[omniORB] orb->destroy SIGSEGV

[jorgefm at cirsa.com]

>> I'm developing under SuSE Linux 8.1 (i386) - Kernel 2.4.19-4GB with
omniORB
>> 4.0.1.
>>
>> When i call the destroy method from orb i get a segmentation fault with
the
>> next backtrace.
>>
>>[...]
>> #0  0x4031e00a in chunk_alloc () from /lib/libc.so.6
>> #1  0x4031de42 in malloc ( ) from /lib/libc.so.6
>> #2  0x4025c94e in operator new(unsigned) () from /usr/lib/libstdc++.so.5
>> #3  0x4025caaf in operator new[ ](unsigned) () from
/usr/lib/libstdc++.so.5
>> #4  0x400adcfd in CORBA::string_alloc(unsigned long) ()
>
>Segfaults in malloc are almost always due to earlier memory
>corruption. Are you able to run your code under valgrind to see if
>that traps it?

Thanks for your clue. It's was a problem related with an extra call to
'_remove_ref()' in a servant.

Thnaks,
Jorge