Hazards of power management (HDD power down)

Hazards of power management (HDD power down)

Post by Benno J. Overeind » Thu, 05 Feb 1998 04:00:00



Hi,

I am curious about other people's experiences with the HDD power down
(power management) facility.

I have an Award BIOS (don't know the version number by heard, but it is
from 1997), and the problems I have is with a Maxtor 1.6GB disk.

When I started using my machine (I'm running Linux) with HDD power down
on 15 minutes, it occasionly caused a corrupt file system after spin up
of an idle disk.  Once, had to remake the file system on the disk, as
the file system check program (e2fsck) could not recover/remap the
inodes.

Well, after these experiences I disabled the power management
facilities, and everything worked well.  At least up to yesterday.  I
thought, I might try to use the power down again as I have not found any
negative references (searched with Deja News).  I rebooted a correctly
running machine, entered the BIOS, set the HDD power down on 15 minutes,
saved the BIOS settings and exited.

During the reboot, the primary master disk was not bootable anymore.
The Linux kernel could not read the partition table.  With DOS, the
first drive was not readable.  A low level format showed that I had bad
blocks in cylinder 0!

The Linux kernel reported this:

Partition check:
 hda: hda: dma_intr: status=0x51 { DriveReady SeekComplete Error}
had: dma_intr: error=0x80 { BadSector }, LBAsect=1, sector=0
end_request: I/O error, dev 03:00, sector 0
 unable to read partition table

A low level format reported the following defect table:

head cyl
 0    0
 1    0
16    0
17    0
32    0
33    0
48    0
49    0

I assume that bad blocks in the first cylinder cannot be remapped.  In
any case, with DOS fdisk /mbr is not able to write the master block to
disk, and with Linux fdisk reports these dma_intr's.

The disk is dead, and I have to replace him.  But still... was it a
coincident that my disk died after I set the HDD power down, or can it
be caused by this?  Are there any other people having the same
experience?

-- Benno
--
 Benno Overeinder                        Department of Computer Science
 voice: (+31) 20 525 7536                University of Amsterdam            
 fax:   (+31) 20 525 7490                Kruislaan 403, 1098 SJ  Amsterdam

 
 
 

Hazards of power management (HDD power down)

Post by Jorg Heijni » Thu, 05 Feb 1998 04:00:00


Quote:

>A low level format reported the following defect table:

>head cyl
> 0    0
> ...>The disk is dead, and I have to replace him.  But still... was it a
>coincident that my disk died after I set the HDD power down, or can it
>be caused by this?  Are there any other people having the same
>experience?

Low level formating an IDE drive without the propper software from the
hardware
manufacturer can cause malfunction in some cases. Never lowlevel format a
IDE drive unless you have the proper software or know for sure you HD
does'nt suffer.