On Fri, 20 Jun 2014 17:28:20 +0200
Antonio Diaz Diaz <[email protected]> wrote:
> Paul L Daniels wrote:
> > I need ddrescue to exit (with non-zero code perhaps) as soon as the
> > read rate hits zero, so that I can then power down the HDD, wait a
> > few minutes, and then have it power back up, all using a script, ie;
>
> What about 'ddrescue --timeout=0'?
Thanks, that seems to be working on my initial tests.
I've written a small tool this evening to help me do the whole
thing cleanly, called it "waitfordevice" ( rather unoriginal
). Available as source at http://ctpc.biz/waitfordevice.c
I just build it using;
gcc waitfordevice.c -o waitfordevice
...and then in the script...
#!/bin/sh
while [ 1 -eq 1 ]; do
./powerSwitch on 3
./waitfordevice -d /dev/sde -v
sleep 2
ddrescue --timeout=0 -v -f /dev/sde drive.img drive.log
./powerSwitch off 3
echo "Sleeping..."
sleep 60
done
It works well for those drives that read a small amount and
then suddenly report a very large error, like 50MB read, 499GB
ERROR.
Using the above setup I've managed to now recover most of the
drive, where's previously I'd have just given up.
The "waitfordevice" tool also lets you put in a drive
signature / serial test to make sure you've got the same drive
again, ie; ./waitfordevice -v -d /dev/sde -s ST9750420AS
Regards,
Paul.
--
Computer Repairs for Charters towers - http://ctpc.biz
A.B.N. 19 500 721 806
_______________________________________________
Bug-ddrescue mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/bug-ddrescue
