On Jan 30, Charles M. Hannum wrote
> Excuse me?  Since when is this is a `known problem'?
> 
> The only time this has ever been reported with IDE disks, AFAIK, is
> when a disk spins down (usually due to APM, but some desktops will
> spin down the disk as well) in the middle of a transfer.  In this
> case, the transfer is retried and unless you get a `hard error
> {read,writ}ing fsbn ...'  message, you can assume that nothing was
> lost.

I've seen these kind of problem on old, slow i386. Chris G. Demetriou
has also these messages on his laptop with APM debug prntf's turned on (you
may want to talk with him directly about that). This let me belive that
there is a timing problem somewhere in the IDE driver, which may show up
on quite slow hardware with fast disks. But I was not able to track this down
as all my machines don't have this problem.

I agree, I should have given more details about the 'known problem'.

--
Manuel Bouyer, LIP6, Universite Paris VI.
email: bouyer@rp.lip6.fr
       bouyer@ensta.fr
--