Restoring Small Business Server 2003 to New Hardware

April 29, 2009

Onsite Computer Services in New Orleans recently received a service call from a Microsoft Small Business Server 2003 (SBS) client in response to a server motherboard failure. The server is five years old. A replacement motherboard of the same type that failed is virtually impossible to locate and limits one’s options for repair.

The following procedure is not the preferred method of restoring a server, according to Microsoft and Onsite. It is provided only as reference for a possible disaster recovery scenario. In addition, this procedure is not recommended by Microsoft.

Steps to Perform Before Moving to New Hardware

1. Clean up the event log. Monitor the event log and clear up any red events. If any yellow warning events exist, determine if they could interfere with a restore to different hardware.

2. Update the Emergency Repair Disk (ERD) and repair directory. You can run the rdisk /s command to update the ERD and write critical information to the repair directory.

3. Perform a full tape backup (including the registry). Back up everything that pertains to the operating system and services. You must also back up the registry.

4. Disable virus scanner software. Before you disable virus scanner software, scan the disks or CD-ROMs on which your third-party drivers are stored.

5. Disable third-party services. This includes virus scanners, UPS software, remote administration programs, and tape backup software (you have already made the tape back up above)

6. Remove existing video drivers. Unless the new computer has the exact same video card, there could be enough of a difference in the hardware that the existing driver could generate an unrecoverable STOP error message. Remove the existing video driver, load the standard VGA driver, and set it to 16 colors (640×480).

7. Install the hard disk controller driver. Preload the hard disk controller that exists on the new computer. If you currently use IDE and the new system is SCSI, restoring a tape with only IDE drivers leaves the computer unbootable. The same applies if you are moving from SCSI to IDE. You may have to physically place the drive controller into the existing computer to load the driver. This is obviously not possible in situations where the controller is integrated.

8. Remove the SCSI hard disk controller driver. If the new computer does not have the same SCSI controller (or none at all), when you attempt to initialize the SCSI driver you may receive an unrecoverable STOP error message. NOTE: You can usually leave IDE drivers in place, which only generate a STOP event in the event log.

9. Install Microsoft Loopback. This is in preparation to remove the existing network adapter. This gives the protocols something to bind to and the protocols and protocol-dependent services do not have to be reloaded after the restore.

10. Remove the network adapter driver. Unless the new computer has the exact same network adapter, there may be enough of a difference in the hardware that the existing driver could generate an unrecoverable STOP error message.

11. Remove the sound card driver. Unless the new computer has the exact same sound card, there may be enough of a difference in the hardware that the existing driver could generate an unrecoverable STOP error message.

12. Remove any drivers for hardware that do not exist on the new computer. Do not attempt to initialize drivers for non-existent hardware on the new computer. This may cause unrecoverable STOP error messages, as well as device driver conflicts.

If you’re the lucky individual in charge of such a procedure, no doubt there is sufficient inspiration for a migraine. The prudent network engineer plans for disaster recovery BEFORE it actually occurs. To increase your chances of recovery after a disaster, Microsoft recommends the following procedures before a disaster happens:


• Check the event logs on a regular basis. A regular check of the event logs can alert you to problems before they become severe. Correct problems as they occur and then back up the computer afterwards.

Create a backup strategy. Nothing provides better recoverability than a good tape backup. Fault tolerant disk setups and imaging are also good solutions. If you create a tape backup and store the tape off-site, however, you have a better chance for recovery if the physical location of your server is damaged.

• Practice restoring the backup quarterly (or at least twice a year).

• OR… secure the services of an experienced Microsoft disaster specialist like Onsite Computer Services in New Orleans for all of your failure recovery needs.

If you’re searching for New Orleans computer service or repair of your Microsoft based system or network, then call the Microsoft specialists at 504-469-6991. Onsite Computer Services, Inc. in New Orleans

You may also be interested in Onsite Computer’s Blog.

This New Orleans On-Site Computer Service post was provided by Eli Lucas. Eli can be reached at Onsite Computer serves the Microsoft computing needs of small businesses throughout the Greater New Orleans area including Uptown, Lakeview, Broadmoor, Mid-City, Metairie, Jefferson, Kenner, Westbank, French Quarter, and the Warehouse District. Please visit our website for New Orleans Computer Service.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: