On Mon, Feb 19, 2001 at 03:36:06PM -0800, Wolfgang Rupprecht wrote:
> 
> Manuel Bouyer writes:
> > On Sun, Feb 18, 2001 at 07:50:51PM -0800, Chuck Silvers wrote:
> > > this kinda looks like the IDE driver is losing the i/o...
> > > even if the hardware isn't responding, I would still think the
> > > driver should time out the request eventually.
> > Wolfgang, do you have another device on the same IDE channel as the
> > dump device ?
> 
> Let me try to shed a bit more light here.  If there is any more info
> that would be useful please speak up.
> 
> Let me put this on a web page, since I don't think anyone enjoys a
> 1k-line email message.
> 
> tests I've done so far:
> 
>   http://www.wsrcc.com/wolfgang/hang/pasillo-diskhang
> 
> boot-time dmesg:
> 
>   http://www.wsrcc.com/wolfgang/hang/pasillo-dmesg
> 
> kernel config files (nested)
> 
>   http://www.wsrcc.com/wolfgang/hang/WSRCC_INTEL
>   http://www.wsrcc.com/wolfgang/hang/WSRCC
> 
> the "hang" directory is listable.  I'll put any newly requested crap
> in there.

Ok. So it looks like you can still access the disk.
Could you try a 'dd if=/dev/rwd0d of=/dev/null bs=64k' while it's hung ?
Let it run until it reaches the end of disk if possible.

But as the disk is still alive I don't think the problem can be in the
IDE driver; although I'll look again tonigth.

--
Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
--