From owner-freebsd-stable  Wed Feb  7  8:11:31 2001
Delivered-To: freebsd-stable@freebsd.org
Received: from clmboh1-smtp3.columbus.rr.com (unknown [65.24.0.112])
	by hub.freebsd.org (Postfix) with ESMTP id 756E837B4EC
	for <stable@freebsd.org>; Wed,  7 Feb 2001 08:11:13 -0800 (PST)
Received: from mail.iowna.com (dhcp065-024-023-038.columbus.rr.com [65.24.23.38])
	by clmboh1-smtp3.columbus.rr.com (8.11.2/8.11.2) with ESMTP id f17G8gr00133;
	Wed, 7 Feb 2001 11:08:47 -0500 (EST)
Message-ID: <3A817294.A0A1DE5C@mail.iowna.com>
Date: Wed, 07 Feb 2001 11:06:44 -0500
From: Bill Moran <wmoran@mail.iowna.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; FreeBSD 4.2-STABLE i386)
X-Accept-Language: en
MIME-Version: 1.0
To: stable@freebsd.org, jwz@jwz.org
Subject: webcollage(ppmmerge)/xscreensaver crashes 4.2-STABLE
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-stable@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

FreeBSD 4.2-STABLE (Jan 20)
xscreensaver-gtk-3.26 installed as FreeBSD package.

I'm not really so worried about seeing this fixed as I thought I should
point it out to others. I have at least one production server that runs
xscreensaver (to keep the local admin happy) and I wanted to let others
know of the problem I found.
When the webcollage screen hack runs it fairly regularly causes a full
reboot - I haven't actually seen it happen yet, but here are the
messages I seen in /var/log/messages:

Feb  6 15:07:50 working /kernel: pid 2405 (ppmmerge), uid 1000: exited
on signal 11 (core dumped)
Feb  6 16:26:35 working /kernel: pid 3497 (ppmmerge), uid 1000: exited
on signal 11 (core dumped)
Feb  6 19:56:58 working /kernel: pid 4071 (ppmmerge), uid 1000: exited
on signal 11 (core dumped)
Feb  6 19:57:35 working /kernel: pid 4515 (ppmmerge), uid 1000: exited
on signal 11 (core dumped)
Feb  7 08:09:24 working /kernel: Copyright (c) 1992-2000 The FreeBSD
Project.
... and the rest of the boot messages

While the times seem to suggest that something was occurring in between
(several hours in there ...) I can't find any other indicator of what
could be causing it, and it doesn't happen when I disable the webcollage
screen hack. Other occurances show varying times between the crash of
ppmmerge and the reboot of the system (as short as 2 hours) but it's
always the same thing ... ppmmerge coredumps and a few hours later the
system reboots.
I haven't caught it happening yet to watch what happens, nor have I been
able to reproduce it by manually running webcollage in a root window. I
suspect some sort of memory leak that eventually brings the system down,
but I don't have details yet.

So this is a) an early warning to others not to run the webcollage
screen hack on critical systems, b) an invitation to others to research
this further if they'd like.
I'll probably take some more time over the next few weeks to try to
isolate this better.
I know this isn't much data, but I just thought I'd pass it on.

-Bill


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message