java.lang.OutOfMemoryError: requested 32744 bytes for ChunkPool::allocate. Out of swap space?

[Vince M. Kostoff]

I am getting the following error:

java.lang.OutOfMemoryError: requested 32744 bytes for ChunkPool::allocate. Out of swap space?
Possible causes:
- not enough swap space left, or
- kernel parameter MAXDSIZ is very small.

My environment is:
Java version: 6.0-011
Operating System: OpenVMS 8.3
java version "6.0"
Java(TM) SE Runtime Environment
Java HotSpot(TM) 64-Bit Server VM (build 1.6.0-1 06/25/2009-12:58 IA64 (c2)W, mixed mode)

I have read a lot of links regarding the use of hpjtune. Is it also available for OpenVMS?

Do I have to use JMX to obtain the same results?
Thanks.
Vince

[P Muralidhar Kini]

Hi Vince,

The Java HotSpot VM cannot expand its heap size if memory is completely
allocated and no swap space is available. This can occur, for example, when
several applications are running simultaneously.
When this happens, the VM will exit after printing the following -
Exception java.lang.OutOfMemoryError: requested bytes.
Out of swap space?

If you see this symptom, consider increasing the available swap space by
allocating more of your disk for virtual memory and/or by limiting the number
of applications you run simultaneously.

For more details, Refer
http://java.sun.com/j2se/1.5.0/relnotes.html
-> Section "HotSpot VMs"

HPjtune is a Java Garbage Collection (GC) visualization tool for analyzing
garbage collection activity in a Java program. This is for HP-UX OS.
For more information -
https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=HPJTUNE

May be there is some program out there which is eating up a lot of heap space
causing it to get exhausted.

Regards,
Murali

[Joseph Huber_1]

I know nothing about Java6 on VMS, but I think the message is at least misleading:
expanding process virtual memory has nothing to do with system swap- or paging-file, and there is no such thing as "kernel MAXDSIZ", all sounds like Unix.

Process virtual memory on VMS is determined by the UAF parameter PGFLQUO: use AUTHORIZE to show the users parameters, increase PGFLQUO to see if resolves the problem.

[Rishi Singhal]

Hi Vince,

This might be due to a memory leak in the application.
I believe you get this after the application runs for long.Try using this VM arguments while running the application.
-XX:+HeapDumpOnOutOfMemoryError
The dump file that is generated can be analyzed by using Eclipse MAT (I usually do that). For that you need to copy the dump file to windows and install Eclipse MAT.

Regards,
Rishi

[Dennis Handly]

>Rishi: This might be due to a memory leak in the application.

Shouldn't java garbage collection take care of that?

[Rishi Singhal]

Hi Dennis,

Memory leak (a misnomer for java) in java mostly occurs when the application keeps on allocating VM but never returns back to the pool.
GC runs on an object only when it is not referenced but due to error in programming it might happen that an object is not dereferenced and always stays in memory. This then becomes a cause of memory leak and if an application is running for long similar objects cause the heap memory to be exhaused unallowing any further memory allocation causing outofmemory.


Chk this
http://java.sun.com/javase/6/webnotes/trouble/TSG-VM/html/memleaks.html

http://www.eclipse.org/mat/


Regards,
Rishi

[Vince M. Kostoff]

Thanks for all input. Investigating application type issues at this time. No specific resolution yet. Was able to use JConsole, but it did not provide enough details.