Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 08 Nov 2006 08:39:32 +0100
From:      Fluffles <etc@fluffles.net>
To:        illoai@gmail.com
Cc:        freebsd-questions@freebsd.org
Subject:   Re: Reading crashed SCSI disk
Message-ID:  <455189B4.7040909@fluffles.net>

next in thread | raw e-mail | index | archive | help
>
> I tried
>
> dd if=/dev/da0s1 of=/somefile
>
> dd stopped after a short while with I/O error, after a number of SCSI
> reading errors.
>
> best regards,
>
> Olivier

Hi Olivier,

You might want to try:

dd if=/dev/da0s1 of=/somefile bs=512 options=sync,noerror

The bs=512 means it will transfer sectors (of 512 bytes), this is the
smallest 'units' harddrives are working with. The options=sync,noerror
means that it will continue reading even after bad blocks have been
found. The sync option will cause DD to write zero's in place where it
couldn't read the data. So this command will recover anything that can
be recovered. But, dd will only try once to read the data. Subsequent
tries might succeed while a first try might fail.

If your data is important enough, consider using SpinRite (see
www.grc.com) or some other utility which can actively recover
physical-damaged sectors. I've had good results with SpinRite.

Good luck!

- Veronica




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?455189B4.7040909>