[jboss-jira] [JBoss JIRA] (WFLY-12671) Memory Leak of PersistenceUnitMetadataImpl

[Scott Marlow (Jira)]

    [ https://issues.jboss.org/browse/WFLY-12671?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13800050#comment-13800050 ] 

Scott Marlow commented on WFLY-12671:
-------------------------------------

[~rschimpf] I wonder if the leak still happens if you change how you redeploy your application.  After you have deployed app.ear and want to redeploy,  could you try:
# Delete the app.ear from wildfly/standalone/deployments folder first.
# Wait a few seconds for the undeploy to complete for app.ear.
# Copy app.ear back into wildfly/standalone/deployments folder.
# Check if this avoids the leak or not.

We just released [WildFly 18.0.0.Final|https://wildfly.org/downloads/] as well,  it would be interesting to know if the leak also recreates with WildFly 18.0.0.Final.

In terms of examining the heap dump, be sure to check that your memory tool is ignoring Java objects that are eligible to be garbage collected but haven't yet.  For example if there are zero strong (Java) references to the various PersistenceUnitMetadataImpl objects, they just haven't been GC'd yet.  What tool are you using for examining the memory leak?

Thanks,
Scott



> Memory Leak of PersistenceUnitMetadataImpl
> ------------------------------------------
>
>                 Key: WFLY-12671
>                 URL: https://issues.jboss.org/browse/WFLY-12671
>             Project: WildFly
>          Issue Type: Bug
>          Components: JPA / Hibernate
>    Affects Versions: 17.0.1.Final
>            Reporter: Robin Schimpf
>            Assignee: Scott Marlow
>            Priority: Major
>         Attachments: Trying_to_catch_memory_leak_with_more_trace_logging.patch, memory leak after 4 deployments.PNG, memory leak after 5 deployments.PNG, persistence-unit-not-stopped.txt, persistence-unit-stopped.txt
>
>
> After updating from Wildfly 13 to Wildfly 17 we noticed a slowdown of our application after multiple deployments to the application server without restarting it between the deployments. Heapdumps are showing that after redeployment the number of {{PersistenceUnitMetadataImpl}} are growing (see screenshots).
> We are deploying 2 ear files and 2 war files. Only one ear has a database connection configured. We are using the bundeled hibernate version of Wildfly 17.
> The ear structure is like the following:
> {noformat}
> - app.ear
>   |
>   |- META-INF
>   |
>   |- lib
>   |   |
>   |   |- all dependencies
>   |
>   |- own jars and wars (e.g. jpa)
> {noformat}
> Since I failed to create a reproducing project for this behaviour I added some trace logging (Patch attached). The only difference I noticed was that sometimes the {{Stopping Persistence Unit}} logstatement is missing before the {{Stopped subdeployment}} logstatement.



--
This message was sent by Atlassian Jira
(v7.13.8#713008)