[omniORB] Problems running omninames on solaris2.6

Duncan Grisby dgrisby@uk.research.att.com
Thu, 15 Jun 2000 11:15:55 +0100


On Tuesday 13 June, Janet Tvedt wrote:

> I never saw a response to this message.  I am seeing the same
> message when several processes attempt to access the omniNames
> at the same time.  Does anyone have an idea as to what would
> cause this?

> > libthread panic: cannot create new lwp ( PID:nnnnnnn LWP: 2)
> > stacktrace
> >      ef4ced20
> >      0

Each new incoming connection to an omniORB server starts a new thread.
The error shows that the Solaris libthread dies horribly when it fails
to create a new thread, rather than returning an error to omniORB. I
don't think this can be an omniORB problem -- I think there must be
something dodgy about your Solaris installation. Maybe it's configured
to severely limit the number of threads a process can have?  Even
then, you'd expect it to complain gracefully rather than dying.

Sorry I can't be more help,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --