[omniORB] OmniNames crashing

Bruce Visscher visschb@rjrt.com
Tue, 06 Jun 2000 10:14:04 -0400


Okay, so now we have three different platforms exhibiting this bug.

Since it apparently started happening somewhat recently, perhaps it has
something to do with the introduction of the omniORB_Ripper thread?

Is there any hope that this problem will go away under omniORB 3?

I recall uncovering a problem in this area before, but the design has
changed somewhat since then.

James Riden wrote:
> 
> "Shelton, Susan L." <Susan.Shelton@viatel.com> writes:
> > I have been successfully running omniNames on an HP unix machine for about 5
> > months.  I never had any problems until just recently, it keeps crashing.
> > It core dumps with a
> >
> > 'SIGSEGV for stack growth failure.  Insufficient memory or swap space, or
> > stack size exceeded maxsize.'

I have noticed that omniNames is moderately recursive in places, so it
is theoretically possible to have a stack overflow if your naming graph
is too deep.  You could try increasing the stack size using the
omni_thread::stacksize member.  OTOH, since you are getting an assertion
failure, this is probably just the application not dying gracefully. 
omniORB 3 seems to be better at this from what I've seen.

> > error.  I then looked at the logging for omniNames and am seeing an
> > assertion error:
> >
> > Assertion failed: pd_refcount >= 0, file ../strand.cc, line 166
> 
> Bruce Visscher mentioned this recently in :
> Subject: [omniORB] omniNames crashes on OpenVMS Alpha with multiple
> kernel threads enabled
> Date: Mon, 05 Jun 2000 17:17:46 -0400
> Message-ID: <393C18FA.2229AC6A@rjrt.com>
> 
> I've seen this on NT 4.0 / x86 but very rarely and not reproducibly -
> just the "Assertion failed: pd_refcount >= 0, file ../strand.cc". I
> can't remember if the line number was the same.
> 
> Sorry, this is the lamest bug report I've ever written, but as I say
> it doesn't happen often.

No need to appologize, IMHO.  It does seem to be an illusive bug.

> cheers,
>   Jamie
> --
> James Riden / jamesr@harlequin.co.uk
> 
> _____________________________________________________________________
> This message has been checked for all known viruses by Star Internet delivered
> through the MessageLabs Virus Control Centre. For further information visit
> http://www.star.net.uk/stats.asp

-- 
All generalities are false - including this one.

Bruce Visscher                                        visschb@rjrt.com