Should bad hardware cause a kernel panic?

[Phil Kernick]

This is a serious question and not intended as a troll.

I have a dying IDE hard drive that I want to try and recover some file from - 
the problem is that with it connected to my box, both FreeBSD 4.9-RELEASE and 
5.2-RC2 kernel panic.

This can be reliably duplicated by booting the machine with the drive not 
connected, then plugging the drive in, and...

# atacontrol reinit 1
panic!

The real problem with this is that I can't use any tools to try and recover 
any data.  The GEOM code in 5.x is slightly more resiliant than the 4.x code, 
but both cause panics.

On the other hand, it I boot a Knoppix Linux distro, it complains about the 
drive, but it *doesn't* panic, crash or otherwise misbehave.

Is this caused by us not properly checking return codes that come back from 
the ata driver, or something more insidious?

I would have thought the the expected behaviour was returning an error and not 
a panic, especially if the problem is detected during initialisation of the bus.


Phil.

-- 
    _-_|\   Phil Kernick                      E-Mail: Phil at Kernick.org
   /     \  ROTFL Enterprises                 Mobile:  041 61 ROTFL
   \_.-*_/
        v   Humourist, satirist, and probably a few more 'ists to boot!