4da
Subj : Re: bad sectors
To   : comp.os.linux
From : erik
Date : Fri Oct 22 2004 04:26 pm

Stuart Fuller wrote:

> erik wrote:
> 
>  
>> Correct. Use 'smart' technology to see what is really happening.
>> Drives automatically remap bad sectors. If you start to see bad
>> sectors, the drive already ran out of its spares... Disaster is
>> lurking...
> 
> Not necessarily.  Consider the case where the drive is commanded to
> read a block, but there are errors in that block that prevent the
> drive from
> obtaining a successful read.  Yes, it is quite appropriate to revector
> the bad block to a spare, but what of the current outstanding read
> that the
> user (operating system)is patiently waiting for?  In this case, the
> drive has no choice but to return a read error to the user (operating
> system). The drive has not necessarily ran out of spare blocks, but it
> needs to return a read error.
> 

That block is lost anyway, and an error should be returned at all times.
When a drive starts remapping, it is _time_ to replace it. There should
be no doubt about that.

EJ
-- 
Remove the obvious part (including the dot) for my email address.
http://www.vanwesten.net for examples of ipf and pf.

.

0