[ 
https://issues.apache.org/jira/browse/TAP5-2411?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14191747#comment-14191747
 ] 

Jochen Kemnade edited comment on TAP5-2411 at 10/31/14 1:02 PM:
----------------------------------------------------------------

Alright, the key is the id/index of the value in the perthread map. So it's 
probably {{org.apache.tapestry5.ioc.internal.ServiceActivityTrackerImpl}} as 
that's the only spot that uses a {{PerThreadValue<Status>}}. Could it be the 
registry startup? Should we call {{registry.cleanupThread()}} at the end of 
{{TapestryFilter.init(FilterConfig)}}?


was (Author: jkemnade):
Alright, the key is the key of the value in the perthread map. So it's probably 
{{org.apache.tapestry5.ioc.internal.ServiceActivityTrackerImpl}} as that's the 
only spot that uses a {{PerThreadValue<Status>}}. Could it be the registry 
startup? Should we call {{registry.cleanupThread()}} at the end of 
{{TapestryFilter.init(FilterConfig)}}?

> Tapestry webapp leaks a ThreadLocal
> -----------------------------------
>
>                 Key: TAP5-2411
>                 URL: https://issues.apache.org/jira/browse/TAP5-2411
>             Project: Tapestry 5
>          Issue Type: Bug
>          Components: tapestry-ioc
>    Affects Versions: 5.4
>            Reporter: Jochen Kemnade
>
> When re-deploying a Tapestry webapp in a Tomcat container, I often see a 
> warning like this:
> {noformat}
> 21-Oct-2014 12:03:37.349 SEVERE [http-nio-9280-exec-14] 
> org.apache.catalina.loader.WebappClassLoader.checkThreadLocalMapForLeaks The 
> web application [/webapp] created a ThreadLocal with key of type 
> [org.apache.tapestry5.ioc.internal.services.PerthreadManagerImpl.
> MapHolder] (value 
> [org.apache.tapestry5.ioc.internal.services.PerthreadManagerImpl$MapHolder@55d6dc57])
>  and a value of type [java.util.HashMap] (value [{1=DEFINED, 2=DEFINED, 
> 3=DEFINED, 4=DEFINED, 5=DEFINED, 6=DEFINED, 7=DEFINED, 8=DEFINED, 9=DEFINED, 
> 10=DEFINED, 
> 11=DEFINED, 12=DEFINED, 13=DEFINED}]) but failed to remove it when the web 
> application was stopped. Threads are going to be renewed over time to try and 
> avoid a probable memory leak.
> {noformat}
> I'm not yet sure where this comes from. It looks like a {{Map<Number, 
> org.apache.tapestry5.ioc.services.Status>}}.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to