Java PermGen out of memory exception

[fredliu]

Hi,

I'm getting the PermGen exception from our web service. Just want to check other than analyzing through memory tools, are there any known bad practices which may cause PermGen memory leak? Thanks

The code of the sevlet
https://cs.corp.google.com/#piper//...acesaver/web/nanny/BookingRemoverServlet.java

I've bumped up the memory size but the issue still exists.
"-XX:MaxPermSize=160m",

E0327 16:52:17.790999 56 GSE worker #22 HttpConnectionImpl.run caught a throwable
java.lang.OutOfMemoryError: PermGen space
at java.lang.ClassLoader.defineClass(ClassLoader.java:795)
at java.security.SecureClassLoader.defineClass(SecureClassLoader.java:142)
at java.net.URLClassLoader.defineClass(URLClassLoader.java:449)
at java.net.URLClassLoader.access$100(URLClassLoader.java:71)
at java.net.URLClassLoader$1.run(URLClassLoader.java:361)
at java.net.URLClassLoader$1.run(URLClassLoader.java:355)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(URLClassLoader.java:354)
at java.lang.ClassLoader.loadClass(ClassLoader.java:427)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.ja

 

[Arne Vajhøj]

I'm getting the PermGen exception from our web service. Just want to check other than analyzing through memory tools, are there any known bad practices which may cause PermGen memory leak? Thanks

The code of the sevlet
https://cs.corp.google.com/#piper//...acesaver/web/nanny/BookingRemoverServlet.java

I've bumped up the memory size but the issue still exists.
"-XX:MaxPermSize=160m",

E0327 16:52:17.790999 56 GSE worker #22 HttpConnectionImpl.run caught a throwable
java.lang.OutOfMemoryError: PermGen space
at java.lang.ClassLoader.defineClass(ClassLoader.java:795)

PermGen is used for classs definitions.

If you are leaking it then it is often because you are dynamically
generating classes and loading them continuously.

But maybe you are not leaking but just need a lot.

160 MB is not that much. You could try with 300 MN and see if the app
can keep running.

Arne

 

[Roedy Green]

I'm getting the PermGen exception

see http://mindprod.com/packratting.html
--
Roedy Green Canadian Mind Products http://mindprod.com
"A great lathe operator commands several times the wage of an average lathe
operator, but a great writer of software code is worth 10,000 times the
price of an average software writer."
~ Bill Gates

 

[Arne Vajhøj]

see http://mindprod.com/packratting.html

Link does not work.

http://mindprod.com/jgloss/packratting.html

does.

But since the text does not relate to PermGen, then it
is irrelevant anyway.

Arne

 

[Jeff Higgins]

But maybe you are not leaking but just need a lot.

This thread led me here. Thanks.
<http://docs.oracle.com/javase/7/docs/webnotes/tsg/TSG-VM/html/toc.html>

 

[Jeff Higgins]

This thread led me here. Thanks.
<http://docs.oracle.com/javase/7/docs/webnotes/tsg/TSG-VM/html/toc.html>

Generally <http://docs.oracle.com/javase/7/docs/webnotes/tsg/>
Specifically
<http://docs.oracle.com/javase/7/docs/webnotes/tsg/TSG-VM/html/memleaks.html>

 

[Jeff Higgins]

Generally <http://docs.oracle.com/javase/7/docs/webnotes/tsg/>
Specifically
<http://docs.oracle.com/javase/7/docs/webnotes/tsg/TSG-VM/html/memleaks.html>

Well, it seems the Java 8 docs are easier to get to:
<http://docs.oracle.com/javase/8/docs/>
<http://docs.oracle.com/javase/8/docs/technotes/guides/tsgvm/index.html>
<http://docs.oracle.com/javase/8/docs/technotes/guides/tsgvm/memleaks.html#sthref38>

 

[Arne Vajhøj]

Well, it seems the Java 8 docs are easier to get to:
<http://docs.oracle.com/javase/8/docs/>
<http://docs.oracle.com/javase/8/docs/technotes/guides/tsgvm/index.html>
<http://docs.oracle.com/javase/8/docs/technotes/guides/tsgvm/memleaks.html#sthref38>

Hmm.

I thought that permgen was gone in Java 8??

Arne

 

[Jeff Higgins]

On 3/28/2014 11:04 PM, Jeff Higgins wrote:

Hmm.

I thought that permgen was gone in Java 8??

? According to JEP 122 removing the permgen space was a funded project.
Maybe that didn't extend to updating the documentation.

 

[Jeff Higgins]

? According to JEP 122 removing the permgen space was a funded project.
Maybe that didn't extend to updating the documentation.

Ah. Current state is listed as Funded, not Complete.
<http://openjdk.java.net/jeps/122>
See: Process States
<http://openjdk.java.net/jeps/1>

 

[Arne Vajhøj]

Ah. Current state is listed as Funded, not Complete.
<http://openjdk.java.net/jeps/122>
See: Process States
<http://openjdk.java.net/jeps/1>

That document also has:

"Updated 2012/12/4"

And
http://www.oracle.com/technetwork/java/javase/8-whats-new-2157071.html
states:

<quote>
* HotSpot
....
* Removal of PermGen.
</quote>

Arne

 

[Jeff Higgins]

That document also has:

"Updated 2012/12/4"

And
http://www.oracle.com/technetwork/java/javase/8-whats-new-2157071.html
states:

<quote>
* HotSpot
...
* Removal of PermGen.
</quote>

Yep. I guess that the third "Impact" listed at the bottom of the page:
"Documentation: References to the permanent generation will need to be
removed." is one item that keeps the page from being updated with a
status of "Complete".