First, add -n to your ddrescue line.
Also, try with -d, sometimes slower, but better.
In a separated terminal, run
, looking at kernel messages, you'll soon learn to anticipate disk failure.
Code: Select all
watch 'dmesg | tail'
When you have done 10 first gigs, go further somewhere the disk is in better condition first ( add -i 50G for example).
If you have latest version of ddrescue (1.14 or more) you can use -R to do reverse copy, and combination with -s to limit high value (like -i but in the other way).
In my opinion, only good script you can do there is "the computer does an action when disk disconnect (beeps, mail, stops and so on...)". If you found time to do this, please release it as opensource script, I'll thank you every day of my life.