Dead disk? READ_DMA Failure
hornetmadness at gmail.com
Fri Jul 29 02:30:54 GMT 2005
On 7/28/05, Erik NÃ¸rgaard <norgaard at math.ku.dk> wrote:
> Please read on although this is a long one... I might panic and
> coredump myself if I don't get this fixed.
> I have a one year old Hitachi Travelstar 60GB/4200RPM disk. This
> night and since then, I get these errors:
> ad0: TIMEOUT - READ_DMA retrying (2 retries left) LBA=48926527
> ad0: TIMEOUT - READ_DMA retrying (1 retry left) LBA=48926527
> ad0: FAILURE - READ_DMA timed out
> ad0: TIMEOUT - READ_DMA retrying (2 retries left) LBA=50299455
> ad0: FAILURE - READ_DMA status=51<READY,DSC,ERROR>
> error=40<UNCORRECTABLE> LBA=50299455
> The LBA number varies, but otherwise it is consistent. Eventually
> the device is removed from /dev and no disk is accessible,
> although mount shows mounted devices.
> I have booted in single user mode and run fsck manually multiple
> times to correct all found errors, most errors are on /, /usr and
> some on /var, yet the problem reappears.
> I have tried to reinstall, toggling newfs to recreate /, /usr and
> /var hoping that any bad blocks or sectors would be removed or
> marked unusable. But the problem remains.
> Is there a way I can reformat or fix the disk so that broken areas
> will not be used again?
> Please, how far from the grave are my data?
> Any suggestion on which new drive to choose? I have looked at
> Western Digital Scorpio WD600VE 60 GB.
> Given the mentioned error, does this indicate an error that would
> be covered by standard guarantee? after all the disk is just one
> year old...
> The system: FreeBSD 5.4, Epia Mini-ITX CL1000, 256MB RAM, 60 GB
> Thanks, Erik
> GnuPG: http://www.locolomo.org/home/norgaard/norgaard.gpg.asc
> pub 1024D/11D11F9E 2003-08-15 Erik Norgaard <norgaard at locolomo.org>
> Key fingerprint = C394 81C4 D137 EEE5 39BE 82D5 3E6B FB3E 11D1 1F9E
> freebsd-questions at freebsd.org mailing list
> To unsubscribe, send any mail to "freebsd-questions-unsubscribe at freebsd.org"
Its been in my experience that once the disk starts getting bad
sectors, it will just get worse with time. If DOS could format around
bad sectors, I'm sure BSD can do it. I would suggest that you backup
your data, and RMA the drive. That is the safest way to go about this.
If you cant/wont RMA the drive then make that 60G drive hold non-vital
information, and use another drive for your system slices.
More information about the freebsd-questions