Re: [gentoo-user] Re: Harddisk trouble ... or not yet?

["W.Kenworthy" <billk@iinet.net.au>]

On Thu, 2010-05-27 at 04:43 +0200, meino.cramer@gmx.de wrote:
> walt <w41ter@gmail.com> [10-05-27 04:08]:
> > On 05/26/2010 05:19 PM, meino.cramer@gmx.de wrote:
> > 
> > >Is there any way to check the disk without damaging its
> > >contents?
> > 
> > Do you know about SMART?  Install sys-apps/smartmontools if you
> > don't already have it, and read the manpage for smartctl.
> > 
> > I add smartd to my default runlevel so the hard drives will test
> > themselves once every month and log the test results in syslog.
> > 
> > Hm.  I just noticed that smartd isn't actually running, so I need
> > to do some debugging now.  But that's the idea, anyway.
> > 
> 
> Hi,

> yes, I know smart...
> 
> I was more thinking of a tool, which test the whole disc surface
> and reports every bad sector.
> 
> Smarts is more of statistical kind: It counts events and tries
> to calculate dooms day from that ;)
> 
> 

Not so easy (coming in late on this thread, sorry if its been covered)

Modern hard drives insulate the outside from whats actually happening
internally.  They have a number of spare locations they can swap into
use when a bad patch develops.  This is invisible except to something
like smart reporting.  Rule of thumb - when a modern drive starts
showing bad sectors to the outside world, its already well past its use
by date.

So something like SMART is the only way for the average Joe to get the
health of a drive.

Google has lots on this sort of thing

BillK