Bad sectors on harddisk (question)

[Lennart Sorensen]

On Thu, Oct 16, 2003 at 06:59:27PM -0400, William Park wrote:
> What is conventional wisdom when you encounter bad sectors on harddisk?
> 
> I am only aware of 
>     umount /dev/hdb1
>     e2fsck -c /dev/hdb1		(read-only check)
> or
>     umount /dev/hdb1
>     badblocks -n -b 4096 -o xxx /dev/hdb1	(non-destructive read-write)
>     e2fsck -c -l xxx /dev/hdb1
> 
> Is there better way of 
>     - identifying bad sectors (/var/log/syslog says so).
>     - then mapping them out, so that filesystem does not use them.

Unless the HD is small by today's standards and very old (over a
decade), it will not show bad sectors.  The hardware remaps them
automatically, unless it has run out of spares, in which case it is time
to get the data off the drive ASAP and discontinue use of the drive.

smart utilities to ask the drive for it's status is handy on modern
drives.  smart is a standard used to predict failures on modern
harddisks.

Lennart Sorensen
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml