Terracotta Discussion Forums (LEGACY READ-ONLY ARCHIVE)

Hi there,

Working on a project, we've recently started to integrate EhCache 2.7.0 as a 2nd level caching mechanism for Hibernate 3.2.7. Recently, we've started getting EXCEPTION_ACCESS_VIOLATIONs being thrown by the JVM. All of these, without fail, are being thrown from

Code:

 #Problematic frame:
 #J org.terracotta.statistics.jsr66e.LongAdder.add(J)V
 ...
 J org.terracotta.statistics.jsr166e.LongAdder.add(J)V
 J org.terracotta.GeneralOperationStatistic.end(Ljava/lang/Enum;)V
 J net.sf.ehcache.store.MemoryStore.get(Ljava/lang/Object;)Lnet/sf/ehcache/Element;
 J net.sf.ehcache.hibernate.regions.EhcacheTransactionalDataRegion.get(Ljava/lang/Object;)Ljava/lang/Object

We don't APPEAR to be running out of PermGen space (I have enabled UseConcMarkSweepGC in conjunction with CMSClassUnloadingEnabled). JRE is 1.6.0_18-b07.

Has anyone else seen this particular error?

At this stage, upgrading the JRE/JDK isn't an option, and upgrading Hibernate to the latest version probably isn't going to fly either. Any other suggestions are welcome, even if they could just help us narrow down the root cause!

Many thanks

Obviously this is JVM bug, but if you can reproduce this on a recent JVM (i.e. the most recent 1.6 or 1.7 release) then it's obviously something we'd want to look in to (to at least find a workaround). Given that you're currently using a 3 year old JVM though, there really isn't much I can think of that would help you at the moment (other than using an earlier version of Ehcache).

Sorry I couldn't be more helpful!

Hi there Chris,

Many thanks for your reply. I've actually rolled us back to EhCache 2.6.0 for the time being, and the issue seems to have gone away. Running some fairly chunky load tests of the kind that cause this JVM exception in 2.7.0 don't seem to phase the earlier version so we'll stick with that for now.

If I get the opportunity I'll try this against a 1.7 JVM and see what happens, as you suggest. I can't really see how it would be a problem with the EhCache release itself, given the failure mode, and the abject lack of anyone else complaining of such a problem, but it's interesting that it always went bang in the LongAdder.

If I find anything conclusive, I'll be sure to report back.

All the best

Hi,

I am using the EhCache 2.8.0.
I was faced this issue using jvm 1.6_32.
With the last 1.6_45 everything is working.