[omniORB] for/execvp from within an ORB client

[Stefan Seefeld]

hi,

I'm writing a test application for the berlin project
(massively using CORBA). I try to fork to run a coprocess
and all seems fine until the server tries to call the client
back, at which point the whole deadlocks.

Is there anything known about conflics between omniORB and
running fork/execvp in a client ?
Fork alone works fine, the problem only happens when issuing
the exec...
Is it possible that some file descriptors are damaged when
doing this ? How can I debug this further ? (sadly, even a
newer version of gdb core dumps with my demo - ah, the ongoing
gdb/thread blues)

Any help would be highly appreciated !

Stefan

_______________________________________________________              
              
Stefan Seefeld
Departement de Physique
Universite de Montreal
email: seefelds@magellan.umontreal.ca

_______________________________________________________

      ...ich hab' noch einen Koffer in Berlin...