> I tried /sbin/badsect and can't seem to get it to start up, it keeps
> saying the device is busy (it's the /usr partition which is bad).

Umm...this could be that bug in badsect. Are you running 1.2? I
think the PR got into that release. If not, I'll try to dig up the
PR and put the patch on my patches page.

cjs

Curt Sampson    curt@portal.ca		Info at http://www.portal.ca/
Internet Portal Services, Inc.	
Vancouver, BC   (604) 257-9400		De gustibus, aut bene aut nihil.

On Mon, 21 Oct 1996, Olaf Schroeder wrote:

> Date: Mon, 21 Oct 1996 16:22:39 -0600
> From: Olaf Schroeder <olaf@digirule.com>
> To: John D Duncan <jddst19+@pitt.edu>
> Cc: port-i386@NetBSD.ORG
> Subject: Re: /dev/wd
> 
> At 05:00 PM 10/21/96 -0400, John D Duncan wrote:
> >It's not whether wd is smart or not--newfs supports the bad144 standard
> >of self-mapping bad sectors at format. The best way to handle this is
> >to format the drive. As that seems a out of the question:) you should
> >probably try to find the bad sectors and use /sbin/badsect to mark them.
> >That is not as satisfactory, though, because bad sectors are stored
> >in a file. But it should work. I'm not sure at the moment how to find
> >exactly which sectors are bad.
> >
> I was thinking about this afterwards and thought it might be the
> file system's job of marking bad sectors, not the device driver.
> I've killed everything in sight but still can't get it to budge.
> I guess I'm going to have to reformat... but when it comes down to
> it seems newfs doesn't mark bad sectors on-the-fly, only at format
> time. Sounds like a bad assumption to me...
> 
>