6 replies to this topic

[Arnold72]

Hi,
Could somebody please explain the difference between cloning and imaging a hard drive please.
Im using macrium reflect and i always create a image for my back ups.
Is it worth my while cloning also?

Thanks.

[RJGoodhouse]

To Clone a Drive means Making an Exact Copy of a Drive on another Drive, sector by Sector. The terminology is probably not as precise as it should be and may vary depending on what software you are using to do the imaging/cloning. In the software I've used, the distinction between imaging a hard drive and cloning a hard drive boils down to when cloning, the mbr is imaged along with all the partitions on the drive and with imaging a hard drive, the mbr is not imaged, but rather just the partitions.
The mbr of a hard drive is the first 512 bytes(i.e. the first sector) which contains a the bootloader(if any) and the partition table which contains the partition info for all partitions on the hard drive. You would normally use the cloning operation when copying a hard drive over to a new hard drive. That way, the partition table would be copied over to the new drive and create the exact same partition structure on the new drive before dumping the data from the old drive onto the new drive. also the bootloader would be copied over to the new drive if the old drive was a bootable drive. Without that bootloader the new drive would not boot. For all practical purposes, an image and a clone are identical - it's just a clone is an exact copy where an image is an exact copy compressed into a single file or set of files. A clone can be installed and booted up, an image would have to be restored to a blank drive using the imaging software's restore process, which can usually be run from a bootable CD. The advantage to images is you can store more "stuff" on the hard drive you are using for storing images, or you can use alternate media (CD, DVD, etc.) for the image.

[David H. Lipman]

Imaging - The process of taking the exact file structure and partition scheme of a given disk and creating a disk file. That disk file may then be used as a a disaster recovery fail safe for a given system or be used to replicate that setup on another system. When used to replicate that setup on another system an image is acting as an intermediate step of cloning.

Cloning - Taking the exact file structure and partition scheme of a given disk and replicating it on another disk.

The actual act of cloning can be physical such as from one SATA to another SATA drive. The two drives do not have to be the same size but usually one would clone a smaller hard disk to a larger hard disk. Thus if I clone a 80GB drive with 20GB free to a 250GB drive I will have the exact same setup on the destination drive with 190GB free.

By the process of imaging a drive you can take the physical connection out of the picture. For example where you have a notebook and want to upgrade a system with a 160GB hard disk to a 320GB hard disk, and you have only one hard disk interface. You can create an image on an external hard disk. Once the image is made you can remove the 160GB hard disk and install a bare 320GB hard disk and restore the image. In the same respect, once an image is made you can use a network protocol to distribute the image to the destination system or multiple systems as a Unicast or as a Multicast. In a Unicast protocol clone you clone the drive of a source image to one destination computer. In a Multicast protocol clone you clone the drive of a source image to multiple computers at the same time.

When using an image in a disaster recovery objective you can restore the image back to the system when the system had a catastrophic failure such as a bad hard disk or a complete OS corruption or when a system becomes too compromised by malware to clean. Often when cleaning a system of malware you may image the drive just in case something goes wrong leaving the system un-bootable or completely unstable. Then you can restore the image and go through the cleaning process using a different modus operandi that won't leave the system un-bootable or completely unstable.

Note when creating an image of a given system you are taking a snapshot of the system at the time the system was imaged. Thus if you restore an image a week or month later it will be in a state of the moment that image was made.

[AdvancedSetup]

If the machine is part of a domain and you wait too long before you bring the reimaged system back online, you may need to either rejoin the domain or use a tool such as Netdom to reset and synchronize the appropriate shared secrets.

[David H. Lipman]

It has been a while since I have done it in a Domain environment but you would SysPrep the source system and then image the computer. Then when you booted the cloned system you would have to join the AD when it ran the mini-setup. That avoided any SID problems

[AdvancedSetup]

We actually never join the domain in the first place when building the image. Don't use SysPrep anymore either (used to). Normally use Ghost and create the image as a non-member workstation. Then join after dropping the image.

In fact having to follow the Microsoft model of imaging alone is reason enough for many shops to stay on XP. It takes a lot of up front time and effort to get things setup the way Microsoft wants you to image Windows 7 but on XP you don't need to setup anything, just Ghost it (or similar products) and deploy often in under 15 minutes. Doing similar on Windows 7 is not as easy, though it does offer more options if one does take the time to do so.


The Machine SID Duplication Myth (and Why Sysprep Matters)

[David H. Lipman]

Good information Ron. Thanx.