On 2014-07-06 05:36, tb75252 wrote:
> I did run Clonezilla Live with the --rescue option. Did not notice
> anything different in the way Clonezilla ran, i.e. got the same error
> message and /dev/sda2 was not saved.
You first said “–resque” - so which it is? Computers are not forgiving
with spelling errors.
Looking at the clonezilla web page, I find this note, with very little help:
http://drbl.org/faq/fine-print.php?path=./2_System/83_rescue_mode_for_clonezilla.faq#83_rescue_mode_for_clonezilla.faq
+++···········································
When I save an image, I see messages like “{ DriveReady Error }” or “{
DriveReady DataRequest Error }”, what can I do?
hda: possible failed opcode: 0xc8
dma_instr: status=0x41 { DriveReady Error }
hda: task_pio_intr: status=0x49 { DriveReady DataRequest Error }
io_all: errno = Input/output error(5)
It means that your hard drive has hardware issues, e.g. bad blocks.
You’d better to copy important data before you continue to use it. If
you want to use Clonezilla to save the image, you can try to enter
expert mode, choose “-q2” and “-rescue” (the rest of options can be
default ones) so that it can use the rescue mode of partclone to save
the image for you. However, this does not mean it always work. Good luck!
···········································+±
Notice that it is neither “–resque” nor “–rescue”, but “-rescue”, and
it has to be used in expert mode.
The clonezilla documentation I’m unable to find without booting the
live, but here there is a link:
http://sourceforge.net/projects/clonezilla/files/clonezilla_live_doc/QuickReference_Card_0.9.5/
It is obsolete, though, it Applies to Clonezilla v.1.2.12-10
But it says:
-rescue Continue reading past block read errors.
-q ntfsclone > partimage > dd
-q1 dd only
-q2 (default) partclone > partimage > dd
Well, that does not say that it will “solve somehow” bad block errors,
only that it will not stop and will continue working - which is what you
need.
(the full clonezilla doc is here:
http://www.clonezilla.org/clonezilla-live-doc.php)
The advanced options:
http://www.clonezilla.org/clonezilla-live/doc/01_Save_disk_image/advanced/09-advanced-param.php
> So… there’s no Linux utility that can repair bad sectors?
Repairing bad blocks is utterly impossible. Repairing bad blocks means
sending the disk back to a manufacturer so that it replaces the platter
surface, so forget it.
What is done is “bypassing” it. You put a notice in the road so that
cars skip the potholes.
For instance, you create a map of the badblocks, and reformat the device
telling the utility where the potholes are, and avoid them. The
resulting filesystem skips the holes.
Or, you cause a /write/ operation on the known holes. This causes the
disk firmware to re-map that particular bad sector to an area that was
reserved by the manufacturer for that purpose.
When I know that a disk has bad sectors, I intentionally write the
entire disk with something; zeroes, for instance. This finds and remaps
all bad sectors. Then I run a long SMART test for verification. If it
passes, perhaps I continue using the disk. If I see the number of bad
sectors increase, it goes to the garbage immediately.
On 2014-07-06 07:26, tb75252 wrote:
>
> I might have solved the problem…
>
> 1) I booted up using a copy of System Rescue CD and went to the command
> prompt;
> 2) I entered “e2fsck -cfpv /dev/sda2”, and when that finished running
> (it took a while)
-c This option causes e2fsck to use badblocks(8)
program to do a read-only scan of the device in
order to find any bad blocks. If any bad blocks
are found, they are added to the bad block inode
to prevent them from being allocated to a file
or directory. If this option is specified
twice, then the bad block scan will be done
using a non-destructive read-write test.
So, it maps the found bad sectors out.
What happens if a sector was used already by a file, huh? It does not
say… 
I would recommend you run the long SMART test ASAP, then consider
“garbaging” the disk.
–
Cheers / Saludos,
Carlos E. R.
(from 13.1 x86_64 “Bottle” at Telcontar)
)