SUMMARY - head disabling with format

From: Ken Britten (ken@monkeybiz.Stanford.Edu)
Date: Sun Jun 07 1992 - 20:16:22 CDT


************************* THE PROBLEM *************************

> We have a Fuji 2333 (driven by a Xylogics 450 controller
> running on a 3/160) which appears to have a head going bad
> - all the errors occur on blocks under head 7. I'd like to
> avoid manually going through and adding all the blocks by
> hand, but if there were an easy way to disable the head....

By the way, this problem is not a crash, since 'format' finds
no surface defects. Hence the desire for a solid patch, since
the drive might live for a while yet.

************************** SOLUTION 1 **************************

If you do have to take the brute-force approach and map out individual
blocks, you may be able to automate that with a script to re-direct
into the format command. You'd need to take special care to get the
right commands at the beginning and end, but there would be a lot of
repetive stuff in the middle. (Just a thought from someone lazy. 8-)

Brent Chivers McLean, VA bchivers@mitre.org

************************** SOLUTION 2 ****************************

>From Perry_Hutchison.Portland@xerox.com Thu Jun 4 12:53:46 1992
Subject: Re: 'format' question
To: ken@monkeybiz.stanford.edu
Reply-To: Perry_Hutchison.Portland@xerox.com

You could define a new format.dat entry which claims that the drive has
only 7 heads (0-6), and reformat. That would disable head 7 (and any
higher-numbered heads as well).

********************** END OF SUMMARY ***************************

Action taken - right now, none, since the problem was alleviated
by taking out the worst offending blocks. But when it gets bad
again (not if), we will use solution 1, because that way we can
save the space under heads 8 and 9, which would be lost under
solution 2.

*****************************************************************

Thanks for the help, guys!

Ken Britten
(ken@monkeyiz.stanford.edu)



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:43 CDT