"dr. Imre Bartfai" wrote:
>
> Thank you for your ingenious advice (really). As the matter of fact,
> before this event the number of bad tracks has very slightly increased
> (from zero to some). After the crash, I tried to read the disk phisically,
> but it was refused on a given place due to physical I/O error. After the
> treatment, the badtracks disappeared.

Good firmware can format around a physically bad part of the surface
by changing the spacing of the inter sector gap so that it lands on
top of the bad media. There was a utility called HDTEST
for old MFM drives that did this very well. Modern drives reserve
many tracks for 'bad block substitution' which gives you the
illusion of a perfect drive (none are with the new ultra high
bit densities) until you try streaming video to them. Then you
find out that the heads are chasing all over the surface doing
block substitution and you get underrun errors. Some A/V rated drives
reseve blocks per track so that the substitution has minimal
effect since no seeks are involved, but the penalty is less total
available storage for the same media area.

> The possibility you mention is fascinated me. Have you an idea, whether
> this accident could be initiated by software event? (e2fsck just run when

Possible, but not likely. Most drive firmware is well tested and robust.
It is more likely a power glitch did it.

> it happened).

You could always try running the program a few times to see if
you can recreate the failure. If you can, the manufacturer should
be contacted since this would represent a huge liability to them.

Robert

> On Tue, 6 Apr 2004, Robert Rolf wrote:
>
> > Sounds like you had a power bump or other glitch that scrambled
> > the on board cache, so when the track was written back, it
> > was full of crud. (big drives cache whole tracks for speed).
> >
> > It would be interesting to get the bad blocks report from the drive
> > to see how many blocks it substituted, and whether you still have
> > an ongoing bad blocks problem.
> >
> > If the utility was able to format the drive with no appreciable increase
> > in bad blocks, you might want to check on your power supply (cold solder
> > joints). If there are a large number of bad blocks, I wouldn't trust
> > the drive to not suddenly start plowing the media with the heads.
> >
> > Robert
> >
> >
> > "dr. Imre Bartfai" wrote:
> > >
> > > Hi,
> > >
> > > I do not know for this or other reason, but IBM has closed the mentioned
> > > plant some times ago. And as info for HDD reliability: I had a 40GB Maxtor
> > > HDD, which crashed very nasty: bad sectors, and a total mix up of an ext3
> > > filesystem, so practically no way to recover the data. However, I did not
> > > want to dump the drive so made a short look. Maxtor offers on their
> > > website a HDD repair program. I downloaded and ran it, and it cleaned up
> > > the drive perfectly. Since there (6 months) the drive runs without any
> > > problem (incl. badtracks). Misterious, eh?
> > >
> > > Regards,
> > > Imre
> > >
> > > On Mon, 5 Apr 2004, Michael Rigby-Jones wrote:
> > >
> > > > >-----Original Message-----
> > > > >From: Philip Pemberton [mailto:philpem@dsl.pipex.com]
> > > >
> > > > >The 75GXPs? I've got one of those in a fileserver. Update the
> > > > >firmware (there's a copy of the flash upgrader on IBM's
> > > > >website) and they're fine. I've also got a 35GXP which seems
> > > > >OK, too. The drive in my 386 is a 4GB Seagate U4 - 1999 fab
> > > > >date, still working fine.
> > > >
> > > > 75 and 60GXP.  The problem was isolated to drives made in one particular
> > > > plant (Hungary IIRC) and was down to build defect, so a firmware update is
> > > > not likely to help any dead drives.  Some peoples drives have lasted well
> > > > with no problems, but a huge number of them did not.
> > > >
> > > > Mike
> > > >
> > > > =======================================================================
> > > > This e-mail is intended for the person it is addressed to only. The
> > > > information contained in it may be confidential and/or protected by
> > > > law. If you are not the intended recipient of this message, you must
> > > > not make any use of this information, or copy or show it to any
> > > > person. Please contact us immediately to tell us that you have
> > > > received this e-mail, and return the original to us. Any use,
> > > > forwarding, printing or copying of this message is strictly prohibited.
> > > > No part of this message can be considered a request for goods or
> > > > services.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.