Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 24 Feb 2007 18:12:39 +1100 (EST)
From:      Ian Smith <>
To:        Jerry McAllister <>
Cc:        Marty Landman <>,
Subject:   Re: input/output error on hd
Message-ID:  <>
In-Reply-To: <>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help
On Fri, 23 Feb 2007, Jerry McAllister wrote:

Jerry, just on a couple of points:

Me to Marty:
 > > Does the above result mean that my boot sector is ok?
 > I think maybe you might be getting some of your terminology shuffled 
 > around.    Your boot sector is not at question here.   It doesn't
 > affect the ability to read the disk partition.

I was talking about sector 0 on ad1.  If the disk were bootable it would
be the 'boot sector', but it's not bootable.  It still contains the DOS
partition table though.  boot0cfg -v <drive> shows this info too.

 > Secondly, it looks like the slice table - which is what fdisk mucks
 > with - is OK.  

Luckily .. though with only one slice on the disk, easily recreatable.

 > What seems to be messed up is either the slice 1 label which contains
 > the partition table.   In fact, with one of your postings, it looked
 > like it thinks it is slice 4, but I don't remember what you did to
 > get that.   

Try 'fdisk ad0sY' where Y is any freebsd slice, and you'll get that (or
at least, I do on both my ad0 freebsd slices).  Maybe fdisk should barf
or at least whinge rather than reporting the first sector of any slice
specified as if it were a sector 0 / MBR / boot sector?

 > If none of this helps at all, then the problem is past the label and
 > in to the filesystem structure itself.   Then I think you are stuck with 
 > tracing superblocks as I pointed out before.   That is going to take some 
 > study but it might work.

Yep, it depends on how precious this data is, but no learning is wasted.

Cheers, Ian

Want to link to this message? Use this URL: <>