Seriously, it doesn't get any better than linux software raid with mdadm
Most distros allow you to create the array at install time with a simple interface.
Tis would be the steps to replace a failed raid1 drive:
/dev/md0= array with failed drive
/dev/hdc1=failed drive
Tell mdadm that the drive is failing:
Code:
mdadm --fail /dev/md0 /dev/hdc1
Tell mdadm to remove the drive from the array:
Code:
mdadm --remove /dev/md0 /dev/hdc1
Insert new drive that is properly formated and has the currect partition layout and add rebuild array:
Code:
mdadm --add /dev/md0 /dev/hdc1
Or if you want to manually make an array on any already running system:
(after creating the partitions with fdisk)
Code:
mdadm --create /dev/md0 --level=5 --raid-devices=4 /dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdd1
mkfs.ext3 /dev/md0
To monitor arrays:
cat /proc/mdstat
or
mdadm -D /dev/md0
Also configure the email address in /etc/mdadm.conf so that all warnings for the array will get emailed to you
To send a test failing email:
mdadm --monitor --scan --test
Also a good read:
http://linux.die.net/man/8/mdadm
It might look confusing but just search for some of the commands that I have used from above to get an understanding of what is going on (--create, --add, --fail, etc)