Re: death by ATA

Andrew Morton (akpm@zip.com.au)
Fri, 16 Nov 2001 17:21:25 -0800


Jens Axboe wrote:
>
> On Thu, Nov 15 2001, Andrew Morton wrote:
> > What does "end-request: buffer-list destroyed" mean?
>
> It means that the request was not sane anymore, or specifically that
> clustered number of sectors was set to lower value than current number
> of sectors (which isn't valid, of course). The buffer-list destroyed
> comment tells you that this corruption is most likely due to the
> buffer_head list on the request having been corrupted -- which in turn
> probably means that someone seriously screwed this request.
>
> hda8: bad access: block=5296, count=-2
> end_request: I/O error, dev 03:08 (hda), sector 5296
> hda8: bad access: block=5298, count=-4
> end_request: I/O error, dev 03:08 (hda), sector 5298
> hda8: bad access: block=5300, count=-6
> end_request: I/O error, dev 03:08 (hda), sector 5300
>
> This errors would seem to backup that theory :-)

'k, thanks.

> Is this an SMP board? Also, is

Uniprocessor VIA C3, running 2.4.15-pre4. The controller is a
VT8231. Running at UDMA100.

> end_request: buffer-list destroyed
>
> the very first error message?

Yes, it is.

It is reproducible after around three few hours. Exactly the
same.

-
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/