[omniORB] LOCATION_FORWARD exception from a python method

[Duncan Grisby]

On Monday 15 December, baileyk at schneider.com wrote:

> Below is a stripped down test program which goes through all the same steps
> as the real system.  I get the same SEGV stack as before on Solaris 8 with
> the latest CVS snapshot of omniORB and omniORBpy.  I'm using Python v2.2.1,
> and the C/C++ compiler is Sun Workshop C/C++ 5.3.

Thanks for the test program. With that I was able to find another bug,
where the object reference in the location forward was accidentally
released before it was used. I've fixed it in CVS, and your test
program works fine for me now.

Cheers,

Duncan.

-- 
 -- Duncan Grisby         --
  -- duncan at grisby.org     --
   -- http://www.grisby.org --