Copy process over hardware failure is endless

[wwp]

Hello Keith!


On Tue, 8 Sep 2009 09:39:21 +0100 (BST) Keith Roberts <keith at karsites.net> wrote:

> On Tue, 8 Sep 2009, wwp wrote:
> 
> > To: mc at gnome.org
> > From: wwp <subscript at free.fr>
> > Subject: Re: Copy process over hardware failure is endless
> > > Hello Frank, Keith,
> >
> >
> > On Mon, 7 Sep 2009 20:42:01 +0200 Frank Dietrich <ABLEsoft at gmx.de> wrote:
> >
> >> Hi,
> >>
> >> wwp <subscript at free.fr> wrote:
> >>> In /var/log/messages I'm getting tons of:
> >>> Sep  7 14:48:03 monolith kernel: sd 8:0:0:0: [sdd] Add. Sense: No
> >>> additional sense information Sep  7 14:48:32 monolith kernel: sd
> >>> 8:0:0:0: [sdd] Sense Key : No Sense [current] Sep  7 14:48:32
> >>> monolith kernel: Info fld=0x0
> >>
> >> Wouldn't it be more secure for your data to save first the whole
> >> partition/disc with ddrescue or similar, to reduce the stress on the
> >> dying disc? And after that copying your important data from this
> >> image? The only reason which could prevent you not to do this could be
> >> the size of the disc.
> >
> > Yes that's probably what I'll do. I appreciate your suggestion (and
> > Keith's) one. Both suggestion will be hard to deploy, as it seems my
> > disk self-powers down regularly.
> 
> Maybe you could change the power-down defaults with
> 
> yum list *hdparm*
> 
> You will need to read *all* the manual page before using hdparm, as there are some options that are not safe, or experimental.
> 
> Use it at your own risk - especially on a faulty drive!
> 
> I use it to send my backup IDE drive into sleep mode, between hourly backups.
> 
> '-S' option
> 
> Set the standby (spindown) timeout for the drive.  This value is used by the drive to determine how long to wait (with no disk activity)  before turning off the spindle motor to save power.

(I know about the hdparm -S thing, thanks anyway)

I'm afraid those power-downs directly or not are due to a hardware
problem. The most amazing thing (and a pretty good news for my data
recovery) is that those power-downs don't happen while ddrescue is
copying data, whereas they are showing up regularly when using mc or
cp.. Maybe a temperature problem and the disk safety mechanisms were
shutting down the disk to prevent breakage.. Apparently this is not due
to a specific area in the disk, since ddrescue completed the copy w/o
problem and I'm now able to mount it.

Thanks again!


Regards,

-- 
wwp
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.midnight-commander.org/pipermail/mc/attachments/20090908/306f570c/attachment.asc>