Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!pepper!cmcmanis
From: cmcmanis%pepper@Sun.COM (Chuck McManis)
Newsgroups: comp.sys.amiga
Subject: Re: Hard drive problems
Message-ID: <46075@sun.uucp>
Date: 18 Mar 88 21:58:58 GMT
References: <2691@crash.cts.com>
Sender: news@sun.uucp
Reply-To: cmcmanis@sun.UUCP (Chuck McManis)
Distribution: na
Organization: Sun Microsystems, Mountain View
Lines: 29

In article <2691@crash.cts.com> billd@crash.cts.com (Bill D'Camp) writes:
> Will the Fast File System (FFS) handle bad blocks, not those found
> during formatting, those that occur as the drive gets used (and abused).

Executive Summary : No.

Long Drawn out explanation :
Well, Steve Beats will probably answer this as well but here goes anyway.
The Fast File System, like the Original File System are just that, File 
Systems. They don't know bad blocks from nothing ok? Now the disk *driver*
or the disk *mechanisim* should know about bad blocks and remap them 
appropriately. If you use a hard disk with a built in SCSI drive there
is usually a facility for telling it that block number "n" is bad and 
that it should re-map it. If the drive is a bare mechanism like the
ST506 type, or SMD type, then either the controller (if it's a smart
one like the SDP) or the driver software need to handle the bad blocks
for you. Drivers like these are not difficult to write, they simply
take some planning. In the case of ST506 drives you need to set aside
some capacity for the eventual bad block, and set aside an area where
you will keep the currently "known" bad blocks. Then when you *first*
initialize the drive, you read in the list of bad blocks from this 
known place and then check every subsequent IO request to see if it
wants to use one of the bad ones, and if so substitute one of the new 
ones. 


--Chuck McManis
uucp: {anywhere}!sun!cmcmanis   BIX: cmcmanis  ARPAnet: cmcmanis@sun.com
These opinions are my own and no one elses, but you knew that didn't you.