Path: utzoo!attcan!uunet!ncrlnk!ncrcae!hubcap!gatech!udel!rochester!pt.cs.cmu.edu!wb1.cs.cmu.edu!avie
From: avie@wb1.cs.cmu.edu (Avadis Tevanian)
Newsgroups: comp.sys.next
Subject: Re: optical disk
Message-ID: <4059@pt.cs.cmu.edu>
Date: 18 Jan 89 23:40:28 GMT
References: <4051@pt.cs.cmu.edu>
Organization: Carnegie-Mellon University, CS/RI
Lines: 37

In article <4051@pt.cs.cmu.edu> eht@f.word.cs.cmu.edu (Eric Thayer) writes:
>This one really is causing problems!  How do work around an
>optical disk with a non-correctable (see Release Notes) bad
>block?

Write down the PHYSICAL disk block address causing problems, then run the
disk program (e.g., /etc/disk /dev/rod0a).  Select the "abort" command, which
disable the abort-on-error feature.  Now, determine how many sectors have bad
ECC.  It is almost always 8 (for 8K) but is sometimes 1.  To do this, execute
the "read" command.  It will prompt for several values:

starting block:		supply the PHYSICAL block you recorded earlier.
# sectors per transfer:	answer 1.
number of transfers:	answer 8.
sector increment:	answer 1.

Now watch the error messages about bad ECC, you'll get either 1 or 8 of them,
corresponding to how many sectors are bad.  If you get no error messages,
you've probably provided the wrong starting block.  Start over!

To "fix" these sectors, issue the "write" command, answering the questions
as above (but for the number of transfers specify 1 or 8 depending on
how many sectors were bad).  Also, it will prompt you for "random data," say
no, which will write zeros.

What this does is to write the bad sectors, forcing good ECC.  Unfortunately,
you previous data there is lost.
--
Avadis Tevanian, Jr.    (Avie)
Chief Operating System Scientist
NeXT, Inc.
avie@cs.cmu.edu or avie@NeXT.com
-- 
Avadis Tevanian, Jr.    (Avie)
Chief Operating System Scientist
NeXT, Inc.
avie@cs.cmu.edu or avie@NeXT.com
--