This topic applies to the Microsoft® Windows Server™ 2003 operating system, Standard Edition, and, except where noted, the x64-based version of Windows Server 2003, Standard Edition.

Getting Ready for an Upgrade

This topic can help you plan for an upgrade to Microsoft® Windows Server™ 2003, Standard Edition, in a network with one to five servers and 100 or fewer clients. The information in Getting Started is designed to help with the upgrade or installation of such a network. For help with planning the upgrade or installation of a larger number of servers, see the Windows Server 2003 Deployment Kit. You can view the Windows Deployment and Resource Kits on the Web at: http://go.microsoft.com/fwlink/?LinkID=29883

For technical information and resources needed to evaluate, deploy, and administer the Windows server operating systems, see the Windows Server TechCenter on the Web at: http://go.microsoft.com/fwlink/?LinkID=28944

For detailed information about how to run Setup, see "Running Setup for an Upgrade" later in this topic.

If you want to plan for a new installation, see "Getting Ready for a New Installation" later in this topic.

After completing the upgrade of the first server, you can get detailed instructions about how to use products in the Windows Server 2003 family by opening Help and Support Center. To open Help and Support Center, click Start, and then click Help and Support.

Upgrades Compared to New Installations

This section compares upgrading to performing a new installation, to help you confirm your decision to upgrade. Upgrading is either replacing Windows NT® 4.0 (with Service Pack 5 or later) with a product in the Windows Server 2003 family, or replacing Windows 2000 with a product in the Windows Server 2003 family. Installing, in contrast with upgrading, means completely removing the previous operating system, or installing a product in the Windows Server 2003 family on a disk or disk partition with no previous operating system.

The following lists describe items to consider when deciding between an upgrade and a new installation:

Points to consider for an upgrade:

With an upgrade, configuration is simpler, and your existing users, settings, groups, rights, and permissions are retained.

With an upgrade, you do not need to re-install files and applications. As with any major changes to the hard disk, however, it is recommended that you back up the disk before beginning an upgrade.

Before planning to perform an upgrade, see "Operating Systems from Which You Can Upgrade" later in this topic.

If you are upgrading in a domain that includes domain controllers running Windows 2000, be sure to read "Upgrades in a Domain Containing Windows 2000 Domain Controllers" later in this topic.

If you are upgrading in a domain where all domain controllers run Windows NT 4.0, be sure to read "Upgrades in a Windows NT 4.0 Domain" later in this topic.

If you want to upgrade and then use the same applications as before, be sure to review applications information in Relnotes.htm (on the Setup CD). Also, for the most recent information about compatible applications for products in the Windows Server 2003 family, see the software compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37604

Points to consider for a new installation:

If you reformat your hard disk and then perform a new installation, the efficiency of your disk might improve (compared to not reformatting it). Reformatting also gives you the opportunity to modify the size or number of disk partitions, to make them match your requirements more closely.

If you want to practice careful configuration management, for example, for a server where high availability is important, you might want to perform a new installation on that server instead of an upgrade. This is especially true on servers on which the operating system has been upgraded several times in the past.

It is possible to install Windows Server 2003, Standard Edition, and also allow the computer to sometimes run another operating system. Setting up the computer this way, however, presents complexities because of file system issues. For more information, see "Deciding Whether a Computer Will Contain More Than One Operating System" in "Getting Ready for a New Installation" later in this topic.

Note

If you used Windows NT 4.0 to create a volume set, mirror set, stripe set, or stripe set with parity, and you want to run Setup for Windows Server 2003, Standard Edition, on that computer, you must prepare the disk set first. For details, see "Working with Volume, Mirror, or Stripe Sets or Stripe Sets with Parity" later in this topic.

Operating Systems from Which You Can Upgrade

If you upgrade, Setup automatically installs Windows Server 2003, Standard Edition, into the same folder as the currently installed operating system. You can upgrade to Windows Server 2003, Standard Edition, from the following versions of Windows:

Windows NT Server 4.0 with Service Pack 5 or later
Windows NT Server 4.0, Terminal Server Edition, with Service Pack 5 or later
Windows 2000 Server

Note

You cannot upgrade from an earlier version of a Windows server operating system to an Itanium-based or x64-based version of a Windows server operating system.

Remote Storage is not included on Windows Server 2003, Standard Edition. If you are using Windows 2000 Server with Remote Storage, you cannot upgrade to Windows Server 2003, Standard Edition. You can, however, upgrade to Windows Server 2003, Enterprise Edition, because Remote Storage is included in this product.

If you have Windows 2000 Server with the Remote Storage component installed, but you are not using Remote Storage, you can remove the component (through Add/Remove Programs in Control Panel) and then upgrade to Windows Server 2003, Standard Edition.

If you have Windows NT 4.0 Server Enterprise Edition, you can upgrade to Windows Server 2003, Enterprise Edition, but not Windows Server 2003, Standard Edition. Before upgrading from Windows NT 4.0, you must apply Service Pack 5 or later.

Note

If you have a version of Windows NT earlier than 4.0, you cannot upgrade directly to a product in the Windows Server 2003 family. You must first upgrade to Windows NT 4.0 and apply Service Pack 5 before upgrading to a product in the Windows Server 2003 family.

System Requirements and Hardware Compatibility

Before upgrading your server, make sure that you have chosen hardware that supports Windows Server 2003, Standard Edition.

System Requirements

To ensure adequate performance, make sure that computers on which you upgrade to Windows Server 2003, Standard Edition, meet the following requirements:

One or more processors with a recommended minimum speed of 550 MHz (minimum supported speed is 133 MHz). A maximum of four processors per computer is supported. Processors from the Intel Pentium/Celeron family, AMD K6/Athlon/Duron family, or compatible processors are recommended.
256 megabytes (MB) of RAM recommended minimum (128 MB minimum supported; 4 gigabytes (GB) maximum).
A hard disk partition or volume with enough free space to accommodate the setup process. To ensure that you have flexibility in your later use of the operating system, it is recommended that you allow considerably more space than the minimum required for running Setup, which is approximately 1.25 GB to 2 GB. The larger amount of space is required if you are running Setup across a network instead of from a CD-ROM, or if you are upgrading on a FAT or FAT32 partition (NTFS is the recommended file system).

In addition, a domain controller upgrade from Windows NT 4.0 could require much more space than other upgrades, because the existing user accounts database can expand by as much as a factor of ten during the upgrade, as Active Directory® functionality is added.

Note

The setup process requires the free disk space described in the previous paragraphs. After Setup is finished, actual hard disk space used for the operating system will be more than the free space required for Setup, because of space needed for the paging file, for any optional components you install, and (on domain controllers) for user accounts and other Active Directory® information. The usual size for the paging file is 1.5 times the size of the RAM. For information about the paging file, optional components, user accounts, and information stored in Active Directory, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

VGA or higher-resolution monitor (Super VGA 800x600 or higher recommended), keyboard, and (optionally) a mouse or other pointing device.
As an alternative, for operation without a monitor or keyboard, you can choose a remote diagnostic and support processor that is designed for products in the Windows Server 2003 family. For details, see the hardware compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37824

For CD-ROM installation:

A CD-ROM or DVD drive.

For network installation:

One or more network adapters and related cables that are designed for products in the Windows Server 2003 family. For details, see the hardware compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37824
A server from which to offer network access for the Setup files.

For additional hardware-based functionality:

Appropriate hardware for the functionality you require. For example, if you require sound, you must have a compatible sound card with speakers or headphones, and if you plan to support network clients, the servers and clients must have appropriate network adapters and cables. For details about your hardware, see the hardware compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37824

Hardware Compatibility

One of the most important steps to take before upgrading your server is to confirm that your hardware is compatible with products in the Windows Server 2003 family. You can do this by running a preupgrade compatibility check from the Setup CD or by checking the hardware compatibility information at the Windows Catalog Web site. Also, as part of confirming hardware compatibility, check to see that you have obtained updated hardware device drivers and an updated system BIOS.

Regardless of whether you run a preupgrade compatibility check, Setup checks hardware and software compatibility at the beginning of an upgrade and displays a report if there are incompatibilities.

Running a preupgrade compatibility check from the Setup CD

You can run a hardware and software compatibility check from the Setup CD. The compatibility check does not require you to actually begin an upgrade. To run the check, insert the Setup CD in the CD-ROM drive and, when a display appears, follow the prompts for checking system compatibility. You will be offered the option to download the latest Setup files (through Dynamic Update) when you run the check. If you have Internet connectivity, it is recommended that you allow the download.

For more information about downloading the latest Setup files, see "Using Dynamic Update for Updated Drivers and Other Setup Files" later in this topic.

Another way to run the compatibility check is to insert the Setup CD in the CD-ROM drive, open a command prompt, and type:

d:\i386\winnt32 /checkupgradeonly

where d represents the CD-ROM drive.

Checking hardware and software compatibility information on the Microsoft Web site

Another way to confirm that your hardware and software are designed for products in the Windows Server 2003 family is to check the hardware and software compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37823

Note that one type of hardware no longer supported is the microchannel bus.

Checking drivers and system BIOS

Check that you have obtained updated drivers for your hardware devices and that you have the latest system BIOS. The device manufacturers can help you obtain these items. For information about the ACPI standard for the BIOS, see "Understanding ACPI BIOS" later in this topic.

Finally, if you have devices that do not use Plug and Play, or you are aware that your Plug and Play devices are not implemented exactly to the standards, consider taking a device inventory of the hardware devices in your computer. For more information, see the next section, "Taking an Inventory of Devices That Do Not Use Plug and Play."

Taking an Inventory of Devices That Do Not Use Plug and Play

This section describes steps you can take if the devices in your computer do not use Plug and Play technology. For important information about hardware compatibility (including the compatibility of devices), be sure to read "System Requirements" and "Hardware Compatibility" earlier in this topic. For specific information about using a mass storage controller (such as a SCSI, RAID, or Fibre Channel adapter) with a driver that was supplied by the manufacturer, see the next section, "Mass Storage Drivers and the Setup Process."

Products in the Windows Server 2003 family include Plug and Play technology so that devices (for example, video and network adapters) can be automatically recognized by the operating system, configuration conflicts are avoided, and you do not have to specify each device’s settings by hand. However, if you have devices that do not use Plug and Play, or you are aware that your Plug and Play devices are not implemented exactly to the standards, you might want to take steps to avoid device configuration conflicts. This section describes steps you can take, if you choose, to understand your device configuration before running Setup.

To take an inventory of your devices, use the existing operating system to obtain the current settings, such as memory address and interrupt request (IRQ), used with your devices. For example, with Windows NT 4.0, you can use Control Panel to view settings (on the Start menu, point to Settings, click Control Panel, and then double-click icons such as Network and Ports). You might also choose to view system BIOS information. To do this, watch the screen while starting the computer, and then press the appropriate key when prompted.

At the beginning of an upgrade, the Setup program automatically takes a device inventory as well. For devices that do not use Plug and Play, or that are not implemented exactly to Plug and Play standards, taking your own inventory helps prevent the following difficulties:

If two or more adapters share IRQ settings or memory addresses, the Setup program might not be able to resolve the conflict. To prevent this, you can take one of two approaches.

You can remove one of the adapters before running Setup and re-install it afterward. For information about installing and configuring adapters and other hardware devices, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

As an alternative, you can modify one adapter’s IRQ settings and memory addresses before running Setup, so that each adapter’s settings are unique.
If adapters do not respond in a standard way to the attempts by Setup to detect or enumerate them, Setup might receive indecipherable or inaccurate information. In this case, you might need to remove these devices before running Setup, and re-install and configure them afterward. For information about installing and configuring adapters and other hardware devices, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

The following table lists the kinds of information to gather if you have devices that do not use Plug and Play, and you decide to take a device inventory before starting Setup.

Adapter	Information to gather
Video	Adapter or chip set type and how many video adapters
Network	IRQ, I/O address, Direct Memory Access (DMA) if used, connector type (for example, BNC or twisted pair), and bus type
SCSI controller	Adapter model or chip set, IRQ, and bus type
Mouse	Mouse type and port (COM1, COM2, or PS/2) or USB
I/O port	IRQ, I/O address, and DMA (if used) for each I/O port
Sound adapter	IRQ, I/O address, and DMA
Universal serial bus (USB)	Which devices and hubs are attached
PC card	Which adapters are inserted and in which slots
Plug and Play	Whether enabled or disabled in BIOS
BIOS settings	BIOS revision and date
External modem	COM port connections (COM1, COM2, and so on)
Internal modem	COM port connections; for nonstandard configurations, IRQ and I/0 address
Advanced Configuration and Power Interface (ACPI); Power Options	Enabled or disabled; current setting
PCI	Which PCI adapters are inserted and in which slots

Mass Storage Drivers and the Setup Process

If you have a mass storage controller (such as a SCSI, RAID, or Fibre Channel adapter) for your hard disk, confirm that the controller is designed for products in the Windows Server 2003 family by checking the hardware and software compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37823

If your controller is compatible, but you are aware that the manufacturer has supplied a separate driver file for use with products in the Windows Server 2003 family, obtain the file (on a floppy disk) before you begin Setup. During the early part of Setup, a line at the bottom of the screen will prompt you to press F6. Further prompts will guide you in supplying the driver file to Setup so that it can gain access to the mass storage controller.

If you are not sure whether you must obtain a separate driver file from the manufacturer of your mass storage controller, you can try running Setup. If the controller is not supported by the driver files on the Setup CD, and therefore requires a driver file that is supplied by the hardware manufacturer, Setup stops and displays a message saying that no disk devices can be found, or it displays an incomplete list of controllers. After you obtain the necessary driver file, restart Setup, and press F6 when you are prompted.

Using a Custom Hardware Abstraction Layer (HAL) File

If you have a custom Hardware Abstraction Layer (HAL) file supplied by your computer manufacturer, before you begin Setup, locate the floppy disk or other medium containing the file. During the early part of Setup, a line at the bottom of the screen will prompt you to press F6: at this time press F5 (not F6). After you press F5, follow the prompts to include your HAL file in the Setup process.

Understanding ACPI BIOS

The basic input/output system (BIOS) is a set of software through which the operating system (or Setup) communicates with the computer’s hardware devices. The Advanced Configuration and Power Interface (ACPI) is the current standard for the way the BIOS works. Products in the Windows Server 2003 family support not only ACPI-compliant BIOS versions, but also some BIOS versions based on older advanced power management (APM) and Plug and Play designs.

Some ACPI-based BIOS versions are not compliant with the standard. The more recent the version of an ACPI BIOS, the more likely that it is compliant. An ACPI-based BIOS that is not compliant with the ACPI standard might not support workable communication between the operating system (or Setup) and your hardware. If workable communication is not supported, Setup stops and displays instructions for contacting your hardware manufacturer and taking other steps to solve the problem. If this happens, follow the instructions provided.

To learn more about the ACPI compliance of your BIOS:

For information about your BIOS version, before running Setup, restart the computer and watch the text on the screen. Pay particular attention to blocks of text containing the words "BIOS" or "ACPI BIOS."
For information about BIOS versions for your hardware, check your hardware documentation and contact your hardware manufacturer.

Using Dynamic Update for Updated Drivers and Other Setup Files

If you have a working Internet connection on the computer on which you run Setup, you can choose Dynamic Update during Setup and obtain the most up-to-date Setup files, including drivers and other files. Whenever an important update is made to any crucial Setup file, that update is made available through Dynamic Update functionality built into the Windows Update Web site. Some of the updated files will be replacements (for example, an updated driver or updated Setup file) and some will be additions (for example, a driver not available at the time that the Setup CD was created). It is recommended that you use Dynamic Update when running Setup.

Dynamic Update has been carefully designed so that it is reliable and easy to use:

The files on the Dynamic Update section of the Windows Update Web site have been carefully tested and selected. Only files that are important in ensuring that Setup runs well are made available through Dynamic Update. Files with minor updates that will not significantly affect Setup are not part of Dynamic Update.
Because Dynamic Update downloads only the files that are required for your computer, the Dynamic Update software briefly examines your computer hardware. No personal information is collected, and no information is saved. The only purpose is to select appropriate drivers for your hardware configuration. This keeps the download as short as possible and ensures that only necessary drivers are downloaded to your hard disk.
You can use Dynamic Update when running a preupgrade compatibility check from the product CD, or when running Setup itself. Either way, you obtain the most up-to-date files for running Setup. For information about running the compatibility check, see "Hardware Compatibility" earlier in this topic.
You can use Dynamic Update with unattended Setup. Preparing for this requires several steps. For a brief description of unattended Setup, see the "Planning for Unattended Setup" section in "Running Setup for an Upgrade" later in this topic. For details about how to use Dynamic Update with unattended Setup (also called automated installation), see "Automating and Customizing Installations" in the Windows Server 2003 Deployment Kit.

The Windows Update Web site offers a variety of updates that you can use after completing Setup. To learn more, go to: http://go.microsoft.com/fwlink/?LinkID=284

Important Files to Review

At some point in your planning process, before you run Setup, familiarize yourself with the Relnotes.htm file found on the Setup CD.

This file contains important usage information about hardware, networking, applications, and printing. Also familiarize yourself with information about hardware compatibility for products in the Windows Server 2003 family. For more information, see "Hardware Compatibility" earlier in this topic.

For the most recent information about compatible applications for products in the Windows Server 2003 family, see the software compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37604

Upgrades in a Domain Containing Windows 2000 Domain Controllers

If you are upgrading servers in a domain containing Windows 2000 domain controllers, there are a few points to keep in mind. The following points are the most important:

The first step in planning the upgrade of a particular server is to check the compatibility of the hardware with products in the Windows Server 2003 family. For more information, see "Hardware Compatibility" earlier in this topic.
Before you upgrade the first domain controller in a domain where one or more domain controllers run Windows 2000, or add a domain controller running a product in the Windows Server 2003 family, you must prepare the domain (and the forest in which it is located) using a simple tool on the Setup CD for Windows Server 2003, Standard Edition. For information, see "Preparing for Upgrades in a Domain Containing Windows 2000 Domain Controllers" later in this topic.
Before preparing the domain (and the forest in which it is located), it is recommended that you apply Service Pack 2 or later to all domain controllers running Windows 2000.
If you have Internet Information Services (IIS) 5.0 installed on a server, when you upgrade the operating system, IIS is upgraded to IIS 6.0. However, for application compatibility, IIS runs in IIS 5.0 isolation mode after the upgrade. For more information about IIS isolation modes, see the IIS 6.0 Help.
The Internet Explorer Enhanced Security Configuration is enabled by default when you upgrade. The security settings in this configuration can help make your computer more secure by limiting its exposure to malicious Web sites. Therefore, with this enhanced level of security, you might find that some Web sites do not display correctly in Internet Explorer when you are browsing Internet and intranet Web sites. Also, you might be prompted to enter your credentials when accessing network resources, such as files in Universal Naming Convention (UNC) shared folders. You can easily change the enhanced security settings. For more information, see Help and Support Center (after completing Setup, click Start, and then click Help and Support).
After upgrading the first server, you can read topics in Help and Support Center about the features that are most useful to you in products in the Windows Server 2003 family.

One concept to read about is domain and forest functional levels, which are levels of Active Directory functionality related to the mix of operating system versions on your domain controllers. As you complete domain controller upgrades and reduce the number of different operating system versions running on them, you can adjust the domain and forest functional levels appropriately.

To view Help and Support Center on a server running Windows Server 2003, after completing Setup, click Start, and then click Help and Support.

If you have servers running Windows NT, also review the following points:

Before running Setup, it is recommended that you review the file systems and partitions that exist on the server. You must have at least one NTFS partition on domain controllers. It is recommended that you use NTFS on all partitions on all the servers in the domain, because any FAT or FAT32 partition lacks many security features. For example, on FAT or FAT32 partitions, a shared folder can be protected only by the permissions set on that shared folder, not on individual files, and there is no software protection against local access to the partition. For more information, see "Reformatting or Converting a Partition to Use NTFS" in "Getting Ready for a New Installation" later in this topic.
If some of your domain controllers run Windows 2000 and some run Windows NT, it is recommended that you upgrade the Windows NT domain controllers as soon as is practical, to reduce the number of version differences between computers, simplify management and troubleshooting, and strengthen security.
Before you begin an upgrade from Windows NT 4.0, you must apply Service Pack 5 or later.
If you have servers or client computers that run Windows NT 3.51, it is recommended that you install or upgrade to a newer operating system on all these computers, or retire them from operation. If you have more than one domain, you must upgrade domain controllers running Windows NT 3.51 for reliable logon validation. In any case, upgrading or retiring computers running Windows NT 3.51 strengthens security and reduces the number of version differences between computers, simplifying management and troubleshooting.

After the domain and forest are prepared, there are two stages in the upgrade of a domain controller. First you run Setup to upgrade the operating system and then, when Setup is complete, you run the Active Directory Installation Wizard (which appears on the screen the first time you log on).

Preparing for Upgrades in a Domain Containing Windows 2000 Domain Controllers

Before beginning the first domain controller upgrade in a domain where one or more domain controllers run Windows 2000, you must first prepare the domain and the forest in which it is located. After preparing the domain, you can begin the domain upgrade through either of the following approaches:

Upgrade an existing domain controller to a product in the Windows Server 2003 family. (Be sure to check hardware compatibility first, as described in "Hardware Compatibility" earlier in this topic.)

If you attempt to upgrade the first domain controller and you have not yet prepared the domain and the forest, Setup stops and a pop-up provides instructions.
Install Windows Server 2003, Standard Edition, on a member server in the domain, and start the Active Directory Installation Wizard to change the member server into a domain controller.

If you install Windows Server 2003, Standard Edition, on a member server in the domain and start the Active Directory Installation Wizard, but you have not yet prepared the domain and the forest, the wizard stops and a pop-up provides instructions.

For additional background information about this type of domain upgrade, see the previous section, "Upgrades in a Domain Containing Windows 2000 Domain Controllers."

To prepare the domain and forest before upgrading a domain containing Windows 2000 domain controllers

Before carrying out this procedure, it is recommended that you apply Service Pack 2 or later to all domain controllers running Windows 2000.
Locate the server that is the schema master and the server that is the infrastructure master (they might be the same server).

In a network of two to five servers (a network within the size of those discussed in this manual), the schema master is usually the server on which you first installed or upgraded to Windows 2000. For information about identifying the schema master or the infrastructure master, search Windows 2000 Help for the topics called "Identify the schema master" and "Identify the infrastructure master." To view Windows 2000 Help, click Start and then click Help. You can also view Windows 2000 Help at: http://go.microsoft.com/fwlink/?LinkID=5197
Back up the schema master.

You can back up other servers as well, as an appropriate part of preparing for upgrading, but be sure to back up the schema master just before taking the next step.
Disconnect the schema master from the network and do not reestablish the connection until step 9 in this procedure.
On the schema master, while Windows 2000 is running, insert the Setup CD for Windows Server 2003, Standard Edition, in the CD-ROM drive.
Open a command prompt, change to the CD-ROM drive, and change to the \i386 folder on the CD.

On the schema master, at the command prompt, type:

adprep /forestprep

Note

To perform this step, you must be a member of the Enterprise Admins group and the Schema Admins group in Active Directory, or you must have been delegated the appropriate authority. As a security best practice, consider using Run as. Run as is a secondary logon method that you can use to start commands or programs using a different security context. For example, you can log on as a member of the Users group and, without logging off, run a command as a member of the Enterprise Admins group. To find more information in Windows 2000 about Run as, on a server running Windows 2000, click Start, click Help, click the Search tab, and then search for runas.

After the preceding command has finished on the schema master, confirm that the preparation of the forest was successful. Steps you can take to do this include:
- Confirm that no error messages have appeared.
- Run any domain controller diagnostic tools that are available, for example, Dcdiag.exe, one of the tools available from the \Support\Tools folder on the Setup CD. Keep in mind that replication errors will appear while the domain controller is disconnected from the network.
- Open Event Viewer and check the system log for errors or unexpected events.
Evaluate the information you gathered in the previous step and choose accordingly:
- If adprep /forestprep ran without errors, reconnect the schema master to the network and continue with the next step of this procedure.
- If adprep /forestprep ran but error messages provided instructions for additional steps to take, follow the instructions and then return to the confirmation process described in the previous step.
- If adprep /forestprep did not run successfully, or diagnostic tools such as Dcdiag.exe revealed a significant problem, restore the schema master from backup and investigate the corrective steps necessary so that adprep /forestprep can be run successfully.
If the schema master is a different computer from the infrastructure master, wait for the changes made by adprep /forestprep to replicate from the schema master to the infrastructure master. Wait at least fifteen minutes, but allow longer (half a day or a day) if the schema and infrastructure masters are not at the same site. If you try to perform the domain preparation on the infrastructure master before the changes have replicated, Adprep will provide an error message to notify you that more time is needed.
On the infrastructure master, while Windows 2000 is running, insert the Setup CD for Windows Server 2003, Standard Edition, in the CD-ROM drive.
Open a command prompt, change to the CD-ROM drive, and change to the \i386 folder on the CD.

On the infrastructure master, type:

adprep /domainprep

Note

To perform this step, you must be a member of the Domain Admins group or the Enterprise Admins group in Active Directory, or you must have been delegated the appropriate authority. As a security best practice, consider using Run as. Run as is described in the previous note in this section.

After the preceding command has finished on the infrastructure master, confirm that the preparation of the domain was successful. Steps you can take to do this include:
- Confirm that no error messages have appeared.
- Run any domain controller diagnostic tools that are available.
- Open Event Viewer and check the system log for errors or unexpected events.
Evaluate the information you gathered in the previous step and choose accordingly:
- If adprep /domainprep ran without errors, continue with the next step of this procedure.
- If adprep /domainprep ran but error messages provided instructions for additional steps to take, follow the instructions and then return to the confirmation process described in the previous step.
Wait for the changes made by adprep /domainprep to replicate from the infrastructure master to other domain controllers. Wait at least fifteen minutes, but allow longer (half a day or a day) if you have a domain controller in a remote site. If you begin to upgrade one of the other domain controllers before the changes have replicated, a pop-up message will notify you that more time is needed.

After you complete this procedure, you can leave your domain controllers running Windows 2000 for an indefinite length of time, or you can begin the domain controller upgrade immediately.

Note

For information about upgrading multiple domains, see the Windows Server 2003 Deployment Kit.

Upgrades in a Windows NT 4.0 Domain

The sections that follow provide information about upgrading servers in a Windows NT 4.0 domain, that is, a domain where all domain controllers run Windows NT 4.0. The guidelines provided are for a network with one to five servers and 100 or fewer clients (the size of network discussed in this manual). For information about upgrading servers in a domain where one or more domain controllers run Windows 2000, see "Upgrades in a Domain Containing Windows 2000 Domain Controllers" earlier in this topic.

The following list describes the sections about upgrading servers in a Windows NT 4.0 domain:

For an overview of the upgrading process, see "Stages in the Process of Upgrading Servers in a Windows NT 4.0 Domain" later in this topic.
For information about initial preparations such as checking hardware compatibility and reviewing the size of partitions on server hard disks, see "Basic Preparations for the Upgrade of Servers in a Windows NT Domain" later in this topic.
For more information about planning steps that help you tailor the upgrade to your specific situation, see the following sections later in this topic:
- "Planning Server Roles and the Order of Server Upgrades from Windows NT"
- "Features Available with the Upgrade of Any Server in a Windows NT Domain"
- "Features Available with the First Domain Controller Upgrade in a Windows NT Domain"
- "Planning DNS for the Upgrade of Domain Controllers Running Windows NT 4.0"
For more information about the final preparations to make just before beginning an upgrade, and guidelines about choices to make during the upgrade of domain controllers, see "Performing Backups and Carrying out Upgrades of Domain Controllers Running Windows NT 4.0" later in this topic.

Other Sources of Information

For additional information, see the following sources:

For conceptual information and procedures for using Active Directory, see the Active Directory topics in Help and Support Center.

To view these topics on a server running Windows Server 2003, after completing Setup, click Start, and then click Help and Support.
For information about deployments and upgrades in multiple domains or in domains larger than five servers, see the Windows Server 2003 Deployment Kit. You can view the Windows Deployment and Resource Kits on the Web at: http://go.microsoft.com/fwlink/?LinkID=29883

Stages in the Process of Upgrading Servers in a Windows NT 4.0 Domain

There are three stages to the process of upgrading servers in a domain where all the domain controllers run Windows NT 4.0:

Basic preparations There are a small number of basic preparations you must carry out before upgrading servers to a product in the Windows Server 2003 family, such as confirming hardware compatibility and reviewing the partitioning of the disks on the servers. For more information, see "Basic Preparations for the Upgrade of Servers in a Windows NT Domain" later in this topic.

Planning server roles and the order in which to upgrade servers Planning server roles and the order in which to upgrade servers involves understanding the server roles available, as well as deciding when to upgrade domain controllers in relation to member servers:

The server roles available for servers running Windows NT are slightly different than for servers running Windows Server 2003. Also, in contrast with servers running Windows NT, you can change the role of a server running Windows Server 2003 without reinstalling the operating system. For more information, see "Planning Server Roles and the Order of Server Upgrades from Windows NT" later in this topic.
If the first Windows NT server you upgrade is a member server, many features become available immediately. For information about reasons for upgrading a particular member server early in the process, see "Features Available with the Upgrade of Any Server in a Windows NT Domain" later in this topic.
When you upgrade the primary domain controller, many Active Directory features become available (in addition to all the other features in the Windows Server 2003 family). For information about how these features can simplify administration, see "Features Available with the First Domain Controller Upgrade in a Windows NT Domain" later in this topic.

However, the upgrade to Active Directory also involves some advance planning. For information about the planning necessary before upgrading to Active Directory, see "Planning DNS for the Upgrade of Domain Controllers Running Windows NT 4.0" later in this topic.

Important

When you begin the upgrade of domain controllers running Windows NT 4.0, you must upgrade the primary domain controller first.

Performing backups and carrying out the upgrade process There are a few preparations to make just before beginning the upgrades, such as ensuring that you have applied Service Pack 5 or later, and performing backups:

For information about the preparations to make just before beginning the upgrade, see "Applying the Service Pack, Performing Backups, and Other Preparations" later in this topic.
For information about choices to make during the upgrade of the primary domain controller, see "Running the Active Directory Installation Wizard After Upgrading the Primary Domain Controller" later in this topic.
For information about Active Directory, see "Learning About Active Directory After the First Server is Upgraded" later in this topic.
For guidelines regarding further upgrades, see "Completing Further Upgrades from Windows NT 4.0" later in this topic.

Basic Preparations for the Upgrade of Servers in a Windows NT Domain

There are a small number of basic preparations you must carry out to lay the groundwork for upgrading servers in a domain where all the domain controllers run Windows NT 4.0. For information about how these preparations fit into the overall process of the upgrade, see "Stages in the Process of Upgrading Servers in a Windows NT 4.0 Domain" earlier in this topic.

Important

If you have servers or client computers that run Windows NT 3.51, it is recommended that you install or upgrade to a newer operating system on all these computers, or retire them from operation. If you have more than one domain, you must upgrade domain controllers running Windows NT 3.51 for reliable logon validation. In any case, upgrading or retiring computers running Windows NT 3.51 strengthens security and reduces the number of version differences between computers, simplifying management and troubleshooting.

Compatibility check Check to see whether the applications and hardware on your systems are compatible with products in the Windows Server 2003 family, and then update or replace them as necessary. For important details about how to do this, see "Hardware Compatibility" and "Important Files to Review" earlier in this topic.

Review partition sizes, especially on domain controllers Review the partitions on the servers to make sure there is plenty of free space available for the Windows Server operating system to run. It is especially important to allow plenty of free space on domain controllers, because the existing user accounts database can expand by as much as a factor of ten when Active Directory functionality is added. For information about the disk requirements for the operating system, see "System Requirements" earlier in this topic.

Review file systems on server partitions You must have at least one NTFS partition on domain controllers. It is recommended that you use NTFS on all partitions on all the servers in the domain, because any FAT or FAT32 partition lacks many security features. For example, on FAT or FAT32 partitions, a shared folder can be protected only by the permissions set on the share, not on individual files, and there is no software protection against local access to the partition. For more information, see "Choosing a File System for the Installation Partition" and "Reformatting or Converting a Partition to Use NTFS" in "Getting Ready for a New Installation" later in this topic.

Service pack Obtain Service Pack 5 or later, because it is necessary for the upgrade.

Mirror sets or other disk sets If you used Windows NT 4.0 to create a volume set, mirror set, stripe set, or stripe set with parity, and you want to run Setup for Windows Server 2003, Standard Edition, on that computer, you must prepare the disk set first. For details, see "Working with Volume, Mirror, or Stripe Sets or Stripe Sets with Parity" later in this topic.

Working with Volume, Mirror, or Stripe Sets or Stripe Sets with Parity

With the disk management technologies in Windows NT 4.0, you could create volume sets, mirror sets, stripe sets, or stripe sets with parity, each with specific capabilities and limitations. By using the dynamic disk technology introduced with Windows 2000, you can use similar technologies, with the added flexibility of being able to extend disk volumes without repartitioning or reformatting.

This transition from the technologies used in Windows NT 4.0 means that you must make certain choices before running Setup for Windows Server 2003, Standard Edition. Any volume sets, mirror sets, stripe sets, or stripe sets with parity that you created with Windows NT 4.0 are not supported in Windows Server 2003, Standard Edition, although they were supported to a limited extent in Windows 2000.

For a mirror set, break the mirror

If you are running Windows NT 4.0 on a computer that has a mirror set and you want to run Setup for Windows Server 2003, Standard Edition, on that computer, first back up the data (as a safeguard — the data will not be erased) and then break the mirror. Ensure that you have applied Service Pack 5 or later, required before running Setup for Windows Server 2003, Standard Edition. Then run Setup.

For a volume set, stripe set, or stripe set with parity, back up the data and delete the set

If you are running Windows NT 4.0 on a computer that has a volume set, stripe set, or stripe set with parity, and you want to run Setup for Windows Server 2003, Standard Edition, on that computer, first back up the data. Next, delete the set (which will delete the data). Ensure that you have applied Service Pack 5 or later, required before running Setup for Windows Server 2003, Standard Edition. Then run Setup.

After running Setup for Windows Server 2003, Standard Edition, you can make the disk dynamic, restore backed-up data as necessary, and make use of the volume options shown in the table later in this section. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

If necessary, use the Ftonline support tool

The preceding methods are the recommended methods for preparing to run Setup for Windows Server 2003, Standard Edition, on a computer that contains a volume set, mirror set, stripe set, or stripe set with parity created with Windows NT 4.0. However, if you do not use these methods and you must access one of these sets after running Setup for Windows Server 2003, Standard Edition, you can use the Ftonline tool. The Ftonline tool is part of the Windows Server 2003 family Support Tools. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Types of Multidisk Volumes on Dynamic Disks

The disk sets described in the previous section have different names in the Windows Server 2003 family than they had in Windows NT 4.0:

Windows NT 4.0 name	Windows Server 2003 family name
Volume set	Spanned volume on a dynamic disk
Mirror set	Mirrored volume on a dynamic disk
Stripe set	Striped volume on a dynamic disk
Stripe set with parity	RAID-5 volume on a dynamic disk

Planning Server Roles and the Order of Server Upgrades from Windows NT

This section provides a list of basic guidelines to use when planning server roles and the order of server upgrades in a domain where all the domain controllers run Windows NT 4.0. For information about how this planning fits into the overall process of the upgrade, see "Stages in the Process of Upgrading Servers in a Windows NT 4.0 Domain" earlier in this topic.

Use the following guidelines when planning server roles and the order of server upgrades from Windows NT:

Familiarize yourself with the slight differences between the naming of server roles for servers running Windows NT as compared to servers running Windows Server 2003:
- In Windows NT, there are two kinds of domain controllers, the primary domain controller, which is limited to one per domain and contains a read-write database, and backup domain controller, which is not limited in number and contains a read-only database.
- For domain controllers running Windows Server 2003, there is only one kind of domain controller, without a "primary" or "backup" designation. All domain controllers contain matching copies of the user accounts and other Active Directory data in a given domain, and provide read-write access to that data.
- In addition to domain controllers, the possible roles for either operating system include member server, which belongs to a domain but does not contain a copy of the Active Directory data, and stand-alone server, which belongs to a workgroup instead of a domain.
Note that in contrast with Windows NT, you can change the role of a server without rerunning Setup. However, as a general practice, it is best to plan the roles ahead of time and change them only as necessary.
As was true with Windows NT, a domain must have at least one domain controller, but it is recommended that a domain have multiple domain controllers for resilience in the handling of logon requests and directory updates.
If the features you need most are not Active Directory features, you can focus on upgrading member servers first. For information about reasons for upgrading a particular member server early in the process, see "Features Available with the Upgrade of Any Server in a Windows NT Domain" later in this topic.

If the features you need most are Active Directory features, focus on upgrading domain controllers first, and carry out the advance planning that is necessary for this part of the upgrade.

Important

When you begin the upgrade of domain controllers running Windows NT 4.0, you must upgrade the primary domain controller first. For information about the features that come with Active Directory, see "Features Available with the First Domain Controller Upgrade in a Windows NT Domain" later in this topic. For information about the planning necessary before upgrading to Active Directory, see "Planning DNS for the Upgrade of Domain Controllers Running Windows NT 4.0" later in this topic.

When you begin upgrading domain controllers, if you have a remote access server that is a member server, it is recommended that you upgrade it before the last domain controller is upgraded. Such a server is dependent on domain controllers for user information, and therefore should not be "left behind" when domain controllers are upgraded. As an alternative, you can weaken the security permissions in Active Directory so that the remote access server running Windows NT can read user attributes from the domain controllers running products in the Windows Server 2003 family. For more information, see the Windows Server 2003 Deployment Kit.

Features Available with the Upgrade of Any Server in a Windows NT Domain

If you are interested in the features in the following list, you might want to focus on upgrading the member servers in your Windows NT 4.0 domain. For information about reasons to upgrade the domain controllers, see "Features Available with the First Domain Controller Upgrade in a Windows NT Domain" later in this topic.

Management tools:
- Microsoft Management Console
- Plug and Play
- Device Manager
- Add Hardware Wizard (in Control Panel)
- Support for universal serial bus
- Enhanced Backup utility
File system support: The version of NTFS in the Windows Server 2003 family includes support for disk quotas, the ability to defragment directory structures, and compressed network I/O.
Application services: Win32 Driver Model, DirectX 7.0, and Windows Script Host.
Printer protocol support: Device and protocol support allowing choices from more than 2,500 different printers. Other printing enhancements are included; for example, Internet Printing Protocol support, which allows users to print directly to a URL over an intranet or the Internet.
Scalability and availability: Improved symmetric multiprocessor support as compared to Windows NT.
Security: Encrypting File System.

Features Available with the First Domain Controller Upgrade in a Windows NT Domain

If you are interested in the features in the following list, all of which are features of Active Directory, focus on upgrading the primary domain controller, as opposed to member servers. (The first domain controller that you upgrade must be the primary domain controller. It is recommended that you also upgrade backup domain controllers.) For information about the DNS planning necessary before beginning the upgrade of domain controllers running Windows NT, see "Planning DNS for the Upgrade of Domain Controllers Running Windows NT 4.0" later in this topic.

For information about reasons to upgrade member servers, see "Features Available with the Upgrade of Any Server in a Windows NT Domain" earlier in this topic.

Note

When a domain controller is upgraded, it has all the features of the Windows Server 2003 family, which means it has all the features in the following list, plus the features in "Features Available with the Upgrade of Any Server in a Windows NT Domain" earlier in this topic.

When you upgrade a domain controller to run the Windows Server 2003 operating system instead of Windows NT, many Active Directory features become available in the domain:

Simplified management of network-resource information and user information.
Group Policy, which you can use to set policies that apply across a given site, domain, or organizational unit in Active Directory.
Security and authentication features, including support for Kerberos V5, Secure Sockets Layer 3.0, and Transport Layer Security using X.509v3 certificates. (For more information, see "What’s New in the Windows Server 2003 Family.")
Directory consolidation, through which you can organize and simplify the management of users, computers, applications, and devices, and make it easier for users to find the information they want. You can take advantage of synchronization support through interfaces based on the Lightweight Directory Access Protocol (LDAP), and work with directory consolidation requirements specific to your applications.
Directory-enabled applications and infrastructure, which make it easier to configure and manage applications and other directory-enabled network components.
Scalability without complexity, a result of Active Directory scaling to millions of objects per domain and using indexing technology and advanced replication techniques to speed performance.
Use of Internet standards, including access through Lightweight Directory Access Protocol and a namespace based on the Domain Name System (DNS).
Active Directory Service Interfaces (ADSI), a powerful development environment.
Additional Active Directory features, as described in the section on Active Directory in "What’s New in the Windows Server 2003 Family."

Planning DNS for the Upgrade of Domain Controllers Running Windows NT 4.0

The Domain Name System (DNS) is a hierarchical naming system used for locating computers on the Internet and private TCP/IP networks. This section describes the planning for DNS that is recommended before you upgrade to Active Directory, that is, before you upgrade the primary domain controller in a Windows NT domain to become a domain controller running Windows Server 2003.

DNS is the name resolution service used by Active Directory. Products in the Windows Server 2003 family can support both the Windows Internet Name Service (WINS) common in Windows NT networks, and DNS. You must deploy DNS to support Active Directory. You can choose to migrate from WINS to DNS or configure your new DNS infrastructure to support your current WINS infrastructure.

For the upgrading of a network with one to five servers and 100 or fewer clients (which is the size of network discussed in this manual), only a simple DNS plan is needed, because there will be no parent or child domains. (For information about upgrading networks with more than five servers, see the Windows Server 2003 Deployment Kit.) DNS names consist of name labels separated by periods. A simple DNS domain name for a Windows Server 2003 family domain could take the form domainname.ext where ext is an established "top-level domain" type such as com or org. With a domain name of that form, the form of a computer name in the domain would be:

computername.domainname.ext

When you upgrade the primary domain controller in a Windows NT 4.0 domain (the primary domain controller must be upgraded first, before backup domain controllers), you will be offered several different options for the handling of DNS. If there is no DNS server available to work with Active Directory, you will be offered, by default, the option of installing DNS on the domain controller you are upgrading (formerly the primary domain controller). In a network with one to five servers, the simplest approach is to accept this default. Then, after completing the upgrade of the server on which DNS was installed, take note of its static IP address, and configure other computers to send any DNS requests to that IP address. For additional basic information about DNS and how to configure it during Setup, see "Name Resolution for TCP/IP" in "Getting Ready for a New Installation" later in this topic. For more detailed information about DNS, see the Windows Resource and Deployment Kits on the Web at: http://go.microsoft.com/fwlink/?LinkID=29883

Another way to view information about DNS is to go to a computer running any product in the Windows Server 2003 family, click Start, and then click Help and Support.

Performing Backups and Carrying out Upgrades of Domain Controllers Running Windows NT 4.0

There are a few preparations to make just before beginning an upgrade from Windows NT 4.0, such as ensuring that you have applied Service Pack 5 or later, and performing backups.

During the upgrade, there are a few choices to make for the upgrade of a network of two to five servers (a network within the size of those discussed in this manual). The following sections describe these final preparations and provide guidelines for making choices.

Before beginning this stage of the upgrade process, be sure to review "Basic Preparations for the Upgrade of Servers in a Windows NT Domain" and "Planning Server Roles and the Order of Server Upgrades from Windows NT" earlier in this topic.

Applying the Service Pack, Performing Backups, and Other Preparations

Before beginning an upgrade of servers in a domain where all the domain controllers run Windows NT 4.0, apply an appropriate Service Pack, perform backups, and consider whether to remove a backup domain controller temporarily from the network:

Be sure that you have applied Service Pack 5 or later to the servers. You must apply Service Pack 5 or later before you can begin an upgrade from Windows NT 4.0.
As with any major change to the contents of hard disks on your servers, it is recommended that you back up each server before upgrading it. Also be sure to review "Preparing Your System for an Upgrade" in "Running Setup for an Upgrade" later in this topic.
If you are preparing to upgrade the primary domain controller, to protect your existing network integrity, consider removing a backup domain controller temporarily from the network. To do this, in your existing Windows NT network, choose a backup domain controller, ensure that it has a current copy of the user accounts database, and back it up. Then disconnect its network cable. After the upgrade of your primary domain controller to become a domain controller running Windows Server 2003 (you must upgrade the primary domain controller first), this disconnected system is available for promotion to a Windows NT primary domain controller if necessary. However, in the course of an uneventful upgrade, you would not and could not promote the Windows NT backup domain controller to primary domain controller. Instead, you would continue the upgrade process, eventually reconnecting the disconnected server and upgrading it.

Running the Active Directory Installation Wizard After Upgrading the Primary Domain Controller

To upgrade the primary domain controller to become a domain controller running Windows Server 2003, you must first run Setup to upgrade the operating system. After the operating system is installed, you can run the Active Directory Installation Wizard (which appears on the screen the first time you log on). This section provides guidelines for making choices in the Active Directory Installation Wizard when upgrading the primary domain controller.

For information about preparations necessary before beginning the upgrade, see "Stages in the Process of Upgrading Servers in a Windows NT 4.0 Domain" earlier in this topic. For information about starting Setup, see "Running Setup for an Upgrade" later in this topic.

Important

When you begin the upgrade of domain controllers running Windows NT 4.0, you must upgrade the primary domain controller first.

The following table shows the choices that are recommended when running the Active Directory Installation Wizard on the server that was formerly the primary domain controller. These recommendations are for a network of two to five servers (a network within the size of those discussed in this manual):

Choices on the server that was formerly the primary domain controller	Recommended action for a network of two to five servers
Domain in a new forest or Child domain in an existing domain tree or Domain tree in an existing forest	Choose Domain in a new forest.
Forest functional level of Windows Server 2003 interim (This choice appears only if Domain in a new forest was previously selected.)	Choose Windows Server 2003 interim. With this domain functional level, you can have domain controllers running Windows NT 4.0 or Windows Server 2003, but no domain controllers running Windows 2000.

During the upgrade, you can choose the location of three important items: the database containing user accounts and other Active Directory data, the log file, and the system volume (Sysvol) folder. The database and the log file can be on any kind of partition (FAT, FAT32, or NTFS); the database can expand by as much as a factor of ten from the size it had with Windows NT, so allow plenty of room for it. (Initially, the log file will take up very little space.) The system volume folder must be on an NTFS partition. (For information about NTFS and other file systems, see "Choosing a File System for the Installation Partition" and "Reformatting or Converting a Partition to Use NTFS" in "Getting Ready for a New Installation" later in this topic.)

Learning About Active Directory After the First Server is Upgraded

As soon as the first server is upgraded, you can open Help and Support and read information about how to use Active Directory. To open Help and Support Center, click Start, and then click Help and Support. The following list provides suggestions about information to review:

Additional information about upgrading from a Windows NT 4.0 domain.
"New ways to do familiar tasks," that is, a table that compares the way you completed domain tasks with Windows NT with the way you complete those tasks with products in the Windows Server 2003 family.
Domains and forests.
Functional levels (domain functional level and forest functional level). After upgrading the last domain controller in a Windows NT domain to become a domain controller running Windows Server 2003, you have the option of changing functional levels, as described in the Active Directory topics.
The global catalog and also operations master roles, which you do not have to take action on as part of the upgrade, but might want to learn about.

For information about deployments and upgrades in multiple domains or in domains larger than five servers, see the Windows Server 2003 Deployment Kit. You can view the Windows Deployment and Resource Kits on the Web at: http://go.microsoft.com/fwlink/?LinkID=29883

Completing Further Upgrades from Windows NT 4.0

After you upgrade your primary domain controller and ensure that it is functioning to your satisfaction, you can begin the upgrade of any backup domain controllers. (You can upgrade member servers at any time.)

Upgrade the backup domain controllers one at a time (ensure that each is backed up before upgrading). Before beginning the upgrade of each backup domain controller, make sure that at least one domain controller running Windows Server 2003 is available on the network, because the user accounts and other Active Directory data will be copied from the domain controllers that have already been upgraded. Start and test each server on the network to ensure that it is functioning to your satisfaction before upgrading another backup domain controller.

An upgraded domain controller will appear as a Windows NT 4.0 primary domain controller to Windows NT 4.0 servers and client computers. However, it is recommended that you complete the upgrade of all servers in the domain relatively quickly (rather than allowing a long delay). This reduces the number of version differences between computers, simplifying management and troubleshooting, and also strengthens security.

The Internet Explorer Enhanced Security Configuration is enabled by default when you upgrade. The security settings in this configuration can help make your computer more secure by limiting its exposure to malicious Web sites. Therefore, with this enhanced level of security, you might find that some Web sites do not display correctly in Internet Explorer when you are browsing Internet and intranet Web sites. Also, you might be prompted to enter your credentials when accessing network resources, such as files in Universal Naming Convention (UNC) shared folders. You can easily change the enhanced security settings.

If you have a remote access server that is a member server, it is recommended that you upgrade it before the last domain controller is upgraded. For more information, see "Planning Server Roles and the Order of Server Upgrades from Windows NT" earlier in this topic.

Top of page

Running Setup for an Upgrade

Preparing Your System for an Upgrade
Starting Setup for an Upgrade
Planning for Unattended Setup
Product Activation for Products in the Windows Server 2003 Family

This topic provides recommendations for preparing your system before running Setup for Microsoft® Windows Server™ 2003, Standard Edition, for an upgrade, and it explains how to run Setup. Before reading this topic or running Setup, be sure to read "Getting Ready for an Upgrade" at the beginning of this document.

For technical information and resources to help you evaluate, deploy, and administer the Windows server operating systems, see the Windows Server TechCenter on the Web at: http://go.microsoft.com/fwlink/?LinkID=28944

Preparing Your System for an Upgrade

This section describes the basic steps to take in preparing your server for an upgrade.

Checking the System Log for Errors

Use Event Viewer to review the system log for recent or recurring errors that could cause problems during the upgrade. For information about viewing errors, see Help for the operating system that you are running.

Backing Up Files

Before upgrading, it is recommended that you back up your current files, including anything containing configuration information, for example, the System State and the system and boot partitions. You can back up files to a variety of different media, such as a tape drive or the hard disk of another computer on the network.

Preparing Mirror Sets and Other Disk Sets for an Upgrade (Windows NT 4.0 only)

With the disk management technologies in Windows NT® 4.0, you could create volume sets, mirror sets, stripe sets, or stripe sets with parity, each with specific capabilities and limitations. By using dynamic disks, introduced with Windows 2000, you can take advantage of similar technologies, and with Windows Server 2003, Standard Edition, you can also extend dynamic volumes without repartitioning or reformatting.

This transition from the technologies used in Windows NT 4.0 means that you must make certain choices before running Setup for Windows Server 2003, Standard Edition. For important information about these choices, see the "Working with Volume, Mirror, or Stripe Sets or Stripe Sets with Parity" section in "Getting Ready for an Upgrade" at the beginning of this document.

Disconnecting UPS Devices

If you have an uninterruptible power supply (UPS) connected to your target computer, disconnect the connecting serial cable before running Setup. Setup automatically attempts to detect devices connected to serial ports, and UPS equipment can cause problems with the detection process.

Reviewing Hardware and Software

When you start Setup for an upgrade, the first process it carries out is a check for compatible hardware and software on your computer. Setup displays a report before continuing. Use this report, along with information in Relnotes.htm (on the Setup CD), to find out whether you need to update your hardware, drivers, or software before upgrading.

Confirm that your hardware and software is designed for products in the Windows Server 2003 family by checking the hardware and software compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37823

If you have hardware devices that do not use Plug and Play, see the "Taking an Inventory of Devices That Do Not Use Plug and Play" section in "Getting Ready for an Upgrade" at the beginning of this document.

Starting Setup for an Upgrade

This section explains how to start Setup for an upgrade on an x86-based computer. You can start Setup for an upgrade from the CD or from a network, as described in "Starting an Upgrade" later in this topic.

For important information about upgrades in the context of a domain, see "Getting Ready for an Upgrade" at the beginning of this document. For information about unattended Setup and other options available when you start Setup, see "Planning for Unattended Setup" later in this topic.

Note

You cannot upgrade from an earlier version of a Windows server operating system to an Itanium-based or x64-based version of a Windows server operating system.

Providing a Mass Storage Driver or a HAL File

If you have a mass storage controller that requires a driver supplied by the manufacturer, or if you have a custom Hardware Abstraction Layer (HAL) file supplied by the manufacturer, provide the appropriate driver file or HAL file during Setup.

Mass Storage Drivers and the Setup Process

If your controller is compatible, but you are aware that the manufacturer has supplied a separate driver file for use with products in the Windows Server 2003 family, obtain the file (on a floppy disk) before you begin Setup. During the early part of Setup, a line at the bottom of the screen prompts you to press F6. Further prompts will guide you in supplying the driver file to Setup so that it can gain access to the mass storage controller.

If you are not sure whether you must obtain a separate driver file from the manufacturer of your mass storage controller, you can try running Setup. If the controller is not supported by the driver files on the Setup CD and therefore requires a driver file that is supplied by the hardware manufacturer, Setup stops and displays a message saying that no disk devices can be found, or it displays an incomplete list of controllers. After you obtain the necessary driver file, restart Setup, and press F6 when you are prompted.

Using a Custom HAL File

If you have a custom Hardware Abstraction Layer (HAL) file supplied by your computer manufacturer, before you begin Setup, locate the floppy disk or other medium containing the file. During the early part of Setup, a line at the bottom of the screen prompts you to press F6: at this time press F5 (not F6). After you press F5, follow the prompts to include your HAL file in the Setup process.

Starting an Upgrade

If you are upgrading to Windows Server 2003, Standard Edition, you can start Setup from the CD or from a network. For information about the versions of Windows from which you can upgrade, see "Operating Systems from Which You Can Upgrade" in "Getting Ready for an Upgrade" at the beginning of this document.

For information about starting Setup for a new installation, see "Running Setup for a New Installation" later in this topic. For information about unattended Setup and other options available when you start Setup, see "Planning for Unattended Setup" later in this topic.

Important

To run Setup for an upgrade, you must be a member of the Administrators group on the local computer. If the computer is joined to a domain, members of the Domain Admins group might be able to perform this procedure. As a security best practice, if you start Setup from a computer running Windows 2000, consider using Run as.

Run as is a secondary logon method that you can use to start commands or programs using a different security context. For example, you can log on as a member of the Users group and, without logging off, run a command as a member of the Administrators group. To find more information in Windows 2000 about Run as, on a server running Windows 2000, click Start, click Help, click the Search tab, and then search for runas.

To start an upgrade from the CD on a computer running Windows

Insert the CD in the drive, and wait for Setup to display a dialog box.
Follow the Setup instructions.

To start an upgrade from a network

On a network server, share the installation files by either inserting the CD and sharing the CD-ROM drive or by copying the files from the I386 folder on the CD to a shared folder.
On the computer on which you want to install Windows Server 2003, Standard Edition, connect to the shared folder or drive that contains the Setup files.
Run Setup.exe.
Follow the Setup instructions.

Planning for Unattended Setup

This section provides general information about unattended Setup. For detailed instructions on running unattended Setup (also called automated installation), see "Automating and Customizing Installations" in the Windows Server 2003 Deployment Kit.

To simplify the process of setting up a product in the Windows Server 2003 family on multiple computers, you can run Setup unattended. To do this, you create and use an answer file, a customized script that answers the Setup questions automatically. Then, you run Winnt32.exe or Winnt.exe with the appropriate options for unattended Setup. Choose the command according to the operating system that is running when you start unattended Setup:

To start unattended Setup on a computer running MS-DOS or Windows 3.x, use Winnt.exe (with the appropriate options).
To start unattended Setup on a computer running Windows 95, Windows 98, Windows Millennium Edition, Windows NT, Windows 2000, Windows XP, or a product in the Windows Server 2003 family, use Winnt32.exe (with the appropriate options). With Windows NT 4.0, before starting unattended Setup, apply Service Pack 5 or later.

To learn about unattended Setup, see the following sources:

For detailed instructions on running unattended Setup (also called automated installation): See "Automating and Customizing Installations" in the Windows Server 2003 Deployment Kit.
To view the command options available for Winnt.exe: On a computer running Windows 3.x or MS-DOS, insert the Setup CD for Windows Server 2003, Standard Edition, in the CD-ROM drive and open the command prompt. Then, change to the CD-ROM drive, change to the I386 directory, and type:

winnt /?
To use an x86-based or x64-based computer to view the command options available for Winnt32.exe: On a computer running Windows 95, Windows 98, Windows Millennium Edition, Windows NT, Windows 2000, Windows XP, or a product in the Windows Server 2003 family, insert the Setup CD for Windows Server 2003, Standard Edition, in the CD-ROM drive, and open the command prompt. Then, change to the CD-ROM drive, change to the I386 directory, and type:

winnt32 /?

Product Activation for Products in the Windows Server 2003 Family

After you install a product in the Windows Server 2003 family, if the product was purchased individually rather than through a volume licensing arrangement, you will have to activate the product unless your hardware manufacturer has preactivated it for you. Product activation is quick, simple, and unobtrusive, and it protects your privacy. It is designed to reduce software piracy (illegal copies of a product). Over time, reduced piracy means that the software industry can invest more in product development, quality, and support. This results in better products and more innovation for customers.

The following details help explain product activation:

Software reminders Until you activate your product, it provides a reminder each time you log on and at common intervals until the end of the activation grace period stated in your End-User License Agreement (30 days is the typical grace period). If your activation grace period passes and you do not activate the product, your computer will continue to function, except that when you log on locally or log on through Remote Desktop for Administration (the new name for the Windows 2000 functionality known as Terminal Services in Remote Administration Mode), you will only be able to use the Activate Windows Wizard.

How the software handles activation Activation is simple and anonymous. When you activate your product, it means that your product key is now associated with the computer (the hardware) it is installed on. After that happens, your product key cannot be used for activation on other computers (unless you are enrolled in a special program that permits additional activations, for example, a program through the Microsoft Developer Network).

Your local operating system does some of the work for creating the association between your product key and the computer it is installed on. It uses your Product ID, which is derived from your product key, and puts it together with a coded number (called a hardware hash) that uniquely represents the hardware components that make up your computer (without representing any personal information or anything about the software). When you activate, you provide this number, through the Internet or by phone, so that it can be confirmed as representing a legal installation. After you activate, if another person has an illegal copy of your software, and that person tries to use your product key to activate the software on another computer, the coded numbers will show that your product key is already associated with a particular set of hardware (your computer). The other person’s activation will not work.

The Installation ID that the local operating system generates, representing the association between your product key and the computer it is installed on, is used only for the purpose of activation.

Methods for activation After your operating system is installed, begin activation by clicking Start and then clicking Activate Windows (you can also click the key icon that appears in the lower right corner of the screen). By following the instructions on the screen, you can activate through the Internet or by phone:

Internet: When you activate through the Internet, your computer transmits coded information that shows that your product key is associated with your computer hardware. Activation is carried out through a secure server. A confirmation ID is passed back to your computer, automatically activating your product. This process normally takes just a few seconds to complete. No personally identifiable information is required to activate your product.
Phone: When you activate by phone, information on the screen guides you through a few simple steps. When you choose the country or region where you are located, a phone number (toll-free, wherever possible) appears on your screen. When you call the number, a customer service representative asks for the Installation ID that is displayed on your screen. The customer service representative enters that number into a secure database, confirms that the number represents a legally installed product, and provides a confirmation ID to you. Then, you type the confirmation ID into the spaces provided on the screen, and activation is complete.

Reactivation (rarely needed) If you overhaul your computer by replacing a substantial number of hardware components (not just a few), the operating system might view your hardware as a completely different computer, not the one on which you activated. In this situation, you can call the telephone number displayed on the telephone activation screen, and, through a quick, simple process, you can reactivate your product.

Product Activation Compared to Product Registration

Product activation is not the same as product registration. As discussed above, product activation is required, and it is completely anonymous. Product registration, in contrast, is completely optional, and it allows you to provide personal information, such as your e-mail address, if you choose. Registration entitles you to receive information about product updates and special offers directly from Microsoft. All registration information provided is stored securely, and no information is ever loaned or sold to other parties.

Top of page

Getting Ready for a New Installation

Upgrades Compared to New Installations
System Requirements and Hardware Compatibility
Important Files to Review
Decisions to Make for a New Installation
Choosing a Licensing Mode
Deciding Whether a Computer Will Contain More Than One Operating System
Choosing a File System for the Installation Partition
Planning Disk Partitions or Volumes for New Installations
Networks: TCP/IP, IP Addresses, and Name Resolution
Deciding Between Workgroups and Domains

This topic can help you plan for the installation of Microsoft® Windows Server™ 2003, Standard Edition, in a network with one to five servers and 100 or fewer clients. The information contained in this topic is designed to help with the upgrade or installation of such a network. For help with planning the upgrade or installation of a larger number of servers, see the Windows Server 2003 Deployment Kit. You can view the Windows Deployment and Resource Kits on the Web at: http://go.microsoft.com/fwlink/?LinkID=29883

For detailed information about how to run Setup, see "Running Setup for a New Installation" later in this topic.

If you want to plan for an upgrade, see "Getting Ready for an Upgrade" at the beginning of this document.

After you make the decisions outlined in this topic, you will be ready to run Setup and, later, to use Manage Your Server, which includes the Configure Your Server Wizard. Together, Setup and Manage Your Server will help you get your servers up and running quickly. To find out more about Manage Your Server, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Upgrades Compared to New Installations

This section compares upgrading to performing a new installation, to help you confirm your decision to perform a new installation. Upgrading is either replacing Microsoft Windows NT® 4.0 (with Service Pack 5 or later) with a product in the Windows Server 2003 family, or replacing Windows 2000 with a product in the Windows Server 2003 family. Installing, in contrast with upgrading, means completely removing the previous operating system, or installing a product in the Windows Server 2003 family on a disk or disk partition with no previous operating system.

The following lists describe items to consider when deciding between an upgrade and a new installation:

Points to consider for an upgrade:

With an upgrade, configuration is simpler, and your existing users, settings, groups, rights, and permissions are retained.
With an upgrade, you do not need to re-install files and applications. As with any major changes to the hard disk, however, it is recommended that you back up the disk before beginning an upgrade.
Before beginning an upgrade, see "Getting Ready for an Upgrade" at the beginning of this document.
If you want to upgrade and then use the same applications as before, be sure to review applications information in Relnotes.htm (on the Setup CD). Also, for the most recent information about compatible applications for products in the Windows Server 2003 family, see the software compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37604

Note

You cannot upgrade from an earlier version of a Windows server operating system to an Itanium-based or x64-based version of a Windows server operating system.

Points to consider for a new installation:

If you reformat your hard disk and then perform a new installation, the efficiency of your disk might improve (compared to not reformatting it). Reformatting also gives you the opportunity to modify the size or number of disk partitions, to make them match your requirements more closely.
If you want to practice careful configuration management, for example, for a server where high availability is important, you might want to perform a new installation on that server instead of an upgrade. This is especially true on servers on which the operating system has been upgraded several times in the past.
It is possible to install Windows Server 2003, Standard Edition, and also allow the computer to sometimes run another operating system. Setting up the computer this way, however, presents complexities because of file system issues. For more information, see "Deciding Whether a Computer Will Contain More Than One Operating System" later in this topic.

Notes

If you want to install Windows Server 2003, Standard Edition, on a computer that previously ran an operating system released before Windows 2000:
Do not install Windows Server 2003, Standard Edition, on a compressed drive unless the drive was compressed with the NTFS file system compression utility. Uncompress a DriveSpace or DoubleSpace volume before running Setup on it.
If you used Windows NT 4.0 to create a volume set, mirror set, stripe set, or stripe set with parity, and you want to run Setup for Windows Server 2003, Standard Edition, on that computer, you must prepare the disk set first. For details, see "Working with Volume, Mirror, or Stripe Sets or Stripe Sets with Parity" later in this topic.

System Requirements and Hardware Compatibility

Before installing your server, make sure that you have chosen hardware that supports Windows Server 2003, Standard Edition.

System Requirements

To ensure adequate performance, make sure that computers on which you install Windows Server 2003, Standard Edition, meet the following requirements:

For an x86-based computer:
- One or more processors with a recommended minimum speed of 550 MHz (minimum supported speed is 133 MHz). A maximum of four processors per computer is supported. Processors from the Intel Pentium/Celeron family, AMD K6/Athlon/Duron family, or compatible processors are recommended.
- 256 megabytes (MB) of RAM recommended minimum (128 MB minimum supported; 4 gigabytes (GB) maximum).
For an x64-based computer:
- One or more processors with a minimum speed of 1.4 GHz. A maximum of four processors per computer is supported.
- 512 MB of RAM minimum.

For computers with more than 4 GB of RAM, be sure to check the hardware compatibility information in the Windows Catalog at:

http://go.microsoft.com/fwlink/?LinkID=37824

A hard disk partition or volume with enough free space to accommodate the setup process. To ensure that you have flexibility in your later use of the operating system, it is recommended that you allow considerably more space than the minimum required for running Setup, which is approximately 1.25 GB to 2 GB for x86-based versions of Windows Server 2003, and 4 GB for x64-based versions of Windows Server 2003. The larger amount of space is required if you are running Setup across a network instead of from a CD-ROM, or if you are installing on a FAT or FAT32 partition (NTFS is the recommended file system).

Note

The setup process requires the free disk space described in the previous paragraph. After Setup is finished, actual hard disk space used for the operating system will be more than the free space required for Setup, because of space needed for the paging file, for any optional components you install, and (on domain controllers) for user accounts and other Active Directory® information. The usual size for the paging file is 1.5 times the size of the RAM. For information about the paging file, optional components, user accounts, and information stored in Active Directory, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

VGA or higher-resolution monitor (Super VGA 800x600 or higher recommended), keyboard, and (optionally) a mouse or other pointing device.

As an alternative, for operation without a monitor or keyboard, you can choose a remote diagnostic and support processor that is designed for products in the Windows Server 2003 family. For details, see the hardware compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37824

For CD-ROM installation:

A CD-ROM or DVD drive.

For network installation:

One or more network adapters and related cables that are designed for products in the Windows Server 2003 family. For details, see the hardware compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37824
A server from which to offer network access for the Setup files.

For additional hardware-based functionality:

Appropriate hardware for the functionality you require. For example, if you require sound, you must have a compatible sound card with speakers or headphones, and if you plan to support network clients, the servers and clients must have appropriate network adapters and cables. For details about your hardware, see the hardware compatibility information in the Windows Catalog at: http://go.microsoft.com/fwlink/?LinkID=37824

Hardware Compatibility

One of the most important steps to take before installing a server is to confirm that your hardware is compatible with products in the Windows Server 2003 family. You can do this by running a preinstallation compatibility check from the Setup CD or by checking the hardware compatibility information at the Windows Catalog Web site. Also, as part of confirming hardware compatibility, check to see that you have obtained updated hardware device drivers and an updated system BIOS.

Regardless of whether you run a preinstallation compatibility check, Setup checks hardware and software compatibility at the beginning of an installation and displays a report if there are incompatibilities.

Running a preinstallation compatibility check from the Setup CD

You can run a hardware and software compatibility check from the Setup CD. The compatibility check does not require you to actually begin an installation. To run the check, insert the Setup CD in the CD-ROM drive and, when a display appears, follow the prompts for checking system compatibility. You will be offered the option to download the latest Setup files (through Dynamic Update) when you run the check. If you have Internet connectivity, it is recommended that you allow the download.

For more information about downloading the latest Setup files, see "Using Dynamic Update for Updated Drivers and Other Setup Files" later in this topic.

Another way to run the compatibility check is to insert the Setup CD in the CD-ROM drive, open a command prompt, and type:

d:\i386\winnt32 /checkupgradeonly

where d represents the CD-ROM drive.

Checking hardware and software compatibility information on the Microsoft Web site

Note that one type of hardware no longer supported is the microchannel bus.

Checking drivers and system BIOS

Taking an Inventory of Devices That Do Not Use Plug and Play

To take an inventory of your devices, if your computer has an existing operating system, use it to obtain the current settings, such as memory address and interrupt request (IRQ), used with your devices. For example, with Windows NT® 4.0, you can use Control Panel to view settings (on the Start menu, point to Settings, click Control Panel, and then double-click icons such as Network and Ports). You might also choose to view system BIOS information. To do this, watch the screen while starting the computer, and then press the appropriate key when prompted.

At the beginning of an installation, the Setup program automatically takes a device inventory as well. For devices that do not use Plug and Play, or that are not implemented exactly to Plug and Play standards, taking your own inventory helps prevent the following difficulties:

If two or more adapters share IRQ settings or memory addresses, the Setup program might not be able to resolve the conflict. To prevent this, you can take one of two approaches.

You can remove one of the adapters before running Setup and re-install it afterward. For information about installing and configuring adapters and other hardware devices, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

As an alternative, you can modify one adapter’s IRQ settings and memory addresses before running Setup, so that each adapter’s settings are unique.
If adapters do not respond in a standard way to the attempts by Setup to detect or enumerate them, Setup might receive indecipherable or inaccurate information. In this case, you might need to remove these devices before running Setup, and re-install and configure them afterward. For information about installing and configuring adapters and other hardware devices, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

The following table lists the kinds of information to gather if you have devices that do not use Plug and Play, and you decide to take a device inventory before starting Setup.

Adapter	Information to gather
Video	Adapter or chip set type and how many video adapters
Network	IRQ, I/O address, Direct Memory Access (DMA) if used, connector type (for example, BNC or twisted pair), and bus type
SCSI controller	Adapter model or chip set, IRQ, and bus type
Mouse	Mouse type and port (COM1, COM2, or PS/2) or USB
I/O port	IRQ, I/O address, and DMA (if used) for each I/O port
Sound adapter	IRQ, I/O address, and DMA
Universal serial bus (USB)	Which devices and hubs are attached
PC card	Which adapters are inserted and in which slots
Plug and Play	Whether enabled or disabled in BIOS
BIOS settings	BIOS revision and date
External modem	COM port connections (COM1, COM2, and so on)
Internal modem	COM port connections; for nonstandard configurations, IRQ and I/0 address
Advanced Configuration and Power Interface (ACPI); Power Options	Enabled or disabled; current setting
PCI	Which PCI adapters are inserted and in which slots

Mass Storage Drivers and the Setup Process

Using a Custom Hardware Abstraction Layer (HAL) File

Understanding ACPI BIOS

The basic input/output system (BIOS) is a set of software through which the operating system (or Setup) communicates with the computer’s hardware devices. The Advanced Configuration and Power Interface (ACPI) is the current standard for the way the BIOS works. Products in the Windows Server 2003 family support not only ACPI-compliant BIOS versions, but also some BIOS versions based on older advanced power management (APM) and Plug and Play designs.

Some ACPI-based BIOS versions are not compliant with the standard. A more recent version of an ACPI BIOS is more likely to be compliant. An ACPI-based BIOS that is not compliant with the ACPI standard might not support workable communication between the operating system (or Setup) and your hardware. If workable communication is not supported, Setup stops and displays instructions for contacting your hardware manufacturer and taking other steps to solve the problem. If this happens, follow the instructions provided.

To learn more about the ACPI compliance of your BIOS:

For information about your BIOS version, before running Setup, restart the computer and watch the text on the screen. Pay particular attention to blocks of text containing the words "BIOS" or "ACPI BIOS."

For information about BIOS versions for your hardware, check your hardware documentation and contact your hardware manufacturer.

Using Dynamic Update for Updated Drivers and Other Setup Files

Dynamic Update has been carefully designed so that it is reliable and easy to use:

The files on the Dynamic Update section of the Windows Update Web site have been carefully tested and selected. Only files that are important in ensuring that Setup runs well are made available through Dynamic Update. Files with minor updates that will not significantly affect Setup are not part of Dynamic Update.
Because Dynamic Update downloads only the files that are required for your computer, the Dynamic Update software briefly examines your computer hardware. No personal information is collected, and no information is saved. The only purpose is to select appropriate drivers for your hardware configuration. This keeps the download as short as possible and ensures that only necessary drivers are downloaded to your hard disk.
You can use Dynamic Update when running a preinstallation compatibility check from the product CD, or when running Setup itself. Either way, you obtain the most up-to-date files for running Setup. For information about running the compatibility check, see "Hardware Compatibility" earlier in this topic.
You can use Dynamic Update with unattended Setup. Preparing for this requires several steps. For a brief description of unattended Setup, see the "Planning for Unattended Setup" section in, "Running Setup for a New Installation" later in this topic. For details about how to use Dynamic Update with unattended Setup (also called automated installation), see "Automating and Customizing Installations" in the Windows Server 2003 Deployment Kit.

The Windows Update Web site offers a variety of updates that you can use after completing Setup. To learn more, go to: http://go.microsoft.com/fwlink/?LinkID=284

Important Files to Review

At some point in your planning process, before you run Setup, familiarize yourself with the Relnotes.htm file found on the Setup CD.

Decisions to Make for a New Installation

This list outlines the basic decisions to make for a new installation.

Note

On a computer with an x64 processor, you can install both 32-bit and 64-bit Windows operating systems.

What licensing mode to use

With products in the Windows Server 2003 family, you can choose between two licensing modes:

Per Device or Per User. This mode requires a separate Client Access License (CAL) for each device or user that accesses a server running a product in the Windows Server 2003 family.
Per Server. This mode requires a separate CAL for each concurrent connection to a server. For more information about licensing, see "Choosing a Licensing Mode" later in this topic.

Whether you want to be able to choose between different operating systems each time you start the computer

You can set up a computer so that each time you restart it, you can choose from several different operating systems. For more information, see "Deciding Whether a Computer Will Contain More Than One Operating System" later in this topic.

What file system to use on the installation partition

You can potentially choose among three file systems for an installation partition: NTFS, FAT, and FAT32. NTFS is strongly recommended in most situations. It is the only file system that supports Active Directory®, which includes many important features such as domains and domain-based security. However, it might be necessary to have a FAT or FAT32 partition on a basic disk, if you must set up the computer so that it sometimes runs Windows Server 2003, Standard Edition, and sometimes runs Windows NT 4.0 or an earlier operating system. For more information, see "Choosing a File System for the Installation Partition" later in this topic.

What partition or volume you plan to install the operating system on

If you are performing a new installation, review your disk partitions or volumes before you run Setup (for an upgrade, you will use existing partitions or volumes). Both partitions and volumes divide a disk into one or more areas that can be formatted for use by one file system. Different partitions and volumes often have different drive letters (for example, C: and D:). After you run Setup, you can make adjustments to the disk configuration, as long as you do not reformat or change the partition or volume that contains the operating system. For information about planning the partitions or volumes for a new installation, see "Planning Disk Partitions or Volumes for New Installations" later in this topic.

How to handle IP addresses and TCP/IP name resolution

With TCP/IP (the protocol used on the Internet), you need to make decisions about how to handle IP addressing and name resolution (the translating of IP addresses into names that users recognize). For more information, see "Networks: TCP/IP, IP Addresses, and Name Resolution" later in this topic.

Whether to create domains or workgroups for your servers

A domain is a group of accounts and network resources that share a common directory database and set of security policies, and might have security relationships with other domains. A workgroup is a more basic grouping, intended only to help users find objects such as printers and shared folders within that group. Domains make it easier for an administrator to control access to resources and keep track of users. For more information, see "Deciding Between Workgroups and Domains" later in this topic.

Choosing a Licensing Mode

Products in the Windows Server 2003 family support two licensing modes:

Per Device or Per User
Per Server

If you choose the Per Device or Per User mode, each device or user that accesses a server running a product in the Windows Server 2003 family requires a separate Client Access License (CAL). With one CAL, a particular device or user can connect to any number of servers running products in the Windows Server 2003 family. This is the most commonly used licensing method for companies with more than one server running products in the Windows Server 2003 family.

In contrast, Per Server licensing means that each concurrent connection to this server requires a separate CAL. In other words, this server can support a fixed number of connections at any one time. For example, if you select the Per Server client-licensing mode with five licenses, this server could have five concurrent connections at any one time (if each client requires one connection, this is five clients at any one time). The clients using the connections do not need any additional licenses.

The Per Server licensing mode is often preferred by small companies with only one server. It is also useful for Internet or remote access servers where the client computers might not be licensed as network clients for products in the Windows Server 2003 family. You can specify a maximum number of concurrent server connections and reject any additional logon requests.

If you are unsure which mode to use, choose Per Server, because you can change once from Per Server mode to Per Device or Per User mode at no cost. After you choose Per Server and complete Setup, you can display topics about licensing modes in Help and Support Center (click Start, and then click Help and Support). If you use Terminal Server, be sure to look for topics about Terminal Server Licensing.

Deciding Whether a Computer Will Contain More Than One Operating System

On a computer with an appropriate disk configuration (outlined in the table later in this section), you can install more than one operating system, and then choose between the operating systems each time you restart the computer. For example, you could set up a server to run Windows Server 2003, Standard Edition, most of the time, but allow it to sometimes run Windows NT Server 4.0 in order to support an older application. (However, to do this you would need to make specific file system choices and would need Service Pack 5 or later, as described in "Multiple Operating Systems and File System Compatibility" and "Choosing a File System for the Installation Partition" later in this topic.) During restarts, a display would appear for a specified number of seconds, allowing you to select between the two operating systems. (You can specify a default operating system that will run if no selection is made during the restart process.)

The following table shows the disk configurations on which you can install more than one operating system.

Important

You must follow the requirements in the following table. For example, on a basic disk, you must install each operating system, including Windows Server 2003, Standard Edition, in a separate partition. This ensures that each operating system does not overwrite crucial files that are needed by another operating system.

Disk configuration

Requirements for multiple operating systems

Basic disk or disks

You can install multiple operating systems, including Windows NT 4.0 and earlier operating systems, on a basic disk. Each operating system must be on a separate partition or logical drive on the disk. A partition or logical drive is a section of the disk that functions as a separate unit. Different partitions often have different drive letters, for example, C: and D:.

Single dynamic disk

You can install only one operating system.

However, if you used Windows 2000 or Windows XP to change a disk with no partitions directly to a dynamic disk, you must revert the disk to basic before you can install an operating system on it. For more information, see the paragraph that follows this table.

Multiple dynamic disks

Each dynamic disk can contain one installation of Windows 2000, Windows XP, or a product in the Windows Server 2003 family. No other operating systems can start from a dynamic disk.

If you used Windows 2000 or Windows XP to change a disk with no partitions directly to a dynamic disk, you must revert the disk to basic before you can install an operating system on it. All data will be lost in this process, so back it up first. You can then use Windows 2000 or Windows XP to revert the disk, or you can use Setup for Windows Server 2003, Standard Edition. To use Windows 2000 or Windows XP, follow the instructions in Help for your operating system. To use Setup for Windows Server 2003, Standard Edition, during the partitioning phase, find the dynamic disk among the list of available partitions and then delete it (which erases all data on all volumes). You will be prompted to confirm your action. After you do this, the disk will contain only unpartitioned space, and you can use Setup to create a new (basic disk) partition on it.

Reasons to Install Only One Operating System

Setting up a computer so that you can choose between two or more operating systems at startup does have an advantage: it allows you to use applications that run only with a particular operating system. However, there are definite reasons to install only one operating system:

Each operating system uses valuable disk space.
Compatibility issues, especially file system compatibility, can be complex. For more information, see "Multiple Operating Systems and File System Compatibility" later in this topic.
On a dynamic disk (a storage type available with products in the Windows Server 2003 family), you can have only one operating system per disk. Dynamic disks also will not work with some operating systems. For more information, see the table in the previous section, "Deciding Whether a Computer Will Contain More Than One Operating System."
It is no longer necessary to maintain multiple operating systems as a safeguard against problems with starting the computer. With products in the Windows Server 2003 family, you have other options for system recovery. For example, if you have a problem with a newly installed device driver, you can use Safe Mode, in which Windows Server 2003, Standard Edition, restarts with default settings and the minimum number of drivers.

Requirements for Setting Up a Computer with Multiple Operating Systems

Before you decide to set up a computer with more than one operating system, review the following restrictions.

On computers that contain MS-DOS and Windows Server 2003, Standard Edition:

Install each operating system on a different partition, and install the applications used with an operating system on the same partition with it. If an application is used with two different operating systems, install it on two partitions.
MS-DOS must be installed on a basic disk on a partition formatted with FAT. If MS-DOS is not installed on the system partition, which is almost always the first partition on the disk, the system partition must also be formatted with FAT.
Windows Server 2003, Standard Edition, must be installed last. Otherwise important files that are needed for starting Windows Server 2003, Standard Edition, could be overwritten.
File system compatibility might be an issue. See "Multiple Operating Systems and File System Compatibility" later in this topic.

On computers that contain Windows 98 or Windows Millennium Edition and Windows Server 2003, Standard Edition:

Install each operating system on a different partition, and install the applications used with an operating system on the same partition with it. If an application is used with two different operating systems, install it on two partitions.
Windows 98 or Windows Millennium Edition must be installed on a basic disk on a partition formatted with FAT or FAT32. If either Windows 98 or Windows Millennium Edition is not installed on the system partition, which is almost always the first partition on the disk, the system partition must also be formatted with FAT or FAT32.
Windows Server 2003, Standard Edition, must be installed last. Otherwise important files that are needed for starting Windows Server 2003, Standard Edition, could be overwritten.
File system compatibility might be an issue. See "Multiple Operating Systems and File System Compatibility" later in this topic.

On computers that contain Windows NT 4.0 and Windows Server 2003, Standard Edition:

See "Multiple Operating Systems and File System Compatibility" and "Computers That Contain Windows NT 4.0 and Windows Server 2003, Standard Edition," later in this topic.

On computers that contain some combination of Windows Server 2003, Standard Edition, with Windows 2000 or Windows XP, or that contain multiple partitions with products in the Windows Server 2003 family:

Install each operating system on a different partition or, for dynamic disks, on a different disk, and install the applications used with an operating system on the same disk or partition with it. If an application is used with two different operating systems, install it in two places.
Choose any product in the Windows Server 2003 family for installation on a specific partition. For example, you could install Windows Server 2003, Standard Edition, in one location and Windows Server 2003, Enterprise Edition, in another.
If Windows 2000 and Windows Server 2003, Standard Edition, are installed, Windows Server 2003, Standard Edition, must be installed last. Otherwise important files that are needed for starting Windows Server 2003, Standard Edition, could be overwritten.
If the computer participates in a domain, use a different computer name for each installation. Because a unique security identifier (SID) is used for each installation on a domain, the computer name for each installation must be unique, even for multiple installations on the same computer.
If you want to use the Encrypting File System (EFS), you must take certain steps to ensure that encrypted files will be available from each of the installations. For more information, see "Multiple Operating Systems and the Encrypting File System" later in this topic.

Multiple Operating Systems and File System Compatibility

On computers that contain multiple operating systems, compatibility becomes more complex when you consider file system choices. The file systems to choose from are NTFS, FAT, and FAT32. (For more information, see "Choosing a File System for the Installation Partition" later in this topic.)

NTFS is normally the recommended file system because it is more efficient and reliable, and supports important features including Active Directory and domain-based security. With NTFS, however, you need to take file system compatibility into account when considering whether to set up a computer to contain more than one operating system, because with Windows 2000 and the Windows Server 2003 family, NTFS has new features in addition to those in Windows NT. Files that use any new features will be completely usable or readable only when the computer is started with Windows 2000 or a product in the Windows Server 2003 family. For example, a file that uses the new encryption feature will not be readable when the computer is started with Windows NT Server 4.0, which was released before the encryption feature existed. (For more information about features that affect file accessibility with products in the Windows Server 2003 family, see "NTFS" later in this topic.)

Important

If you want to set up a computer with both Windows NT and Windows Server 2003, Standard Edition, and you want to have an NTFS partition, the only appropriate version of Windows NT is version 4.0 with the latest released Service Pack. Using the latest Service Pack maximizes compatibility between Windows NT 4.0 and Windows Server 2003, Standard Edition. (Specifically, you must have Service Pack 5 or later.) Even the latest Service Pack, however, does not provide access to files using the new features in NTFS.

Using NTFS as the only file system on a computer that contains both Windows Server 2003, Standard Edition, and Windows NT is not recommended. On these computers, a FAT partition containing the Windows NT 4.0 operating system ensures that when started with Windows NT 4.0, the computer will have access to needed files. In addition, if Windows NT is not installed on the system partition, which is almost always the first partition on the disk, it is recommended that the system partition also be formatted with FAT.

If you set up a computer so that it starts with Windows NT 3.51 or earlier on a FAT partition, and Windows Server 2003, Standard Edition, on an NTFS partition, when that computer starts with Windows NT 3.51, the NTFS partition will not be visible. If you set up a computer this way, and the partition containing Windows NT 3.51 is not the system partition (which is almost always the first partition on the disk), the system partition must also be formatted with FAT.

Computers That Contain Windows NT 4.0 and Windows Server 2003, Standard Edition

If you plan to set up a computer so that it contains Windows NT 4.0 and Windows Server 2003, Standard Edition, first review the following precautions:

It is no longer necessary to maintain multiple operating systems as a safeguard against problems with starting the computer. With products in the Windows Server 2003 family, you have other options for system recovery. For example, if you have a problem with a newly installed device driver, you can use Safe Mode, in which Windows Server 2003, Standard Edition, restarts with default settings and the minimum number of drivers.
Using NTFS as the only file system on a computer that contains both Windows Server 2003, Standard Edition, and Windows NT is not recommended. Follow the guidelines in "Multiple Operating Systems and File System Compatibility" earlier in this topic.
Make sure that Windows NT 4.0 has been updated with the latest released Service Pack. For details, see "Multiple Operating Systems and File System Compatibility" earlier in this topic.
Install each operating system on a different partition, and install the applications used with an operating system on the same partition with it. If an application is used with two different operating systems, install it on two partitions.
Do not install Windows Server 2003, Standard Edition, on a compressed drive unless the drive was compressed with the NTFS file system compression feature.
Windows Server 2003, Standard Edition, must be installed last. Otherwise important files that are needed for starting Windows Server 2003, Standard Edition, could be overwritten.
If the computer participates in a domain, use a different computer name for each installation.

Multiple Operating Systems and the Encrypting File System

If you set up a server so that it contains some combination of Windows Server 2003, Standard Edition, with Windows 2000 or Windows XP, or contains multiple partitions with products in the Windows Server 2003 family, and you want to use the Encrypting File System (EFS) on the computer, you must take certain steps. These steps make encrypted files readable between the different installations.

One approach is to ensure that all the installations are in the same domain and that the user of these installations has a roaming profile.
Another approach is to export the user’s file encryption certificate and associated private key from one installation and import it into the other installations.

For more information about EFS, roaming user profiles, and importing and exporting certificates, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Choosing a File System for the Installation Partition

You can choose among three file systems for an installation partition: NTFS, FAT, and FAT32. NTFS is strongly recommended in most situations.

Note

You can use important features such as Active Directory and domain-based security only by choosing NTFS as your file system.

The following table lists a number of installation scenarios (the last two scenarios are fairly uncommon) and provides file system guidelines for each one:

Installation or upgrade scenario	File system to use and additional information to read
The computer currently uses NTFS only (no FAT or FAT32).	Continue to use NTFS. No additional information about file systems needed.
The computer has one or more FAT or FAT32 partitions.ANDThe computer contains only one operating system, or the operating systems on the computer include Windows 2000, Windows XP, or products in the Windows Server 2003 family, but no other operating systems.	Consider reformatting or converting partitions so all partitions use NTFS. For more information, see "Reformatting or Converting a Partition to Use NTFS" later in this topic.
The computer will contain multiple operating systems, one of which is MS-DOS, Windows 95, Windows 98, or Windows Millennium Edition	For any partition that must be accessible from MS-DOS, Windows 95, Windows 98, or Windows Millennium Edition, use FAT (or when appropriate, FAT32). For more information, see "Requirements for Setting Up a Computer with Multiple Operating Systems" earlier in this topic.
The computer will contain multiple operating systems, one of which is Windows NT.	Read "Multiple Operating Systems and File System Compatibility" earlier in this topic.

The sections that follow provide information about reformatting or converting a FAT or FAT32 partition to use NTFS, as well as additional background information about NTFS, FAT, and FAT32.

Reformatting or Converting a Partition to Use NTFS

If you have a FAT or FAT32 partition on which you want to install a product in the Windows Server 2003 family, and you want to use NTFS instead, you have two choices:

You can convert the FAT or FAT32 partition to NTFS. This leaves files intact, although the partition may have somewhat more fragmentation and slower performance than a partition formatted with NTFS. However, it is still advantageous to use NTFS, regardless of whether the partition was formatted with NTFS or converted.

If you install a product in the Windows Server 2003 family on a FAT or FAT32 partition, you are offered the option to convert the partition to NTFS. You can also convert a FAT or FAT32 partition after Setup by using Convert.exe. For more information about Convert.exe, after completing Setup, click Start, click Run, type cmd and then press ENTER. In the command window, type help convert, and then press ENTER.
You can reformat the partition with NTFS. This erases all files on the partition, but results in less fragmentation and better performance than with a converted partition.

If you format a partition during Setup, the file systems choices are listed as NTFS and FAT. The following table provides information about the relationship between partition size and file system choices during Setup.

State and size of partition	Setup choices and responses (when formatting the partition)
Unformatted,less than 2 GB.	Setup offers NTFS or FAT.Setup uses the format chosen.
Unformatted, 2 GB or larger, up to a maximum of 32 GB.	Setup offers NTFS or FAT.If FAT is chosen, Setup uses FAT32.
Unformatted, larger than 32 GB.	Setup allows only NTFS.
Previously formatted with FAT32 and larger than 32 GB. (Partition created with Windows 95, Windows 98, or Windows Millennium Edition.)	No formatting needed, even though an unformatted partition of this size, when formatted during or after Setup for a product in the Windows Server 2003 family, would have to use NTFS. In other words, the Windows Server 2003 family continues to support previously-formatted FAT32 partitions of this size.

If you format a partition during Setup, you can choose between a quick format and a full format:

Quick format Quick format creates the file system structure on the disk without verifying the integrity of every sector. Choose this method for any disk that has no bad sectors and no history of file-corruption problems that might be related to bad sectors.

Full format A full format identifies and tracks bad sectors so that they are not used for storing data. Choose this method for any disk that has bad sectors or has a history of file-corruption problems that might be related to bad sectors.

NTFS Compared to FAT and FAT32

NTFS has always been a more powerful file system than FAT and FAT32. Windows 2000, Windows XP, and the Windows Server 2003 family include a new version of NTFS, with support for a variety of features including Active Directory, which is needed for domains, user accounts, and other important security features. For more details about features in NTFS, see "NTFS" later in this topic.

FAT and FAT32 are similar to each other, except that FAT32 is designed for larger disks than FAT. The file system that works most easily with large disks is NTFS.

The following table describes the compatibility of each file system with various operating systems.

Note

File system choices have no effect on access to files across the network. For example, using NTFS on all partitions on a server does not affect clients connecting across a network to shared folders or shared files on that server, even if those clients run an earlier operating system such as Windows 98 or Windows NT.

NTFS	FAT	FAT32
A computer running Windows 2000, Windows XP, or a product in the Windows Server 2003 family can access files on a local NTFS partition. A computer running Windows NT 4.0 with Service Pack 5 or later might be able to access some files. Other operating systems allow no local access.	Access to files on a local partition is available through MS-DOS, all versions of Windows, and OS/2.	Access to files on a local partition is available only through Windows 95 OSR2, Windows 98, Windows Millennium Edition, Windows 2000, Windows XP, and products in the Windows Server 2003 family.

The following table compares disk and file sizes possible with each file system.

NTFS	FAT	FAT32
Recommended minimum volume size is approximately 10 MB. Maximum volume and partition sizes start at 2 terabytes (TB) and range upward. For example, a dynamic disk formatted with a standard allocation unit size (4 KB) can have partitions of 16 TB minus 4 KB. Cannot be used on floppy disks.	Volumes from floppy disk size up to 4 GB. Does not support domains.	Volumes from 33 MB to 2 TB can be written to or read using products in the Windows Server 2003 family. Volumes up to 32 GB can be formatted as FAT32 using products in the Windows Server 2003 family. Does not support domains.
Maximum file size is potentially 16 TB minus 64 KB, although files cannot be larger than the volume or partition they are located on.	Maximum file size is 2 GB.	Maximum file size is 4 GB.

NTFS

This section provides background information about the features available with NTFS. Some of these features include:

Better scalability to large drives. The maximum partition or volume size for NTFS is much greater than that for FAT, and as volume or partition sizes increase, performance with NTFS does not degrade as it does with FAT.
Active Directory (and domains, which are part of Active Directory). With Active Directory, you can view and control network resources easily. With domains, you can fine-tune security options while keeping administration simple. Domain controllers and Active Directory require NTFS.
Compression features, including the ability to compress or uncompress a drive, a folder, or a specific file. (However, a file cannot be both compressed and encrypted at the same time.)
File encryption, which greatly enhances security. (However, a file cannot be both compressed and encrypted at the same time.)
Permissions that can be set on individual files rather than just folders.
Remote Storage, which provides an extension to your disk space by making removable media such as tapes more accessible. (This feature is not included with Windows Server 2003, Web Edition, or with Windows Server 2003, Standard Edition.)
Recovery logging of disk activities, which allows NTFS to restore information quickly in the event of power failure or other system problems.
Sparse files. These are very large files created by applications in such a way that only limited disk space is needed. That is, NTFS allocates disk space only to the portions of a file that are written to.
Disk quotas, which you can use to monitor and control the amount of disk space used by individual users.

For information about converting or reformatting a FAT or FAT32 partition, see "Reformatting or Converting a Partition to Use NTFS" earlier in this topic.

Planning Disk Partitions or Volumes for New Installations

You must plan your disk partitions before you run Setup only if both of the following conditions are true:

You are performing a new installation, not an upgrade.
The disk on which you are installing is a basic disk, not a dynamic disk. Basic disks are the disk type that existed before Windows 2000; most disks are basic disks. Dynamic disks are disks that once were basic but were changed to dynamic using Windows 2000, Windows XP, or a product in the Windows Server 2003 family. If you plan to install to a dynamic disk, you cannot change the volume or partition sizes on the disk during Setup, and therefore no planning is needed regarding partition sizes. Instead, review the guidelines in "Working with Dynamic Disks and Setup" later in this topic.

Disk partitioning is a way of dividing your physical disk so that each section functions as a separate unit. When you create partitions on a basic disk, you divide the disk into one or more areas that can be formatted for use by a file system, such as FAT or NTFS. Different partitions often have different drive letters (for example, C: and D:). A basic disk can have up to four primary partitions, or three primary partitions and one extended partition. (An extended partition can be subdivided into logical drives, while a primary partition cannot be subdivided.)

Important

If you plan to delete or create partitions on a hard disk, be sure to back up the disk contents beforehand, because these actions will destroy any existing data. As with any major change to disk contents, it is recommended that you back up the entire contents of the hard disk before working with partitions, even if you plan to leave one or more of your partitions alone.

Do not install Windows Server 2003, Standard Edition, on a compressed drive unless the partition was compressed with the NTFS file system compression feature. (Uncompress a DriveSpace or DoubleSpace drive before running Setup on it.)

Before you run Setup to perform a new installation, determine the size of the partition on which to install. There is no set formula for figuring a partition size. The basic principle is to allow plenty of room for the operating system, applications, and other files that you plan to put on the installation partition. The files for setting up Windows Server 2003, Standard Edition, require approximately 1.25 GB to 2 GB, as described in "System Requirements" earlier in this topic. It is recommended that you allow considerably more disk space than the minimum amount. It is not unreasonable to allow 4 GB to 10 GB on the partition, or more for large installations. This allows space for a variety of items, including optional components, user accounts, Active Directory information, logs, future service packs, the paging file used by the operating system, and other items.

When you perform a new installation, you can specify the partition on which to install. If you specify a partition on which another operating system exists, you will be prompted to confirm your choice.

During Setup, create and size only the partition on which you want to install Windows Server 2003, Standard Edition. After installation is complete, you can use Disk Management to manage new and existing disks and volumes. This includes creating new partitions from unpartitioned space; deleting, renaming, and reformatting existing partitions; adding and removing hard disks; and changing a basic disk to the dynamic disk storage type, or dynamic to basic. (If you want to have a dynamic disk on a computer that contains more than one operating system, be sure to read "Deciding Whether a Computer Will Contain More Than One Operating System" earlier in this topic.)

Important

If you are setting up a computer so that it contains multiple operating systems, you must install Windows Server 2003, Standard Edition, on its own partition or logical drive. This ensures that Windows Server 2003, Standard Edition, will not overwrite crucial files that are needed by the other operating system. For more information, see "Deciding Whether a Computer Will Contain More Than One Operating System" earlier in this topic.

Disk Partition Requirements for Remote Installation Services

If you plan to use Remote Installation Services on this server so that you can install operating systems onto other computers, a separate partition for use by Remote Installation Services is necessary. Plan on using NTFS on this partition: NTFS is required for the Single Instance Store feature of Remote Installation Services.

If you need to create a new partition for Remote Installation Services, plan on doing it after Setup, and leave enough unpartitioned disk space so that you can create it (at least 4 GB of space is recommended). As an alternative, you can plan to make the disk a dynamic disk, which allows more flexibility in the use of the disk space than a basic disk. (However, if you want to have a dynamic disk on a computer that contains more than one operating system, see "Deciding Whether a Computer Will Contain More Than One Operating System" earlier in this topic.)

For more information about Remote Installation Services and about disk and partition choices, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Options When Partitioning a Disk

You can change the partitions on your disk during Setup only if you are performing a new installation, not an upgrade. You can modify the partitioning of the disk after Setup by using Disk Management.

If you are performing a new installation, Setup examines the hard disk to determine its existing configuration, and then offers the following options:

If the hard disk is unpartitioned, you can create and size the partition on which you will install a product in the Windows Server 2003 family.
If the hard disk is partitioned but has enough unpartitioned disk space, you can create the partition for your Windows Server 2003 family product by using the unpartitioned space.
If the hard disk has an existing partition that is large enough, you can install a product in the Windows Server 2003 family on that partition, with or without reformatting the partition first. Reformatting a partition erases all data on the partition. If you do not reformat the partition, but you do install a Windows Server 2003 family product where there is already an operating system, that operating system will be overwritten, and you must reinstall any applications you want to use with the Windows Server 2003 family product.
If the hard disk has an existing partition, you can delete it to create more unpartitioned disk space for a partition for a product in the Windows Server 2003 family. Deleting an existing partition also erases any data on that partition.

Working with Dynamic Disks and Setup

A dynamic disk is a disk using the new storage type introduced with Windows 2000. If you changed a disk to dynamic and you want to perform a new installation on the disk, review the following.

If you used Windows 2000 or Windows XP to change a disk with no partitions directly to a dynamic disk, you must revert the disk to basic before you can install an operating system on it.

Caution

All data will be lost in the process of reverting the disk to basic, so back it up first.

You can use Windows 2000 or Windows XP to revert the disk, or you can use the Setup program for a product in the Windows Server 2003 family. To use Windows 2000 or Windows XP, follow the instructions in Help for your operating system. To use Setup for a product in the Windows Server 2003 family, during the partitioning phase, find the dynamic disk among the list of available partitions and then delete it (which erases all data on all volumes). You will be prompted to confirm your action. After you do this, the disk will contain only unpartitioned space, and you can use Setup to create a new (basic disk) partition on it.
If you plan to rerun Setup on a computer on which a product in the Windows Server 2003 family was already installed, and the computer contains dynamic disks, be sure to read about the limitations for installing operating systems on disks that are changed to dynamic using a product in the Windows Server 2003 family. For more information, see topics about dynamic disks and the partition table in Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Working with Volume, Mirror, or Stripe Sets or Stripe Sets with Parity

For a mirror set, break the mirror

If you are running Windows NT 4.0 on a computer that has a mirror set and you want to run Setup for Windows Server 2003, Standard Edition, on that computer, first back up the data (as a safeguard — the data will not be erased) and then break the mirror. Ensure that you have applied Service Pack 5 or later, which is required before running Setup for Windows Server 2003, Standard Edition. Then run Setup.

For a volume set, stripe set, or stripe set with parity, back up the data and delete the set

If you are running Windows NT 4.0 on a computer that has a volume set, stripe set, or stripe set with parity, and you want to run Setup for Windows Server 2003, Standard Edition, on that computer, first back up the data. Next, delete the set (which will delete the data). Ensure that you have applied Service Pack 5 or later, which is required before running Setup for Windows Server 2003, Standard Edition. Then run Setup.

After running Setup for Windows Server 2003, Standard Edition, you can make the disk dynamic, restore backed-up data as necessary, and make use of the volume options shown in the table later in this section. For more information about dynamic disks, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

If necessary, use the Ftonline support tool

The preceding methods are the recommended methods for preparing to run Setup for Windows Server 2003, Standard Edition, on a computer that contains a volume set, mirror set, stripe set, or stripe set with parity created with Windows NT 4.0. However, if you do not use these methods and you must access one of these sets after running Setup for Windows Server 2003, Standard Edition, you can use the Ftonline tool. The Ftonline tool is part of Support Tools in the products in the Windows Server 2003 family. For more information, see topics on Support Tools and on Ftonline in Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Types of Multidisk Volumes on Dynamic Disks

The disk sets described in the previous section have different names in the Windows Server 2003 family than they had in Windows NT 4.0:

Windows NT 4.0 name	Windows Server 2003 family name
Volume set	Spanned volume on a dynamic disk
Mirror set	Mirrored volume on a dynamic disk
Stripe set	Striped volume on a dynamic disk
Stripe set with parity	RAID-5 volume on a dynamic disk

Networks: TCP/IP, IP Addresses, and Name Resolution

TCP/IP is the network protocol that provides Internet access. It is the protocol used by most servers, although you can use additional or different network adapters and their associated protocols on your servers. Setup and Manage Your Server (which includes the Configure Your Server Wizard) are designed to make it easy to configure TCP/IP and the services that support it.

To use TCP/IP, make sure that each server is provided with an IP address, either a dynamic or automatic address provided through software, or a static address that you obtain and set. Because these addresses are numbers and therefore hard to remember, you will also have to provide users with names that are easier to use. Mapping this type of name to an IP address is called name resolution, and can be accomplished by various methods, primarily the Domain Name System (DNS) and Windows Internet Name Service (WINS). The following sections provide more information.

For detailed information about TCP/IP, DHCP, DNS, and WINS, see Help and Support Center as well as the Windows Server 2003 Technical Reference. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

IP Addresses

As outlined in the preceding section, using TCP/IP requires that an IP address be provided for each computer. This section describes IP addressing options. For descriptions of name resolution options, which are also needed in a TCP/IP network, see "Name Resolution for TCP/IP" later in this topic.

There are two basic approaches for providing an IP address for a server you are installing:

Dynamic Host Configuration Protocol (DHCP)

You can provide IP addresses to the computers on your network by configuring one or more DHCP servers, which provide IP addresses dynamically to other computers. A DHCP server must itself be assigned a static IP address.

One server or several servers can provide DHCP along with one or more name resolution services, which are called Domain Name System (DNS) and Windows Internet Name Service (WINS). The name resolution services are described in the next section, "Name Resolution for TCP/IP."

If you want to run Setup before you have finalized your decisions about which server to use as your DHCP server and what static IP address to assign to that server, you can choose Typical settings in the Networking Settings dialog box during Setup and complete the network configuration later. If you do this and there is no DHCP server in the network, Setup will use a limited IP addressing option called Automatic Private IP Addressing (APIPA). During the time that a server is using APIPA, it can communicate only with other computers using APIPA on the same network segment. A server that is using APIPA cannot make connections to the Internet (for browsing or e-mail), and cannot be used with DNS or Active Directory (which depends on DNS).

If you know which server you want to use as your DHCP server, when installing that server, in the Networking Settings dialog box in Setup, choose Custom settings, and specify a static IP address and related network settings. For more information about how to do this, see "Specifying Networking Settings" in, "Running Setup for a New Installation" later in this topic. After Setup, use Manage Your Server, along with information in Help and Support Center (click Start and then click Help and Support) to install the DHCP component and complete the configuration of the DHCP server.

Static IP addressing

For certain types of servers, you must assign a static IP address and subnet mask during or after Setup. These servers include DHCP servers, DNS servers, WINS servers, and any server providing access to users on the Internet. It is also recommended that you assign a static IP address and subnet mask for each domain controller. If a computer has more than one network adapter, you must assign a separate IP address for each adapter.

If you want to run Setup on a server before you have finalized your decision about the static IP address you want to assign to that server, you can choose Typical settings in the Networking Settings dialog box during Setup and configure that server later. In this situation, if there is a DHCP server in the network, Setup will obtain an IP address configuration from DHCP. If there is no DHCP server in the network, Setup will use Automatic Private IP Addressing (APIPA). APIPA is described in the previous item in this list, "Dynamic Host Configuration Protocol (DHCP)."

For more information about static IP addresses, including private IP addresses (which you choose from certain ranges of addresses) and public IP addresses (which you obtain from an Internet service provider), see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

After you have formed a plan for IP addressing, the next components to consider are those for name resolution, which is the process of mapping a computer name (something that users can recognize and remember) to the appropriate IP address. The following sections describe options for name resolution.

Name Resolution for TCP/IP

Name resolution is a process that provides users with easy-to-remember server names, instead of requiring them to use the numerical IP addresses by which servers identify themselves on the TCP/IP network. The name-resolution services are Domain Name System (DNS) and Windows Internet Name Service (WINS).

DNS

DNS is a hierarchical naming system used for locating computers on the Internet and private TCP/IP networks. One or more DNS servers are needed in most installations. DNS is required for Internet e-mail, Web browsing, and Active Directory. DNS is often used as a name resolution service in domains with clients running Windows 2000, Windows XP, or products in the Windows Server 2003 family.

DNS is installed automatically when you create a domain controller (or when you install Active Directory on an existing member server, which makes it a domain controller), unless the software for Windows Server 2003, Standard Edition, detects that a DNS server already exists for that domain. You can also install DNS by choosing the DNS server role in Manage Your Server or by using Add/Remove Windows Components, which is part of Add or Remove Programs in Control Panel.

If you plan to install DNS on a server, specify a static IP address on that server and configure that server to use that IP address for its own name resolution. For information about assigning a static IP address, see "Static IP Addressing" earlier in this topic For more information about configuring DNS, see Help and Support Center (after completing Setup, click Start, and then click Help and Support).

WINS

If you provide support for clients running Windows NT or any earlier Microsoft operating system, you might need to install Windows Internet Name Service (WINS) on one or more servers in the domain. You might also need to install WINS if it is required by your applications. You can install WINS after Setup by choosing the WINS server role in Manage Your Server or by using Add/Remove Windows Components, which is part of Add or Remove Programs in Control Panel.

If you plan to install WINS on a server, specify a static IP address on that server. For information about assigning a static IP address, see "Static IP Addressing" earlier in this topic. For more information about configuring WINS, see Help and Support Center (after completing Setup, click Start, and then click Help and Support).

Deciding Between Workgroups and Domains

A domain is a group of accounts and network resources that share a common directory database and set of security policies, and might have security relationships with other domains. A workgroup is a more basic grouping, intended only to help users find objects such as printers and shared folders within that group. Domains are the recommended choice for all networks except very small ones with few users.

In a workgroup, users might have to remember multiple passwords, one for each network resource. (In addition, different users can use different passwords for each resource.) In a domain, passwords and permissions are simpler to keep track of, because a domain has a single, centralized database of user accounts, permissions, and other network details. The information in this database is replicated automatically among domain controllers. You determine which servers are domain controllers and which are simply members of the domain. You can determine these roles not only during Setup but afterward.

Domains, and the Active Directory directory system of which they are a part, provide many options for making resources easily available to users while maintaining good monitoring and security. For more information about Active Directory, see Help and Support Center (after completing Setup, click Start, and then click Help and Support).

Planning for Domain Controllers and Member Servers

With Windows Server 2003, Standard Edition; Windows Server 2003, Enterprise Edition; or Windows Server 2003, Datacenter Edition, servers in a domain can have one of two roles: domain controllers, which contain matching copies of the user accounts and other Active Directory data in a given domain, and member servers, which belong to a domain but do not contain a copy of the Active Directory data. (A server that belongs to a workgroup, not a domain, is called a stand-alone server.) It is possible to change the role of a server back and forth from domain controller to member server (or stand-alone server), even after Setup is complete. However, it is recommended that you plan your domain before running Setup and change server roles (and server names) only when necessary.

Multiple domain controllers provide better support for users, compared to a single domain controller. With multiple domain controllers, you have multiple copies of user account data and other Active Directory data; however, it is still important to perform regular backups, including Automated System Recovery backups, and familiarize yourself with the methods for restoring a domain controller. In addition, multiple domain controllers work together to support domain controller functions, such as carrying out logon validations. For more information about domain controllers and other Active Directory topics, see Help and Support Center (after completing Setup, click Start, and then click Help and Support). Special Aspects of Windows Server 2003 Family Domain Controllers

As you manage your Windows Server 2003 family domains, you might want to learn more about operations master roles. This section describes operations master roles briefly.

Operations master roles are special roles that are assigned to one or more domain controllers in an Active Directory domain. The domain controllers that are assigned these roles perform operations that are single-master (not permitted to occur at different places in a network at the same time). For example, a single domain controller must oversee the creation of security identifiers for new resources (such as new computers) to ensure that the identifiers are unique.

The first domain controller installed in a domain is automatically assigned all the operations master roles. You can change the assignment of operations master roles after Setup, but in most cases this is not necessary. You will need to be particularly aware of operations masters roles if problems develop on an operations master or if you plan to take one out of service. You can find more information in Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Additional Information About Domains

For information about upgrading a domain, see "Getting Ready for an Upgrade" at the beginning of this document.

For detailed information about domains and forests, operations master roles, the global catalog, and other Active Directory topics, see Help and Support Center, which is available after you install (or upgrade to) a product in the Windows Server 2003 family on your first server. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

For information about setting up multiple domains with structured relationships, see the Windows Server 2003 Deployment Kit. You can view the Windows Deployment and Resource Kits on the Web at: http://go.microsoft.com/fwlink/?LinkID=29883

Top of page

Running Setup for a New Installation

Preparing Your System for a New Installation
Starting Setup
Planning for Unattended Setup
Entering Server Settings for a New Installation
Configuring Your Server
Product Activation for Products in the Windows Server 2003 Family

This topic provides recommendations for preparing your system for a new installation, and it explains how to run Setup for Microsoft® Windows Server™ 2003, Standard Edition. Before reading this topic or running Setup, be sure to read "Getting Ready for a New Installation" earlier in this topic.

This topic also provides a description of Manage Your Server, which appears after Setup completes the installation. Together, Setup and Manage Your Server help you get your servers up and running quickly. For more information about Manage Your Server, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

Preparing Your System for a New Installation

This section describes the basic steps to take in preparing your server for a new installation.

Checking the System Log for Errors

If the computer already has a working operating system, review the system log for recent or recurring errors (especially hardware errors) that could cause problems during the installation. For information about viewing event logs, see Help for the operating system on your computer.

Backing Up Files

Before you perform a new installation, it is recommended that you back up your current files, unless the computer has no files or the current operating system files have been damaged. You can back up files to a variety of different media, such as a tape drive or the hard disk of another computer on the network.

Uncompressing the Drive

Uncompress any DriveSpace or DoubleSpace volumes before installing. Do not install Windows Server 2003, Standard Edition, on a compressed drive unless the drive was compressed with the NTFS file system compression feature.

Preparing Mirror Sets and Other Disk Sets (Windows NT 4.0 only)

Disconnecting UPS Devices

Starting Setup

This section explains how to start Setup for a new installation.

Setup works in several stages, prompting you for information, copying files, and restarting. Setup concludes with Manage Your Server, which you can use to adjust the server configuration for your specific needs. For more information about Manage Your Server, see Help and Support Center after you run Setup. To view Help and Support Center, click Start, and then click Help and Support.

For important information to help you make the initial choices offered by Setup, see "Getting Ready for a New Installation" earlier in this topic. For information about unattended Setup and other options available when you start Setup, see "Planning for Unattended Setup" later in this topic.

Providing a Mass Storage Driver or a HAL File

Mass Storage Drivers and the Setup Process

If your controller is compatible, but you are aware that the manufacturer has supplied a separate driver file for use with products in the Windows Server 2003 family, obtain the file (on a floppy disk) before you begin Setup. During the early part of Setup, a line at the bottom of the screen prompts you to press F6. Further prompts will guide you in supplying the driver file to Setup so that it can gain access to the mass storage controller.

If you are not sure whether you must obtain a separate driver file from the manufacturer of your mass storage controller, you can try running Setup. If the controller is not supported by the driver files on the Setup CD and therefore requires a driver file that is supplied by the hardware manufacturer, Setup stops and displays a message saying that no disk devices can be found, or it displays an incomplete list of controllers. After you obtain the necessary driver file, restart Setup, and press F6 when you are prompted.

Using a Custom HAL File

If you have a custom Hardware Abstraction Layer (HAL) file supplied by your computer manufacturer, before you begin Setup, locate the floppy disk or other medium containing the file. During the early part of Setup, a line at the bottom of the screen prompts you to press F6: at this time press F5 (not F6). After you press F5, follow the prompts to include your HAL file in the Setup process.

Starting Setup

The sections that follow, "Starting a New Installation from a CD" and "Starting a New Installation from a Network," explain how to start Setup for a new installation. For information about planning for and starting Setup for an upgrade, see "Getting Ready for an Upgrade" and "Running Setup for an Upgrade" earlier in this topic.

For information about unattended Setup and other options available when you start Setup, see "Planning for Unattended Setup" later in this topic.

Starting a New Installation from a CD

If you use the Setup CD, you have several options for starting Setup, as explained in the following procedures:

Note

If you are running Setup on a computer running Windows 3.x or MS-DOS, for best efficiency, use disk caching. Otherwise, the Setup process (started from Winnt.exe) could take a long time. To enable disk caching on a computer running Windows 3.x or MS-DOS, you can use SMARTDrive. For information about SMARTDrive, see the documentation for Windows 3.x or MS-DOS.

To start Setup from the CD on a computer running MS-DOS

Insert the CD in the drive.
At the command prompt, type:

d :

where d is the drive letter of the CD-ROM drive.
Type:

cd i386
Type:

winnt
Follow the setup instructions.

Note

This procedure does not apply to computers with x64 processors.

To start Setup from the CD on a computer running Windows

Before starting this procedure on a computer running Windows NT® 4.0, apply Service Pack 5 or later.
Insert the CD in the drive.
To begin Setup, do one of the following:
For a computer running any version of Windows other than Windows 3.x, wait for Setup to display a dialog box.
For a computer running Windows 3.x, use File Manager to change to the CD-ROM drive and to change to the I386 directory, and then double-click Winnt.exe.
Follow the Setup instructions.

Note

This procedure does not apply to computers with x64 processors.

Another way of using the Setup CD is to start the computer from the CD-ROM drive. This method applies only if you want to perform a new installation, not an upgrade. Using this method, you can perform an installation on a computer that does not have an operating system, although you can also use this method on computers that have operating systems.

To start Setup for a new installation from the CD

Determine whether the computer on which you want to start Setup can be started from the CD-ROM drive and whether you want to perform a new installation (not an upgrade). Continue only if both are true.
Insert the CD in the drive, and then restart the computer.
Follow the instructions for your operating system to boot the computer from the CD.
Wait for Setup to display a dialog box, and then follow the Setup instructions.

Starting a New Installation from a Network

To install Windows Server 2003, Standard Edition, from a network, you either share the files directly from the CD or copy them to a shared folder. Then, you start the appropriate program to run Setup.

To install Windows Server 2003, Standard Edition, from a network

On a network server, share the installation files, either by inserting the CD and sharing the CD-ROM drive or by copying the files from the I386 folder on the CD to a shared folder.
On the computer on which you want to install Windows Server 2003, Standard Edition, connect to the shared Setup files:
- If you are sharing the CD-ROM drive, connect to the shared drive and change to the I386 folder.
- If you are sharing a folder, connect to that folder.
Find and run the appropriate file in the I386 directory of the CD or in the shared folder:
- From a computer running MS-DOS or Windows 3.x, run Winnt.exe.
- From a computer running Windows 95, Windows 98, Windows Millennium Edition, Windows NT with Service Pack 5 or later, Windows 2000, or Windows XP, run Winnt32.exe.
Follow the Setup instructions.

Note

This procedure does not apply to computers with x64 processors.

Planning for Unattended Setup

This section provides general information about unattended Setup. For detailed instructions on running unattended Setup (also called automated installation), see the Microsoft Windows Resource Kits Web site at: http://go.microsoft.com/fwlink/?LinkId=29883

To start unattended Setup on a computer running MS-DOS or Windows 3.x, use Winnt.exe (with the appropriate options).
To start unattended Setup on a computer running Windows 95, Windows 98, Windows Millennium Edition, Windows NT, Windows 2000, Windows XP, or a product in the Windows Server 2003 family, use Winnt32.exe (with the appropriate options). With Windows NT 4.0, before starting unattended Setup, apply Service Pack 5 or later.

To learn about unattended Setup, see the following sources:

For detailed instructions on running unattended Setup (also called automated installation): See "Automating and Customizing Installations" in the Windows Server 2003 Deployment Kit (described in the first paragraph of this section).
To view the command options available for Winnt.exe: On a computer running Windows 3.x or MS-DOS, insert the Setup CD for Windows Server 2003, Standard Edition, in the CD-ROM drive and open the command prompt. Then, change to the CD-ROM drive, change to the I386 directory, and type:

winnt /?
To use an x86-based or x64-based computer to view the command options available for Winnt32.exe: On a computer running Windows 95, Windows 98, Windows Millennium Edition, Windows NT, Windows 2000, Windows XP, or a product in the Windows Server 2003 family, insert the Setup CD for Windows Server 2003, Standard Edition, in the CD-ROM drive, and open the command prompt (click Start, click Run, and then type cmd). Then, change to the CD-ROM drive, change to the I386 directory, and type:

winnt32 /?

Entering Server Settings for a New Installation

If you are upgrading, you can skip this section because Setup will use your previous settings.

After you start Setup, a process begins in which necessary Setup files are copied to the hard disk. During this process, Setup displays dialog boxes that you can use to select various options.

The following sections outline the items that you specify as you run Setup. Before working with these sections, on your Setup CD, see "Getting Ready for a New Installation" earlier in this topic for important background information about issues such as partitions, file system choices, and the handling of TCP/IP name resolution. You can modify some settings after running Setup by using the Configure Your Server Wizard or other configuration tools.

Notes

If you have a mass storage controller (such as a SCSI, RAID, or Fibre Channel adapter) for your hard disk, see "Providing a Mass Storage Driver or a HAL File" earlier in this topic for information about how to ensure that the controller driver file is available to the Setup program.
If you have specialized hardware that requires a custom Hardware Abstraction Layer (HAL) file supplied by your computer manufacturer, see "Using a Custom HAL File" earlier in this topic for information about how to make the HAL file available during Setup.

Choosing or Creating a Partition for Windows Server 2003, Standard Edition

During a new installation of Windows Server 2003, Standard Edition, a dialog box gives you the opportunity to create or specify a partition on which you want to install. You can create a partition from the available unpartitioned space, specify an existing partition, or delete an existing partition to create more unpartitioned disk space for the new installation. If you specify any action that will cause information to be erased, you will be prompted to confirm your choice.

For more information, see the section on planning partitions for new installations in "Getting Ready for a New Installation" earlier in this topic.

Important

If you delete an existing partition, all data on that partition is erased.
Performing a new installation of Windows Server 2003, Standard Edition, on a partition that contains another operating system overwrites the existing operating system.

Selecting Regional and Language Options

You can set up Windows Server 2003, Standard Edition, to use multiple languages and regional options.

If you select a European country or region in the list of countries/regions, or if you live in a country or region where the euro has been introduced, it is a good idea to verify that the default currency settings in Regional and Language Options meet your needs. After you run Setup, you can modify these options by clicking Regional and Language Options in Control Panel.

Personalizing Windows

Enter your name and, as an option, your organization.

If you select a European country or region in the list of countries/regions, or if you live in a country or region where the euro has been introduced, it is a good idea to verify that the default currency settings in Regional and Language Options meet your needs. After you run Setup, you can modify regional and language options by clicking Regional and Language Options in Control Panel.

Choosing a Licensing Mode

Select your client licensing mode. For information about licensing modes, see "Getting Ready for a New Installation" earlier in this topic. If you are unsure which mode to use, select Per Server, because you can change once from Per Server mode to Per Device or Per User mode at no cost.

Entering Your Computer Name

During Setup, in the Computer Name and Administrator Password dialog box, follow the instructions for entering your computer name. The recommended length for most languages is 15 characters or less. For languages that require more storage space per character, such as Chinese, Japanese, and Korean, the recommended length is 7 characters or less.

It is recommended that you use only Internet-standard characters in the computer name. The standard characters are the numbers from 0 through 9, uppercase and lowercase letters from A through Z, and the hyphen (-) character. Computer names cannot consist entirely of numbers.

If you are using DNS on your network, you can use a wider variety of characters, including Unicode characters and other nonstandard characters, such as the ampersand (&). Using nonstandard characters might affect the ability of non-Microsoft software to operate on your network. For more information about DNS, see the "Name Resolution for TCP/IP" section in "Getting Ready for a New Installation" earlier in this topic.

The maximum length for a computer name is 63 bytes. If the name is longer than 15 bytes (15 characters in most languages, 7 characters in some), computers running Windows NT Server 4.0 and earlier will recognize this computer by the first 15 bytes of the name only. In addition, there are additional configuration steps for a name that is longer than 15 bytes. For more information, in Help and Support Center, see "Namespace planning for DNS." To view Help and Support Center, after running Setup, click Start, and then click Help and Support.

If a computer is part of a domain, you must choose a computer name that is different from any other computer in the domain. To avoid name conflicts, the computer should be unique on the domain, workgroup, or network. If this computer is part of a domain, and it contains more than one operating system, you must use a unique computer name for each operating system that is installed. For example, if the computer name is FileServerNT when the computer is started with Windows NT Server 4.0, the computer must have a different name, perhaps FileServerNew, when it is started with a product in the Windows Server 2003 family. This requirement also applies to a computer that contains multiple installations of the same operating system. For more information, in Help and Support Center, see "Planning your domain structure."

Setting the Administrator Account Password

During Setup, in the Computer Name and Administrator Password dialog box, type a password of up to 127 characters in the Administrator Password box. For the strongest system security, use a password of at least 7 characters, and use a mixture of uppercase and lowercase letters, numbers, and other characters, such as *, ?, or $.

Important

After Setup is completed, for best security, change the name of the Administrator account (it cannot be deleted) and keep a strong password on the account at all times. For more information about security and the administrative rights and permissions held by the Administrator account and the Administrators group, see Help and Support Center. You can view Help and Support Center after Setup is complete by clicking Start, and then clicking Help and Support.

Setting the Date and Time

During Setup, in the Date and Time Settings dialog box, set the date, time, and time zone. If you want the system to automatically adjust for daylight saving time, select the Automatically adjust clock for daylight saving changes check box.

You can change your computer's date and time after Setup is complete. If your computer is a member of a domain, your computer clock is probably synchronized automatically by a network time server. If your computer is not a member of a domain, you can synchronize your computer clock with an Internet time server.

Specifying Networking Settings

You can specify networking information for TCP/IP or other protocols during Setup, or you can use typical settings and then make any necessary changes to your networking configuration after installation.

For basic background information about TCP/IP configuration and Setup, see "Networks: TCP/IP, IP Addresses, and Name Resolution" in "Getting Ready for a New Installation" earlier in this topic. For additional information about TCP/IP configuration, see Help and Support Center. To open Help and Support Center, after completing Setup, click Start, and then click Help and Support.

To allow Setup to assign or obtain an IP address

When you click Typical settings in the Networking Settings dialog box, Setup checks to see if there is a DHCP server on your network. If there is a DHCP server on your network, DHCP provides an IP address. If there is no DHCP server on your network, Setup will use a limited IP addressing option called Automatic Private IP Addressing (APIPA). On a server using APIPA, complete the network configuration after Setup, because a server using APIPA can communicate only with other computers using APIPA on the same network segment.

For basic background information about IP address configuration, including additional information about the limitations of APIPA, see "IP Addresses" in "Getting Ready for a New Installation" earlier in this topic.

To specify a static IP address and settings needed for DNS and WINS

During Setup, in the Networking Settings dialog box, click Custom settings, and then click Next.
In the Networking Components dialog box, click Internet Protocol (TCP/IP).
Click Properties.
In the Internet Protocol (TCP/IP) Properties dialog box, click Use the following IP address.
In IP address, Subnet mask, and Default gateway, type the appropriate addresses.
Under Use the following DNS server addresses, type the address of a preferred DNS server and, optionally, an alternate DNS server.

If the local server is the preferred or alternate DNS server, type the same IP address as assigned in the previous step.
If you will use a WINS server, click Advanced, and then click the WINS tab in the Advanced TCP/IP Settings dialog box to add the IP address of one or more WINS servers.
Click OK in each dialog box, and continue with Setup.

Specifying the Workgroup or Domain Name

A domain is a group of accounts and network resources that share a common directory database and set of security policies and might have security relationships with other domains. A workgroup is a more basic grouping, intended only to help users find objects such as printers and shared folders within that group. Domains make it easier for an administrator to control access to resources and keep track of users. For more information, see "Getting Ready for a New Installation" earlier in this topic.

For more information about name conflicts, see "Entering Your Computer Name" earlier in this topic.

Configuring Your Server

When Setup is complete, the computer restarts. Setup has now completed the basic installation. Manage Your Server appears on the screen the first time you log on as the computer's administrator. You can use Manage Your Server to install and configure server roles, including file servers, print servers, Web and media servers, and networking and communications servers. You can start Manage Your Server at any time if you are logged on as an administrator. To start Manage Your Server, click Start, and then either click Manage Your Server or point to All Programs, point to Administrative Tools, and then click Manage Your Server.

Also at this point, you can view Help and Support Center by clicking Start, and then clicking Help and Support. If you have used Windows NT in the past, one topic you might find useful is "New ways to do familiar tasks," which is available in Help and Support Center.

Choosing Server Components

You can use the Windows Components Wizard to select the appropriate components for your server. To use this wizard, after running Setup, click Start, and then click Control Panel. In Control Panel, double-click Add or Remove Programs, and then, on the left side of the dialog box, click Add/Remove Windows Components. With this wizard you can choose and install individual components.

Product Activation for Products in the Windows Server 2003 Family

The following details help explain product activation:

Your local operating system does some of the work of creating the association between your product key and the computer it is installed on. It uses your Product ID, which is derived from your product key, and puts it together with a coded number (called a hardware hash) that uniquely represents the hardware components that make up your computer (without representing any personal information or anything about the software). When you activate, you provide this number, through the Internet or by phone, so that it can be confirmed as representing a legal installation. After you activate, if another person has an illegal copy of your software, and that person tries to use your product key to activate the software on another computer, the coded numbers will show that your product key is already associated with a particular set of hardware (your computer). The other person’s activation will not work.

The Installation ID that the local operating system generates, representing the association between your product key and the computer it is install on, is used only for the purpose of activation.

Methods for activation After your operating system is installed, begin activation by clicking Start, and then clicking Activate Windows. (You can also click the key icon that appears in the lower right corner of the screen.) By following the instructions on the screen, you can activate through the Internet or by phone:

Internet: When you activate through the Internet, your computer transmits coded information that shows that your product key is associated with your computer hardware. Activation is carried out through a secure server. A confirmation ID is passed back to your computer, automatically activating your product. This process normally takes just a few seconds to complete. No personally identifiable information is required to activate your product.
Phone: When you activate by phone, information on the screen guides you through a few simple steps. When you choose the country or region where you are located, a phone number (toll-free, wherever possible) appears on your screen. When you call the number, a customer service representative asks for the Installation ID that is displayed on your screen. The customer service representative enters that number into a secure database, confirms that the number represents a legally installed product, and provides a confirmation ID to you. Then, you type the confirmation ID into the spaces provided on the screen, and activation is complete.

Product Activation Compared to Product Registration

Top of page

Setting Up Windows Server 2003, Standard Edition

Contents

Getting Ready for an Upgrade

Upgrades Compared to New Installations

Points to consider for an upgrade:

Points to consider for a new installation:

Operating Systems from Which You Can Upgrade

System Requirements and Hardware Compatibility

System Requirements

For CD-ROM installation:

For network installation:

For additional hardware-based functionality:

Hardware Compatibility

Taking an Inventory of Devices That Do Not Use Plug and Play

Mass Storage Drivers and the Setup Process

Using a Custom Hardware Abstraction Layer (HAL) File

Understanding ACPI BIOS

Using Dynamic Update for Updated Drivers and Other Setup Files

Important Files to Review

Upgrades in a Domain Containing Windows 2000 Domain Controllers

Preparing for Upgrades in a Domain Containing Windows 2000 Domain Controllers

Upgrades in a Windows NT 4.0 Domain

Other Sources of Information

Stages in the Process of Upgrading Servers in a Windows NT 4.0 Domain

Basic Preparations for the Upgrade of Servers in a Windows NT Domain

Working with Volume, Mirror, or Stripe Sets or Stripe Sets with Parity

Types of Multidisk Volumes on Dynamic Disks

Planning Server Roles and the Order of Server Upgrades from Windows NT

Features Available with the Upgrade of Any Server in a Windows NT Domain

Features Available with the First Domain Controller Upgrade in a Windows NT Domain

Planning DNS for the Upgrade of Domain Controllers Running Windows NT 4.0

Performing Backups and Carrying out Upgrades of Domain Controllers Running Windows NT 4.0

Applying the Service Pack, Performing Backups, and Other Preparations

Running the Active Directory Installation Wizard After Upgrading the Primary Domain Controller

Learning About Active Directory After the First Server is Upgraded

Completing Further Upgrades from Windows NT 4.0

Running Setup for an Upgrade

Preparing Your System for an Upgrade

Checking the System Log for Errors

Backing Up Files

Preparing Mirror Sets and Other Disk Sets for an Upgrade (Windows NT 4.0 only)

Disconnecting UPS Devices

Reviewing Hardware and Software

Starting Setup for an Upgrade

Providing a Mass Storage Driver or a HAL File

Mass Storage Drivers and the Setup Process

Using a Custom HAL File

Starting an Upgrade

Planning for Unattended Setup

Product Activation for Products in the Windows Server 2003 Family

Product Activation Compared to Product Registration

Getting Ready for a New Installation

Upgrades Compared to New Installations

Points to consider for an upgrade:

Points to consider for a new installation:

System Requirements and Hardware Compatibility

System Requirements

Hardware Compatibility

Taking an Inventory of Devices That Do Not Use Plug and Play

Mass Storage Drivers and the Setup Process

Using a Custom Hardware Abstraction Layer (HAL) File

Understanding ACPI BIOS

Using Dynamic Update for Updated Drivers and Other Setup Files

Important Files to Review

Decisions to Make for a New Installation

What licensing mode to use

Whether you want to be able to choose between different operating systems each time you start the computer

What file system to use on the installation partition

What partition or volume you plan to install the operating system on

How to handle IP addresses and TCP/IP name resolution

Whether to create domains or workgroups for your servers

Choosing a Licensing Mode

Deciding Whether a Computer Will Contain More Than One Operating System

Reasons to Install Only One Operating System

Requirements for Setting Up a Computer with Multiple Operating Systems

On computers that contain MS-DOS and Windows Server 2003, Standard Edition:

On computers that contain Windows 98 or Windows Millennium Edition and Windows Server 2003, Standard Edition:

On computers that contain Windows NT 4.0 and Windows Server 2003, Standard Edition:

On computers that contain some combination of Windows Server 2003, Standard Edition, with Windows 2000 or Windows XP, or that contain multiple partitions with products in the Windows Server 2003 family:

Multiple Operating Systems and File System Compatibility