Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have a Windows SBS (2003 R2) which is currently running Exchange 2003. It's now beginning to max out it's 25GB HDD and I have been given the job of upgrading it to a larger one (lucky me...).

Since the servers job is to host Exchange and a couple of very small/low traffic sites I am probably going to look at a Western Digital (Blue or Green) as the new HDD. I have read a lot of good things about them in terms of reliability/warrantys etc. (feel free to recommend any others).

My actual question is what is the best strategy for going about migrating Exchange over? I don't have any previous experience with this so I am assuming it goes along the lines of:

  1. Install Exchange onto new HDD
  2. Export existing Exchange database
  3. Import database into new Exchange server
  4. Decommission old HDD

I hope it's pretty much as straightforward as that, however, given that it's Exchange server I won't hold me breath...

share|improve this question
A a minimum consider the WDC Red drives, as they're meant to be used in RAID configurations. That "meant for RAID" is mostly important in the case of failure; it's the difference between a crashing server and just a failed drive. – Chris S Dec 31 '12 at 16:22
@ChrisS from speaking to other people I don't think those particular drives are the correct ones I should be looking at. I think I need to look at enterprise-class ones instead. – James Dec 31 '12 at 16:38
Well the XE and RE drives would be preferred over the Red drives; but the Blue and Green that you mentioned in your Question don't have TLER, which is vital to graceful failure. – Chris S Dec 31 '12 at 17:09
@ChrisS I think I would need to look at the XE/RE ones and avoid the Green/Blue completely as they aren't enterprise specific ones. I assume the server won't have a SAS controller so I would need to go with SATA. – James Dec 31 '12 at 17:12
The XE drives are SAS only. The RE have some of each. Red are SATA only... I'm not sure what point you're trying to make though. What particular "enterprise" features are you looking for? – Chris S Dec 31 '12 at 17:13

Ignoring for one minute the issue of running Exchange on desktop-class rather than enterprise class hardware...

I would look into using a disk cloning solution that can expand the volumes during the clone. That way, your existing configuration is unchanged and the process will take much less time that a fresh install of Exchange and a migration of the data and settings.

There are plenty of both paid-for and free options available alhtough not all will support growing the volumes during the clone. The Ultimate boot CD contains a number of options as a starting point and can be downloaded from

share|improve this answer
thanks for the response. How would I differentiate "desktop-class" hardware from enterprise-class? I don't have the spec of the machine in front of me so I can't give all the details, however, I seem to have said something that has made everyone think otherwise? Is it the choice of HDD? Like I mentioned in a comment in the other answer, I didn't purchase the server, however, it was bought for server specific reasons by someone who apparently does this for a living so I can only assume it is a proper server. – James Dec 31 '12 at 16:03
The Western Digital enterprise range can be found at – Phil Dec 31 '12 at 16:06
I've used Clonezilla with much success – chue x Dec 31 '12 at 16:52

Given the target audience of this site, no one here will agree with your recommendation of running Exchange on a desktop-class computer with desktop-class storage.

Get a real server with real enterprise grade storage.

Here's a generic list of reasons you want a real server for server-type workloads:

  • ECC memory: No silent corruption of important data before being committed to disk.
  • Dual power supplies: Resilience against power supply failure and branch circuit failure (if you use two different circuits).
  • Hot-swap RAID: replace failed drives, upgrade existing drives, or add new drives without shutting down the server.
  • Better MTBF: the components in a server are designed to last longer than desktop computers because they rated for continuous operation. Desktop computer component MTBFs are usually rated based on 40-80 hours a week.
share|improve this answer
Don't really see what your getting at here? The machine is more than capable of doing the job it was bought to has done so for the past 2 years. The HDD is now at max capacity and I need to upgrade it. I don't need enterprise grade storage, this is a small office which runs at max 3 people. Also, this is not an answer to my question, please delete and make it a comment. – James Dec 31 '12 at 15:20
Well, assuming it was not some "it challenged" Person buying SBS 2003 R2 after 2008... that machine is OLD. This means, it WILL fail. Replacing it after 5-6 years is what most sensible companies do. – TomTom Dec 31 '12 at 15:23
@TomTom Just to be clear, I didn't purchase the server and I agree SBS 2003 is old. However, it hasn't failed for the past 2 years and the strain/demand from the server is most likely not going to change so I don't understand why your adament it will fail? Obviously hardware in general will hit EOF at some point, however, there is no immediate reason to go out and spend money a new server when it's not needed. – James Dec 31 '12 at 15:27
I am not saying you bought it. I just Point out some (not smart People) may hahve purchased that in 2008 or 2009, so it may not be that old. If it is as old as the Version, it WILL fail. THere is a lifetime Limit on elements like Motherboard. Larger (and some smaller companeis) thus run regular replacement cycle. FOr example my Business will replace ALL (except some Special newer) Desktops this year, and we also replace all our core Servers which still are quad core machines. Not that tey fail, it just is cheaper (we Need more capacity anyway) and... well, I can PLAN when we do that. – TomTom Dec 31 '12 at 15:31
This is exactly the reason you should get a Real Server now. If you have a real server, you could 1) pop out one of the mirrored drives and insert a bigger one, 2) wait for it to resync, 3) replace the other mirrored drive, 4) wait for it to resync, and 5) grow the volume. Don't extend past mistakes; plan for the future! – longneck Dec 31 '12 at 15:39

Another option, if there is room in the server, is simpliy to ADD, not replace the existing HDD. Add the new disk (or disks, if you can add RAID when you didn't before) and format the drive, then simply use the tools available to move the Exchange data store. This link has detailed instructions on how to do so:

share|improve this answer

I recently did this on a desktop, so it seems to work:

It may be possible to "move" the server without actually needing to reinstall anything at all.

  1. Install your second hard drive in your server.
  2. Set your main hard drive as a dynamic disk in the "Disk Management" pane of the computer management administrative tool.
  3. Add the new hard drive as a mirror of the old hard drive.
  4. Wait for drives to synchronize.
  5. Break the mirror by removing the old drive.
  6. Expand the partition to take up the full size of the new hard-drive.

It let me move my desktop from a hard drive that was having SMART errors without any trouble, so it should work for you too.

share|improve this answer
I like the idea of this would be ideal if I didn't need to re-install anything. I have never mirrored a drive before I assume it migrates everything including the OS? – James Jan 1 '13 at 13:29

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.