[
https://issues.apache.org/jira/browse/GERONIMO-3343?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12515003
]
Kevan Miller commented on GERONIMO-3343:
----------------------------------------
This problem was discussed previously in GERONIMO-3305.
There may be more problems, but I would simplify the above operations. Simply
deploy/undeploy DT. You'll see a ClassLoader memory leak caused by OpenJPA.
Likely that Matt's additional steps are just speeding up the memory leak.
I've had discussions about the problem on OpenJPA's dev list. A Jira was
created for the problem -- https://issues.apache.org/jira/browse/OPENJPA-285. I
don't think the patch attached to the issue will fix the problem.
I had my own attempt to fix the issue. However, ran into one final problem. I
don't currently see a way around the problem -- other than introducing an
OpenJPA-proprietary interface to inform OpenJPA of ClassLoader lifecycle events
(in particular when a ClassLoader is going away...).
I've been distracted for a bit on other issues. I'll dig up where I've gotten
to...
> Out Of Memory Error when deploying and undeploying applications
> ---------------------------------------------------------------
>
> Key: GERONIMO-3343
> URL: https://issues.apache.org/jira/browse/GERONIMO-3343
> Project: Geronimo
> Issue Type: Bug
> Security Level: public(Regular issues)
> Reporter: Matt Hogstrom
> Fix For: 2.0
>
>
> Using DayTrader 2.0 if you deploy the app.
> Create the tables and populate them
> Log in and do a few trans
> Undeploy the app
> Deploy the app
> We get an out of memory error in permgen space.
> Here is the app and plan used to re-create this problem:
> http://people.apache.org/~hogstrom/daytrader-2.0-SNAPSHOT/
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
