All commands will be inside of a little fancy box like this. Comments will start off with a hash mark. Commands will be posted in order that they should be typed. Some commands will take a while to process. Some commands will do nothing to the system (readonly command) meaning they only spit out information for you to see. Other commands will do writes, meaning they can do some damage, or some good (hopefully the later). All commands in the computer universe are either read or write or both.

# comment, read this, but dont type this line (if you do type it in, make sure to include the # with it, that way bash/linux will know to ignore it). bash & linux ignore lines that start off with #. 
command1
# this is command2, and should be run after command1
command2

# this next command shows all of the detected drives and thier partitions. We need to know this because once we add more drives, we will know what are the old drives (which we will now see) and what are the new drives (which we will see later)

cat /proc/partitions 

# I recommend to record what you see here as the "before adding source drive", just note it on a piece of paper or copy it to a notepad program

# dmesg shows a running system log of the linux system (so it will show when new hardware is attached or failed or etc), we dont need to see that whole dmesg output, so we just see the last few lines by "piping" it to tail (which just shows the last 5 or so lines). if that wasnt enough for you, you can view the whole output with: "dmesg"
dmesg | tail
# now that we added the source drive and dmesg should of reported it being added we will see it added to "cat /proc/partitions". If you dont see anything, wait a minute. If you dont see anything and your sure the drive has been added, then it means the drive could be completely damaged - I recommend trying a different port/slot, but if that doesnt work, take it to data recovery solution that can clone completely gone-bad drives (such a vendor is http://www.datarecoverynj.com/)
cat /proc/partitions
# here is another way to view the partition details of all drives. you will see the source drive
fdisk -l
# its important to realize the letter of the drive, we attached the drive, and we see it came up as "sda" and it probably also had 3 partitions (more or less) "sda1", "sda2", "sda3"
# we simply need to rememember "sda" is our source drive

# similar process as when we attached the source drive
dmesg | tail
cat /proc/partitions
fdisk -l
# the destination drive probably shows up as "sdb" and it probably has no partitions as its a brand new drive, so we dont see things like "sdb1", "sdb2", etc...

smartctl -a /dev/sda
# smartctl has alot of output with alot of whitespace, here is a trick to make all the empty lines of the output go away for easier reading:
# smartctl -a /dev/sda | grep .
# or since we are only interested in the serial number
# smartctl -a /dev/sda | grep -i "serial" 
# grep is a line by line search utility in linux. When we grep for a ., it means to find every none empty line. when we ask grep to do -i, it means dont be case sensitive, so when we ask to find "serial" it will also find "SeRIAL" or the more likely "Serial".

# this is a mini script that i recommend copy pasteing in, but if you cant, re-write it. it will run smartctl on every drive letter a thru z, and it will only show serial number, ata errors, model number, user capacity etc...(the info we need):

for i in a b c d e f g h i j k l m n o p q r s t u v w x y z; do echo "===drive sd$i==="; smartctl -a /dev/sd$i | egrep -i "reallocated_sector|ata error|serial|model|user capacity"; done;

# or try this one which also looks at current_pending sectors & offline_uncorrectable (some disks dont have reallocated sectors or ata errors, but are still badvia other variables such as current_pending sector and offline_uncorrectable)

for i in a b c d e f g h i j k l m n o p q r s t u v w x y z; do echo "===drive sd$i==="; smartctl -a /dev/sd$i | egrep -i "reallocated_sector|ata error|serial|model|user capacity|current_pending|offline_uncorrectable"; done;

SIDENOTE: above I talked about sda being the source drive, and sdb being the sdb the destination drive. Here we switch gears to something more realistic (as the system your using probably already has another system drive in it), so the source drive will have a letter like sdc and the destination drive will have sdd

cd ~
pwd

SIDENOTE: the above just gets you to the currently logged in users home directory. Which should be /root/ if your the root user, or /home/user1 if your user1 user. Or /home/knoppix if your the knoppix user (knoppix operating system has a knoppix user which is like their main admin user – note that the root user still has more priviledges than the knoppix user, but for this article the knoppix default user should be sufficient for use).  Keep this in mind, if for any command, you get an error saying “you dont have permissions” or something like this, preceed the command with “sudo”, so if you didnt have permission to run a command like “ls -lisah” then run it like this “sudo ls -lisah” (note in real life ls -lisah is always allowed by any user, im just trying to use it to prove a point).

# the first command copies/clones data from sdc to sdd, but it doesnt copy over any areas that give it trouble (the error areas are skipped). it keeps track of what it copied and didnt copy to ddrlog.txt, so the next time the same command (in case the command got cancelled & you had to rerun the commamd) or the next command is ran it will know where to begin to clone. The second command will read ddrlog.txt, which after the first command finishes has all of the info of where the errors are (its the areas that didnt get copied), and it will copy over the bad/error areas. If any blocks give it troubles, it will retry them an extra time (-r1). So it tries bad blocks a total of 2 times before giving up and moving on to next block. If you want it to try 6 times change it to -r5 (the first time counts as try along with 5 retries, thats a total of 6 times).

ddrescue -v -n /dev/sdc /dev/sdd ddrlog.txt
ddrescue -v -r1 /dev/sdc /dev/sdd ddrlog.txt

# --- everything below, is if the commands didnt run due to permission error, or the program asking to use "--force" --- #

# --- move on to next step if above commands worked --- #

# if program complains about some permissions issue try it with sudo (it will then ask for the users password, actually nevermind it wont as knoppix doesnt require sudo password):

sudo ddrescue -v -n /dev/sdc /dev/sdd ddrlog.txt
sudo ddrescue -v -r1 /dev/sdc /dev/sdd ddrlog.txt  

# --- move on to next step if above commands worked --- #

# if program complains about needing to use "--force", run it like this:

ddrescue -v -f -n /dev/sdc /dev/sdd ddrlog.txt
ddrescue -v -f -r1 /dev/sdc /dev/sdd ddrlog.txt 

# or if you need to run it with sudo and force:

sudo ddrescue -v -f -n /dev/sdc /dev/sdd ddrlog.txt
sudo ddrescue -v -f -r1 /dev/sdc /dev/sdd ddrlog.txt 

# We add "-R" for reverse (again this is optional as the gddrescue algorithm attacks the cloning problem in its own interesting way that probably doesnt require going in reverse -- well atleast thats according to their "man" page)

ddrescue -v -R -n /dev/sdc /dev/sdd ddrlog.txt
ddrescue -v -R -r1 /dev/sdc /dev/sdd ddrlog.txt

# -- if it complains about permissions use sudo -- #

ddrescue -v -R -n /dev/sdc /dev/sdd ddrlog.txt
ddrescue -v -R -r1 /dev/sdc /dev/sdd ddrlog.txt

# -- if it complains about "-force" -- #

ddrescue -v -f -R -n /dev/sdc /dev/sdd ddrlog.txt
ddrescue -v -f -R -r1 /dev/sdc /dev/sdd ddrlog.txt

# -- if it complains about "-force" and you need to use sudo -- #

sudo ddrescue -v -f -R -n /dev/sdc /dev/sdd ddrlog.txt
sudo ddrescue -v -f -R -r1 /dev/sdc /dev/sdd ddrlog.txt

# just like from previous step:

for i in a b c d e f g h i j k l m n o p q r s t u v w x y z; do echo "===drive sd$i==="; smartctl -a /dev/sd$i | egrep -i "reallocated_sector|ata error|serial|model|user capacity"; done;

# or try this one which also looks at current_pending sectors & offline_uncorrectable (some disks dont have reallocated sectors or ata errors, but are still badvia other variables such as current_pending sector and offline_uncorrectable)

for i in a b c d e f g h i j k l m n o p q r s t u v w x y z; do echo "===drive sd$i==="; smartctl -a /dev/sd$i | egrep -i "reallocated_sector|ata error|serial|model|user capacity|current_pending|offline_uncorrectable"; done;

# ps aux lists the processes in linux for all current users
ps aux
# well we are only interested in ddrescue, grep usualy finds itself in the process list, so we tell grep to exclude itself using 'grep -v "grep"', which means dont show me lines that have the word "grep in them
ps aux | grep "ddrescue" | grep -v "grep"

# another variation (note you can put the [] around any character in ddrescue), the [] makes it so grep doesnt show up.
ps aux | grep "[d]drescue"
# SIDENOTE: here is an article explaining why [] works: http://unix.stackexchange.com/questions/74185/how-can-i-prevent-grep-from-showing-up-in-ps-results

shutdown -h now

ddrescue vs gddrescue (this article used gddrescue, which has the command line name: ddrescue – the older program is annoyingly called ddrescue and it has the command line name: dd_rescue, just remember the underscore command is older)