[Top] [All Lists]

SMART (was Re: LBA to File?)

To: Chris Wedgwood <cw@xxxxxxxx>
Subject: SMART (was Re: LBA to File?)
From: Bogdan Costescu <bogdan.costescu@xxxxxxxxxxxxxxxxxxxxx>
Date: Mon, 10 Mar 2003 14:25:29 +0100 (CET)
Cc: Bogdan Costescu <bogdan.costescu@xxxxxxxxxxxxxxxxxxxxx>, Michel Machado <michel@xxxxxxxxxxxxxxx>, <linux-xfs@xxxxxxxxxxx>
In-reply-to: <20030310130657.GA25467@f00f.org>
Sender: linux-xfs-bounce@xxxxxxxxxxx
On Mon, 10 Mar 2003, Chris Wedgwood wrote:

> There a vendor-specific and vendor-neutral things you can poll.

But that's the best you can get while online... Offline, I think that each 
manufacturer provides some diagnostic tool that can go and test each 
sector (and destroy any data that was there before :-)) and report if 
there are problems. 'badblocks' is also useful while offline, although it 
only reports the present situation and doesn't know about transparent 
remapping done by the drive firmware.

> I'm not 100% sure about this, but I don't think controllers and/or
> mainboards need to know anything about SMART at all...

They need to know how to read and interpret a bunch of parameters which 
give the health status of the drive. If some warning value is triggered, 
you get when you boot some message like:

Warning: Disk failure. Please backup and replace disk 1.
Press any key to continue.

(this is all from memory, so I might get it wrong). When BIOS does the 
checking it's also supposed to leave the drive with SMART enabled, so more 
sofisticated tools can read parameters and logs after the OS is loaded.
I did however used safely (= no data loss) such drives for months, so it 
might be that because of some combination of BIOS and drive firmware the 
early warning was too early...

Bogdan Costescu

IWR - Interdisziplinaeres Zentrum fuer Wissenschaftliches Rechnen
Universitaet Heidelberg, INF 368, D-69120 Heidelberg, GERMANY
Telephone: +49 6221 54 8869, Telefax: +49 6221 54 8868
E-mail: Bogdan.Costescu@xxxxxxxxxxxxxxxxxxxxx

<Prev in Thread] Current Thread [Next in Thread>