Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 28 Jul 1996 20:09:59 +0000 ()
From:      James Raynard <fqueries@jraynard.demon.co.uk>
To:        "Keith Beattie[SFSU Student]" <beattie@george.lbl.gov>
Cc:        questions@freebsd.org
Subject:   Re: vnode_pager_input: I/O read error
Message-ID:  <199607282009.UAA02702@jraynard.demon.co.uk>
In-Reply-To: <199607271929.MAA09047@george.lbl.gov> from "Keith Beattie[SFSU Student]" at Jul 27, 96 12:29:37 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> 
> Things are getting worse.  When I boot, fsck fails and running it
> manually it succeds on my first ide drive (the one with / on it) but
> then while checking my SCSI drive it dies from a sig 8:
> 
> --- manual fsck run ---
> *** /dev/rsd0s1e
> BAD SUPERBLOCK: VALUES IN SUPERBLOCK DISAGREE WITH THOSE IN FIRST ALTERNATE
> pid 13: fsck: uid 0: exited on signal 8
> floating exception
> --- manual fsck run ---
> 
> I tried it several times and it always dies the same way.
 
Hmm. Signal 8 is SIGFPE (floating point exception).  This could
just be as simple as fsck trying to divide something by zero.
 
> I don't have any spare ones.  I'd just have to remove one bank at a
> time.  But are sig 8's indicative of bad SIMMs?

Not that I know of.  I would have thought this indicated some
kind of motherboard problem, at a guess - or maybe the filesystem
on /dev/rsd0s1e has been trashed beyond fsck's ability to repair it.

> I *need* this to be a software problem, I can spare the time to
> reformat/reintsall/whatever, new hardware is a different story...

I would certainly try installing from scratch again.  

Let's try and be more positive about the hardware possibilities - 
maybe your cache settings are a bit too aggressive, in which case 
it's worth tweaking about with the BIOS settings (after making a 
note of them, of course).

Or you may have an overheating CPU - try taking the case off the 
machine, or switching on the CPU fan if you have one.  Also, it 
has been known for unscrupulous vendors to try and pass off
CPUs as being higher-rated than they really are (eg selling P90's
as P100's).  FreeBSD prints out the true CPU speed when it boots.




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199607282009.UAA02702>