Re: Memory leak found in java.util.concurrent.ForkJoinPool

2017-01-23 Thread Rory O'Donnell 



On 21/01/2017 16:56, Mark Thomas wrote:

On 21/01/2017 09:31, Konstantin Kolinko wrote:

2017-01-09 23:58 GMT+03:00 Mark Thomas :

All,

As a result of a thread on the users list, a memory leak has been
discovered in java.util.concurrent.ForkJoinPool.commonPool. It appears
that use of this pool is hard-coded in multiple places such as
Collection.parallelStream().forEach().

I have some ideas about how to work-around the leaks with
JreMemoryLeakPreventionListener but any code that depends on the Tomcat
class loader being set (e.g. JNDI lookup) is going to break if the user
tries to use Collection.parallelStream().


Rory,

The review Id for my report is 9046841. Anything you can do to get this
accepted and in front of the right folks to fix it would be appreciated.

Mark this bug is already under investigation!

By the way, JDK-8163449 - Allow per protocol setting for URLConnection 
defaultUseCaches

is included in b153, now available on java.net.

Rgds,Rory


Kind regards,
Mark

Hi!

I filed this into Tomcat Bugzilla, to keep track of our workaround code
https://bz.apache.org/bugzilla/show_bug.cgi?id=60620


As for Oracle bug number, I do not see this issue with a quick search
at http://bugs.java.com/

The review ID does not get re-used as the bug number. The bug, as
references in the Javadoc in JreMemoryLeakPreventionListener, is:
http://bugs.java.com/bugdatabase/view_bug.do?bug_id=JDK-8172726

Mark


Best regards,
Konstantin Kolinko

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org



-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org



--
Rgds,Rory O'Donnell
Quality Engineering Manager
Oracle EMEA , Dublin, Ireland


-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

Re: Memory leak found in java.util.concurrent.ForkJoinPool

2017-01-21 Thread Mark Thomas 
On 21/01/2017 09:31, Konstantin Kolinko wrote:
> 2017-01-09 23:58 GMT+03:00 Mark Thomas :
>> All,
>>
>> As a result of a thread on the users list, a memory leak has been
>> discovered in java.util.concurrent.ForkJoinPool.commonPool. It appears
>> that use of this pool is hard-coded in multiple places such as
>> Collection.parallelStream().forEach().
>>
>> I have some ideas about how to work-around the leaks with
>> JreMemoryLeakPreventionListener but any code that depends on the Tomcat
>> class loader being set (e.g. JNDI lookup) is going to break if the user
>> tries to use Collection.parallelStream().
>>
>>
>> Rory,
>>
>> The review Id for my report is 9046841. Anything you can do to get this
>> accepted and in front of the right folks to fix it would be appreciated.
>>
>> Kind regards,
>> Mark
> 
> Hi!
> 
> I filed this into Tomcat Bugzilla, to keep track of our workaround code
> https://bz.apache.org/bugzilla/show_bug.cgi?id=60620
> 
> 
> As for Oracle bug number, I do not see this issue with a quick search
> at http://bugs.java.com/

The review ID does not get re-used as the bug number. The bug, as
references in the Javadoc in JreMemoryLeakPreventionListener, is:
http://bugs.java.com/bugdatabase/view_bug.do?bug_id=JDK-8172726

Mark
> 
> 
> Best regards,
> Konstantin Kolinko
> 
> -
> To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: dev-h...@tomcat.apache.org
> 


-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

Re: Memory leak found in java.util.concurrent.ForkJoinPool

2017-01-21 Thread Konstantin Kolinko 
2017-01-09 23:58 GMT+03:00 Mark Thomas :
> All,
>
> As a result of a thread on the users list, a memory leak has been
> discovered in java.util.concurrent.ForkJoinPool.commonPool. It appears
> that use of this pool is hard-coded in multiple places such as
> Collection.parallelStream().forEach().
>
> I have some ideas about how to work-around the leaks with
> JreMemoryLeakPreventionListener but any code that depends on the Tomcat
> class loader being set (e.g. JNDI lookup) is going to break if the user
> tries to use Collection.parallelStream().
>
>
> Rory,
>
> The review Id for my report is 9046841. Anything you can do to get this
> accepted and in front of the right folks to fix it would be appreciated.
>
> Kind regards,
> Mark

Hi!

I filed this into Tomcat Bugzilla, to keep track of our workaround code
https://bz.apache.org/bugzilla/show_bug.cgi?id=60620


As for Oracle bug number, I do not see this issue with a quick search
at http://bugs.java.com/


Best regards,
Konstantin Kolinko

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org