Re: [Wrapper-user] JVM process exited with a code of -1073741571

[Leif M. <le...@ta...>]

Vadim,
    Also, if it is fairly easy for you to reproduce this problem, I 
would appreciate it if you could
try out this new candidate for the 3.1.2 release.   When the JVM 
crashes, it should now display
an error message with the exception name even if debug output is disabled.

http://wrapper.tanukisoftware.org/tmp/3.1.2-e/wrapper_win32_3.1.2-e.zip

Cheers,
Leif

Leif Mortenson wrote:

> Vadim,
>    Doing some research, I figured out that the function call I am 
> making to get the process exit
> code is capable of returning the value of an uncaught exception that 
> killed the process.   I will
> modify the Wrapper for the 3.1.2 release to display more useful 
> information in this case.
>
>    In your case, the exception being thrown was 
> STATUS_STACK_OVERFLOW.   This is
> a low level error in the JVM implementation and not a Java level stack 
> overflow.  Try looking
> through the Sun bug database, or trying a newer JVM version.  It may 
> have been fixed.
> This was the kind of thing that the Wrapper was originally built for. :-)
>
> Cheers,
> Leif
>
> Vadim Punski wrote:
>
>> Hi!
>> I've found a lot of posts started with " JVM process exited with a 
>> code", but no one with such a strange exit code.
>> This could be reproduced by clicking 10-20 times on some button of 
>> the application in the browser.
>> During button pressing the system not overloaded at all.
>> The same thing works perfect during simple jvm execution w/o java 
>> wrapper.
>> All timings and timeouts were extended up to 3-4 minutes, and pinging 
>> works perfect.
>> wrapper.conf configuration added after log messages.
>>
>> Any ideas ?!
>>
>> Best Regards,
>>    Vadim
>>
>> DEBUG  | wrapper  | 2004/09/22 18:22:40 | JVM process exited with a 
>> code of -1073741571, setting the wrapper exit code to -1073741571.
>> ERROR  | wrapper  | 2004/09/22 18:22:40 | JVM exited unexpectedly.
>