[jbossts-issues] [JBoss JIRA] Updated: (JBTM-352) Memory leak in TransactionReaper._timeouts

Ivan Szanto (JIRA) jira-events at lists.jboss.org
Fri Mar 28 09:38:54 EDT 2008


     [ http://jira.jboss.com/jira/browse/JBTM-352?page=all ]

Ivan Szanto updated JBTM-352:
-----------------------------

    Attachment: TransactionReaper.zip

Andrew, thanks for the feedback with lots of useful info! 

I need a fix quickly, therefore I attach a new fix based on this information:

"The reason this is failing is because the comparison is being done the wrong way round i.e. by calling Uid.equals(ControllWrapper) rather than ControlWrapper.equals(Uid)."

In this fix, if a Uid is supplied to TransactionReaper.remove, then it retrieves the keyset from _timeouts, iterates over it and finds the Reapable that has the Uid supplied. It then uses the Reapable it found to remove it from _timeouts.

It will probably cause performance problems if there are many items in _timeouts, but at least it makes the memory leak go away (I tested it with the same test case as the previous one). 

Do you think it's a valid fix?

> Memory leak in TransactionReaper._timeouts
> ------------------------------------------
>
>                 Key: JBTM-352
>                 URL: http://jira.jboss.com/jira/browse/JBTM-352
>             Project: JBoss Transaction Manager
>          Issue Type: Bug
>      Security Level: Public(Everyone can see) 
>          Components: Transaction Core
>    Affects Versions: 4.3.0.GA
>         Environment: Compiled 4.3.0.GA for the EAP 4.3 as JTS according to the instructions (by setting JBOSS_HOME to jboss-5.0.0.Beta4) on RHEL4.
> Run ~ 1500 JMS transactions in EAP 4.3 all configuration on RHEL4 to see it leaks.
> Used Sun jdk 1.5.0.14 
>            Reporter: Ivan Szanto
>         Assigned To: Andrew Dinn
>             Fix For: 4.4.CR1
>
>         Attachments: TransactionReaper.java, TransactionReaper.zip
>
>
> The leak can also be seen by using the debugger to check the number of entries in TransactionReaper._timeouts after a couple of transactions.
> It appears that the TransactionReaper.insert method uses ControlWrapper objects as keys when putting elements into the hashtable, but the TransactionReaper.remove method sometimes unsuccessfully tries to use Uid ojbects to remove them.
> I will post a corrected version of TransactionReaper.java.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        



More information about the jbossts-issues mailing list