
> up and see if we can figure out what is going on.   A couple of questions:
> 

Thanks, as I said in the ticket below I know there probably isn't enough
to figure out the problem, but I'm happy to set up further
instrumentation as needed.

> 1. Was this not a problem with earlier releases of jdk6 on RHEL5?  Did
> this just start with 1.6.0_23?
> 

It occurred with 1.6.0_20 as well.

> 2.  Running with valgrind would be a good idea.  It has been used to
> find leaks in the VM in the past (6980262) but the VM mostly uses fixed
> memory in the permgen for internal VM structures and other fixed size
> arenas, so the leak may also be in the jdk native libraries or the
> application native libraries.
> 

I'll look into doing this.

> This isn't really a hotspot open source development question, can you
> file a bug?
>  http://java.sun.com/webapps/bugreport/crash.jsp
> 

Created but not yet visible.  I'll move any further discussion there:

http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7037080