Riccardo Riva

Hi all, the following will be a quick and dirt list of commands for use “dd” in a linux system for various task.

First of all you should create an Hard Disk Image, for example if you want to create an image of your /dev/sda device, you should have another disk (with a writable partition) (i.e. /dev/sdb) with a directory (i.e. /home mounted on it) and simply type :

dd if=/dev/sda of=/home/sda.bin

Or even a partition backup using the same disk device for source and destination like the following (if for example you have /home mounted on /dev/sda2) :

dd if=/dev/sda1 of=/home/sda1.bin

You should even create a compress image of the same disk above, using GZIP, simply type the following :

dd if=/dev/sda1 | gzip > /home/sda1.bin.gz

One of the most useful use of dd (it’s saved my life a lot of time) will be :

dd if=/dev/sda of=sda.boot.mbr bs=512 count=1

With the above command, you have backuped up your MBR (Master Boot Record) of your /dev/sda device and in case of disaster you should restore it, for example booting with a live CD using the following :

dd if=sda.boot.mbr of=/dev/sda bs=512 count=1

You should also create a manual RAID1 between two disk by executing the followinf script with a cron job :

#!/bin/sh
#

LOG="/var/log/mirror.log"
ADMIN="yourmail@yourprovider.ext"
ERROR=0

echo `/bin/date` >$LOG 2>&1
/bin/dd if=/dev/sda of=/dev/sdb bs=1M >>$LOG 2>&1
ERROR=$?
echo `/bin/date` >>$LOG 2>&1

if [ $ERROR -ne 0 ]; then
cat $LOG | /bin/mail -s "Report mirror `uname -n`" $ADMIN
fi

The above sript will copy the entire /dev/sda to /dev/sdb logging the process and send an email to your email address at the end for debug and monitoring purpose.
If the first hard disk (/dev/sda) fail, you should phisicaly umount it, and subsitute it with /dev/sdb then reboot the system.

Hope this help

Bye
Riccardo

Bookmark It