From owner-freebsd-questions  Wed Sep  3 16:29:20 1997
Return-Path: <owner-freebsd-questions>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id QAA15032
          for questions-outgoing; Wed, 3 Sep 1997 16:29:20 -0700 (PDT)
Received: from zeus.gel.usherb.ca (zeus.gel.usherb.ca [132.210.70.7])
          by hub.freebsd.org (8.8.7/8.8.7) with SMTP id QAA15027
          for <freebsd-questions@FreeBSD.ORG>; Wed, 3 Sep 1997 16:29:17 -0700 (PDT)
Received: from castor.gel.usherb.ca by zeus.gel.usherb.ca (4.1/SMI-4.1)
	id AA18227; Wed, 3 Sep 97 19:29:15 EDT
Received: by castor.gel.usherb.ca (SMI-8.6/SMI-SVR4)
	id TAA21212; Wed, 3 Sep 1997 19:29:10 -0400
Date: Wed, 3 Sep 1997 19:29:10 -0400 (EDT)
From: "Alex.Boisvert" <boia01@castor.GEL.USherb.CA>
To: "John P. Reddy" <jreddy@lightning.net>
Cc: freebsd-questions@FreeBSD.ORG
Subject: Re: Fatal double fault
In-Reply-To: <3.0.3.32.19970903105606.007fd4d0@blizzard.lightning.net>
Message-Id: <Pine.SOL.3.91.970903192339.20997A-100000@castor>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-questions@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> 
> I have a system that keeps giving "Fatal double faults" after having an
> uptime of a few hours.  The system freezes , and I get the message:
> 
> Fatal double fault:
> eip = 0x171c6
> esp = 0xefbfcc48
> ebp = 0xefbfcc74
> panic:double fault
> 
> syncing disks . . .
> 
> System stats are:
> Cyrix 6x86L PR200+
> 128 MB Ram
> 1 IDE hard drive (root)
> 3 Seagate Cheetahs striped together using ccd
> 
> Running:
> Diablo (The news feed management program)
> 
> My only guess is that having a Cyrix chip in the same box as the 3 Cheetahs
> is overheating the chip, but I can't tell for certain.  The case is a full
> tower with extra fans blowing through it.  Help me, please! :)

I had the same kind of problem as you are describing with a Cyrix P200+ CPU.
I had to take it down from 150 MHz (2x75MHz) to 133 MHz (2x66MHz).

My gut feeling was that there was a problem with memory access at 2x75
MHz... but I didn't bother to research this problem further.  Now, the
system runs like a charm (but a bit slower).  It's been up for about 20
days (including running a few "make world" to test stability...).

Regards,
Alex.