From owner-freebsd-questions@FreeBSD.ORG Fri May 28 10:40:17 2004 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4820E16A4CE for ; Fri, 28 May 2004 10:40:17 -0700 (PDT) Received: from smtp.owt.com (smtp.owt.com [204.118.6.19]) by mx1.FreeBSD.org (Postfix) with ESMTP id CE2D543D31 for ; Fri, 28 May 2004 10:40:16 -0700 (PDT) (envelope-from kstewart@owt.com) Received: from [207.41.94.233] (owt-207-41-94-233.owt.com [207.41.94.233]) by smtp.owt.com (8.12.8/8.12.8) with ESMTP id i4SHdEbI029099; Fri, 28 May 2004 10:39:15 -0700 From: Kent Stewart To: freebsd-questions@freebsd.org Date: Fri, 28 May 2004 10:39:48 -0700 User-Agent: KMail/1.6.2 References: <40B77019.9020108@lineone.net> In-Reply-To: <40B77019.9020108@lineone.net> MIME-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <200405281039.48083.kstewart@owt.com> cc: Robert Downes Subject: Re: buildworld actually crashed X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 May 2004 17:40:17 -0000 On Friday 28 May 2004 10:00 am, Robert Downes wrote: > During buildworld, I wandered off. When I returned, my machine was, > alarmingly, in single user mode, demanding that I run fsck manually. > > I'm running fsck right now, and it's finding all sorts of block size > errors, to which I'm simply hitting 'y' and agreeing that things > should be salvaged and corrected. > > Before running fsck, I had a look at the buildworld.out script that > was being written to during the buildworld process. I can't tell you > exactly what it says, but it definitely came to a stop in the middle > of a 'sentence' of output. I.e. it looks like my new machine (yeah, > the soon-to-be-fanless EPIA again) must have crashed during > buildworld. > > What could cause buildworld to crash like that? I'm now worried that > my PSU board *was* damaged the other day. Is a damaged PSU the most > likely cause of this incident? Just one of the possibilities. There are connectors that will test ATX psu's. My experience with psu failures is the system dies and you don't have something like HDs. I had what I think is a cpu failure that caused a voltage regulator IC to explode on the mobo. I was looking at the mobo from about 2 feet when it did this. That was about the most exciting thing that happended that day :). I was telling a friend about it and he made the comment that they don't throw shrapnel but disappear into dust. I never saw or felt anything and assume that he was right. Insufficient size will cause intermitant errors depending on the cpu. I usually go at least the next 50 watts up from what is recommended for a cpu. I also have a tendancy to like 350+'s with dual fans. > > All advice very welcome. In the FAQs trouble shooting link is a section on signal 11's. I would look at the options. Very little will cause a buildwold to die but hardware errors. You have to use wavy hands at this point because -stable and -current occasionaly have errors but none of the software errors are likely to cause the system to reset it self during a build. Kent -- Kent Stewart Richland, WA http://users.owt.com/kstewart/index.html