Re: [Wrapper-user] Memory leak calling getLastErrorText()

[Tim L. <tim...@gm...>]

The leak occurs in 3.5.25 and earlier (64-bit), running on an linux machine
(linux kernel 3.2.30-49.59).
It is located in the #ifdef UNICODE block of the method:
the buffer used in the WIN32 code could also be used in the UNICODE block.


I am not sure this is the root cause of the problems we are seeing.
I wrote a small test program which used the Java logging framework and is
continuously logging a string with some unicode characters in it to
reproduce the problem.
That is also creating memory issues with the wrapper (3.5.25).
I am currently suspecting locale/glibc problems due to security updates.
I'll keep you posted if the root cause is in the service wrapper.

Regards,
Tim

On Tue, Sep 9, 2014 at 11:17 AM, Leif Mortenson <
lei...@ta...> wrote:

> Tim,
> I wanted to follow up as well.  The getLastErrorText method is known to
> have a small leak, but it is not normally called repeatedly.
> Where it is we are specifically freeing the returned string.
>
> What is the version you are seeing the leak in?  Is it something that you
> have been able to reproduce?
> There have been some other leaks fixed over time:
> http://wrapper.tanukisoftware.com/doc/english/release-notes.html
>
> Cheers,
> Leif
>
>
> On Tue, Sep 9, 2014 at 4:34 PM, Alexandre Klein <
> ale...@ta...> wrote:
>
>> Tim,
>>
>> Thank you for your email.
>> We are aware of this issue with getLastErrorText() and it is in our bug
>> tracking system.
>>
>> We will fix it in a future version of the Java Service Wrapper.
>>
>> If you see any other problem in our code, please let us know. We really
>> want to improve the Wrapper and deliver a great product to our users.
>>
>> Regards,
>> Alexandre Klein
>>
>> On Fri, Sep 5, 2014 at 9:37 PM, Tim Lammens <tim...@gm...>
>> wrote:
>>
>>> Hi,
>>>
>>> We are using your wrapper and sometimes the wrapper itself goes out of
>>> memory.
>>> It seems when the method getLastErrorText() in the C code of the wrapper
>>> is called, the memory allocated with malloc is not being freed afterwards.
>>> Is this possible or did I miss something?
>>>
>>> Regards,
>>> Tim
>>>
>>
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce.
> Perforce version control. Predictably reliable.
>
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>