"This is very likely to create a memory leak"

[f***@atlassian.com]

Searched on this and didn't see anything, which surprised me. My search ability is weak I guess.

New install of 4.3 stand-alone
CentOS 5.6
MySQL 5.1
Java 1.6.0_17

Everything is working fine, but when I shutdown Jira I get about 50 messages like this:

{code}Apr 20, 2011 2:28:37 PM org.apache.catalina.loader.WebappClassLoader clearThreadLocalMap
SEVERE: The web application [] created a ThreadLocal with key of type [com.atlassian.plugin.util.ClassLoaderStack$1] (value [com.atlassian.plugin.util.ClassLoaderStack$***@4de3afa9]) and a value of type [java.util.LinkedList] (value [[]]) but failed to remove it when the web application was stopped. This is very likely to create a memory leak.
{code}

Any help in figuring out why we're getting these messages?

Thanks!
--
Post by crsoftware - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

Take a look at your disk space on your server and make sure that you are not running out.
--
Post by jmacleod - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

Hi Jo-Anne. Thanks for the quick response.

It's a brand new server. 12 cores, 16GB RAM, and Jira is installed on one of three 300GB SAS drives.

Let me also add that the CentOS 5.6 is 64-bit. I should have included that originally.
--
Post by crsoftware - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

Hmm...then I guess it won't be disk space. :)

Is this a big system? Have you tried to allocate more memory.
http://confluence.atlassian.com/display/CONF28/Fix+Out+of+Memory+errors+by+Increasing+Available+Memory
--
Post by jmacleod - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

Hi Jo-Anne,

I did allocate more memory from the beginning. The current values in setenv.sh are:

{code}JVM_MINIMUM_MEMORY="512m"
JVM_MAXIMUM_MEMORY="1024m"{code}

That's a little more than double the default values I think.

MySQL 5.1

Thanks
--
Post by crsoftware - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

I have seen some other posts where "To prevent a memory leak, the ThreadLocal has been forcibly removed." was added to the end of their error messages, but there's not clause like that in mine; that's why I'm concerned.
--
Post by crsoftware - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

I +think+ I can ignore the errors safely. I've stopped and started Jira about 6 times while monitoring memory usage and each time it gives me back all the RAM it used while it was running.

I hate having all of these errors cluttering catalina.out, but I guess there's nothing I can do about it. I've seen other errors like this reported by Tomcat users and the answer in those cases was basically "don't worry about it." My errors don't say that ThreadLocal is being forcibly removed, but in my tests it seems like it is.
--
Post by crsoftware - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

Probably the best idea.
--
Post by jmacleod - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

If you take a look at this thread: http://echo.nextapp.com/site/node/6254 It does point to the shutdown using more memory. I would certainly try increasing your memory allocated to JIRA to see if that resolves the problem.
--
Post by jmacleod - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

<a href="http://www.baiyunbag.com/"><strong>replica burberry handbags</strong></a>
<a href="http://www.baiyunbag.com/"><strong>fendi scarves</strong></a>
<a href="http://www.baiyunbag.com/"><strong>replica gucci sale</strong></a>
<a href="http://www.paypalcoachpurses.com/Cheap-DG-Handbags-category148.html">Cheap D&G Handbags</a>
<a href="http://www.paypalcoachpurses.com/Cheap-Versace-Handbags-category149.html">Cheap Versace Handbags</a>
<a href="http://www.paypalcoachpurses.com/Cheap-Louis-VuittonHandbags-category150.html">Cheap Louis VuittonHandbags</a>
<a href="http://www.cndesignerbags.com/"><strong>fake burberry handbags</strong></a>
<a href="http://www.cndesignerbags.com/"><strong>replica burberry handbags</strong></a>
<a href="http://www.cndesignerbags.com/"><strong>imitation burberry handbags</strong></a>
--
Post by mbtdeng - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

You are actually correct that these errors are essentially harmless. It's a new memory leak detection feature introduced in Tomcat 6.0.26, which will drop these messages into the logs when it has to forcibly unregister objects. There's more information in [this knowledge base article|http://confluence.atlassian.com/display/CONFKB/Tomcat+6.0.26+or+higher+Shutdown+Reports+'A+web+application+created+a+ThreadLocal+....+ThreadLocal+has+been+forcibly+removed'] that should be of help.

Cheers,
Adam Laskowski
Atlassian Support
--
Post by alaskowski - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348

[f***@atlassian.com]

Thanks Adam, that removes my worries. FYI I saw this: http://wiki.apache.org/tomcat/MemoryLeakProtection

I see Jira 4.3 standalone uses Tomcat 6.0.32.
--
Post by crsoftware - online at:
http://forums.atlassian.com/thread.jspa?forumID=46&threadID=53348