Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 1 Dec 2002 21:47:00 +0100
From:      Cliff Sarginson <cls@raggedclown.net>
To:        freebsd-questions@freebsd.org
Subject:   Re: Memory test?
Message-ID:  <20021201204700.GB1424@raggedclown.net>
In-Reply-To: <200212011418.GB1EIVI40435@asarian-host.net>
References:  <200211291228.GATCSZI18555@asarian-host.net> <44vg2dq522.fsf@be-well.ilk.org> <200212011418.GB1EIVI40435@asarian-host.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Dec 01, 2002 at 03:18:26PM +0100, Mark wrote:
> ----- Original Message -----
> From: "Lowell Gilbert" <freebsd-questions-local@be-well.no-ip.com>
> To: "Mark" <admin@asarian-host.net>
> Cc: <freebsd-questions@FreeBSD.ORG>
> Sent: Sunday, December 01, 2002 1:47 PM
> Subject: Re: Memory test?
> 
> 
> > Mark <admin@asarian-host.net> writes:
> >
> > > I just did a makeshift memory stress test on a new machine; that is,
> > > I did a "make buildworld". Regrettably, it failed, saying it had a
> syntax
> > > error somewhere. :(
> > >
> > > So, I ran it again; and this time it went fine; then I reinstalled
> > > FreeBSD 4.7 altogether, and rebuilt world again. And, again,
> > > no problems.
> > >
> > > This is a bit worrysome. Does FreeBSD even notice bad memory?
> > > Would the kernel put a message in /var/log/messages? It did not
> > > show anything there when buildworld failed.
> >
> > I'm not sure what you're getting at.  You've seen just one failure,
> > and it is of a type that is highly unlikely to be memory-related.
> 
> 
> Au contraire; a syntax error in a "make buildworld" that cannot be repeated
> when issued again is almost always indicative of a memory error --
> especially with a new out of box FreeBSD 4.7R -- and is generally considered
> a good memory stress test.
> 
> I was rightly alerted to a memory error. I ran Memtest86, suggested by a
> list-member here, and indeed, one of the Kingston 512M DIMM's was acting up.
> I did some research; and Kingston memory appears to have known issues with
> the ASUS A7V333 board.
> 
As an aside, memtest *can* show false positivies on some Asus
motherboards with AMD chips. Tests 5 and 7 are not to be trusted. I
found this out by own observations, and I believe (although I cannot
remember where I read it) that tests 5 and 7 can be a bit iffy. The
memory in question when in use and under stress showed no memory typical
problems at all. The same memory tested on a system with a Pentium III
also showed no failures with repeated running of these particular
tests.

-- 
Regards
   Cliff Sarginson 
   The Netherlands

[ This mail has been checked as virus-free ]

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021201204700.GB1424>