1. Installing Windows 2000 Professional

1.1   Perform an attended installation of Windows 2000 Professional.

 Windows 2000 Professional requires 2 GB with a minimum of 650MB free space on the boot partition. If your computer only has a 1GB hard disk with 1GB free space, it meets the minimum free space requirement but does not satisfy the hard disk requirement. You still cannot install Windows 2000 on your computer.

Boot Partition contains the Windows operating system files (i.e. \winnt) and its support files (i.e. \winnt\system32). To check which drive is your boot partition, you need to go to Command Prompt (From Start -> Run, and type CMD) and type echo %systemroot%

In the above example, your Windows operating system files are in C:\winnt and your boot partition is on C: drive.

For the Windows 2000 exams, besides boot partition, you also need to know the system partition.

The System Partition is the volume that has the hardware-specific files (i.e., NTLDR or OSLOADER.EXE, BOOT.INI, etc.) needed to load Windows NT. On x86 based computers, it must be a primary partition that has been marked active for startup purposes. The user should be careful not to change the drive letter of the system partition because many MS-DOS and Windows programs make reference to the C: drive. The system partition can be (but does not have to be) the same partition as the boot partition.

 

Some people might mix the System partition with Boot Partition because they thought the system partition contains system files. Please remember the following two sentences for the Windows 2000 exams.

1.       The System Partition includes the files to boot Windows 2000.

2.       The Boot Partition includes the System files.

A domain is a logical grouping of computers that share common security and user account information. A domain employs the user-level security model. As such, a user can use the same account to log on from any machine within the domain and get the same permission to access the resources within the domain. A domain model provides the following advantages:

1.       Single user logon: Users access network resources according to the rights assigned by the corresponding domain. For example, after a user ww log on to computer1, he can access the authorized resources in computer1, as well as the authorized resources in other computers without extra login requests. Moreover, users can logon to the domain in any computer of the domain.

2.       Centralized administration: A centralized view of the entire network from any workstation on the network provides the ability to track and manage information on users, groups, and resources in a distributed network.

3.       Resource universal access: The administrator can grant users access to shared resources regardless of his location. Resource universal access makes 'Single User logon' possible.

 

A FQDN (fully qualified domain name) is a hierarchical name that provides a naming scheme for TCP/IP hosts. FQDNs are most commonly used in the Internet with a Domain Name System (DNS). FQDN consists of two parts: the domain name and the host name. For example, using the fictional domain, an FQDN for a web server could be www.yahoo.com. Note that the DNS name can actually be multi-part with each part of the name separated by a period (.).  Microsoft suggests people use the domain part of the FQDN as the domain name in Windows 2000.

 

A workgroup model distributes the administration tasks. Each computer has its own database for user¡¯s name and password. It is not saying that you cannot access the resource on other computers using workgroup model, but your login name could be different on another computer. Therefore, in a workgroup model, you assign passwords to each shared resource. The shared resource distinguishes people from the password provided. For example, if you shared a directory c:\ww on a computer, you assign "wwread" as the read-only password and "wwfull" for Full Access. Anyone knows the password "wwread" can only read the file on shared the directory, while those who knows the password "wwfull" can modify the files.

You can set up a Windows 2000 Professional computer join the domain or the workgroup. 

 

Active Directory is the directory service that stores user accounts; computer accounts and other network objects and makes this information available to users and network administrators. Active Directory gives network users access to permitted resources anywhere on the network using a single logon process. It provides network administrators with an intuitive hierarchical view of the network and a single point of administration for all network objects.

Every domain has its own active directory. You cannot implement the active directory in a workgroup environment because a workgroup model distributes the administration tasks while the active directory centralize the management of the user accounts. You can install the Active Directory on a Windows 2000 server computer while not the Windows 2000 Professional. When you install the Active Directory, you can create a new domain tree. (Figure 1.1.1) As long as a Windows 2000 domain is created, you can let the Windows 2000 Professional computers to join the domain and user domain user account while not the local user account to logon to the domain.

  

Figure 1.1.1 Active Directory Installation Wizard

Winnt.exe and winnt32.exe are the commands to perform an installation of or upgrade to Windows 2000. You can find the winnt.exe and winnt32.exe files under the i386 directory of the Windows 2000 Retail CD. Some people might be wondering when to use winnt.exe and when to use winnt32.exe. The rule is quite simple. When you install Windows 2000 from a 16-bit operating system, you need to run winnt.exe and when you install Windows 2000 from a 32-bit operating system, you need to run winnt32.exe. For those Microsoft operating systems we mentioned before, MS-DOS is a 16-bit operating system and Windows 95, Windows 98, Windows NT and Windows 2000 are 32-bit operating systems. Therefore, if you want to install Windows 2000 on a Windows 98 computer, you must use winnt32.exe.

 

Microsoft has another program called setup.exe (in the root directory of the Windows 2000 retail CD). If you are headache of figuring out whether your current OS is 16-bit or 32-bit, you can just run setup.exe for the installation and the setup.exe will detect the current OS and select the correct files for you. If your computer supports the bootable CD-ROM nowadays, you can also boot the computer using the Windows 2000 CD and perform the Windows 2000 installation. Windows 2000 supports FAT, FAT32 and NTFS three file systems. However, you might encounter other file systems such as HPFS on the Windows 2000 exam. Therefore, let's study the definitions of these file systems first:

 

FAT (File Allocation Table) is a table maintained by some operating systems such as MS-DOS, Windows 95, Windows NT, Windows 2000 as well as OS/2 to keep track of the status of various segments of disk space used for file storage. FAT is also called the FAT file system. FAT is the only file system that is supported by all the Microsoft operating systems. Because MS-DOS can only support FAT file system, if you want to dual boot MS-DOS with Windows 2000, you must implement the FAT file system. On a Windows 95 or DOS system, FAT is limited to 2 GB hard drives. Some people may have found that when they purchased a 6GB hard disk, they had to partition the disk to three 2GB disks using the fdisk command and ended up with (c:, d: and e: three disk drives in their system. This is because of he 2GB limitation of the FAT file system.

 

A dual boot is a configuration that you have multiple operating systems installed on your computer. You can choose which operating systems you want to start each time you reboot the computer.

 

FAT32 is the 32-bit version of FAT. Although FAT is a 16-bit file system, we usually don¡¯t call FAT FAT16. FAT32 was available on Windows 95 since late 1996 and it increased the drive limitation to 2TB. FAT32 is supported on Windows 95, Windows 98 and Windows 2000 while not MS-DOS and Windows NT. Therefore, if you want to dual boot Windows 2000 with Windows 98, you can either use FAT or FAT32 file system. However, if you want to dual boot Windows 2000 with MS-DOS, you can only use FAT.

To check what your current file system is, you can open Windows Explorer and right clicks the C: drive, click properties to see if your system is FAT, FAT32 or NTFS. Because FAT file system does not support partitions larger than 2 GB, if you attempt to format a partition larger than 2 GB with FAT in Windows 2000 Setup automatically formats the partition with FAT32.

 

NTFS (NT File System) is an advanced file system that was designed for use specifically within the Windows NT operating system. NTFS supports file system recovery, extremely large storage media, long filenames, and various features for the POSIX subsystem. It also supports object-oriented applications by treating all files as objects with user-defined and system-defined attributes.

For the Windows 2000 exam, you need to know three things about the NTFS:

1.       NTFS only works on the Windows NT or Windows 2000 platform

2.       NTFS has more security features than FAT

3.       You can run most of the MS-DOS and Windows application on NTFS

Suppose we have two hard disks in the system. We format C: to FAT and D: to NTFS. When you open the Properties of C: and D: drivers, you will see the following diagram:

Figure 1.1.2 compare FAT and NTFS

In the figure 1.1.2, you will find that D: drive has four more tabs called Hardware, security, Web Sharing and Quota. Hence, you can implement more security issues with NTFS. In fact, when you logon to a Windows 2000 computer locally, anybody can access the folders on the FAT file system. There is no way to prevent anyone from accessing these FAT folders if he can login locally. That¡¯s the reason why Microsoft suggests people use NTFS on Windows 2000. For the drives using NTFS, user-level security can be implemented on different folders and you can prevent users from accessing local system files.

For small directories, the FAT file system may be faster to get to the file, because:

-          The FAT directory structure is simpler.

-          The FAT directory size is smaller for an equal number of files.

-          FAT doesn¡¯t have to check permissions for an individual file or whether a specific user has access to the file or directory.

 

The NTFS file system uses a binary tree structure for all directories. This structure minimizes the number of disk accesses required to find a file, which means that the NTFS file system should be faster for larger directories. The FAT file system is best used on smaller volumes than approximately 400 - 500 MB. That¡¯s because the FAT file system starts out with very little overhead. However, the FAT file system is very inefficient for the volumes large that 1 Gigabyte (GB). Because Windows 2000 requires a 2GB hard disk, you¡¯d better always implement NTFS instead of FAT except that you need to dual boot the system with Windows 95/98 and MS-DOS.

 

HPFS (High Performance File System) from IBM is the file system introduced with OS/2 that handles large disks (2TB volumes; 2GB files) and long file names (256 bytes). It coexists with the existing FAT system.  Please note that HPFS has nothing to do with HP Corporation but is an IBM product. As far as Microsoft Products go HPFS is compatible only with Windows NT 3.5x.  To reduce the system overhead, Windows NT 4.0 and Windows 2000 no longer support HPFS. It is easy to understand that when not many people use Microsoft¡¯s products, Microsoft needs to support as many file systems as possible. After Microsoft acquires the market share, Microsoft is focusing more on its operating system¡¯s performance. Supporting more file systems will definitely increase the system¡¯s overhead.

 The following table lists the supported file system on MS-DOS, Windows 3.1, 95/98, NT as well as Windows 2000.

 

 

MS-DOS Windows 3.1

Windows 95

Windows 98

Windows NT 3.1, 3.5x

Windows NT 4.0

Windows 2000

FAT

Y

Y

Y

Y

Y

FAT32

N

Y

N

N

Y

HPFS

N

N

Y

N

N

NTFS

N

N

Y

Y

Y

 

CDFS is used on a CD-ROM driver. It is read-only and when the files on CD-ROM are copied to the Windows 2000 system, it converts the file to FAT, FAT32 or NTFS depending on where the file is copied to.

 

1.2 Perform an unattended installation of Windows 2000 Professional.

 

1.2.1 Install Windows 2000 Professional by using Windows 2000 Server Remote Installation Services (RIS).

More and more devices support plug and play nowadays. In the future, a computer itself can be plug-and-played. People had a dream that after connecting the computer to the network, you don¡¯t need to configure the network setting, you do not need to install the software, and you can just use the computer. Microsoft¡¯s RIS technology makes this dream come true. RIS technology represents the future of the computers. No wonder there are lots of RIS related questions in the MCSE exams.

 

RIS (Remote Installation Services) is a technology that allows an administrator to remotely set up new client computers that support remote booting, without having to visit each client. Remote Installation Services must be installed on a Windows 2000 Server computer. Currently, Windows 2000 Professional is the only operating system that can be installed using the RIS. You cannot install a Windows 2000 server through the RIS. To implement RIS in your network, you must have Active Directory, DHCP server and DNS server running in your network. The Active directory is used to create the user and computer account for the new computer. DHCP server is used to provide the IP address. An A record and SRV record in the DNS server can be used to locate the domain controller where the Active Directory is located. 

 

PXE (Preboot EXecution Environment) is a technology that enables a PC to boot from the server. PXE supports remote booting (boot the OS) or boot the installation program to install the OS. For a client computer, it must use PXE DHCP¨Cbased remote boot technology to connect to the Windows 2000 RIS server and install the operating system.

Therefore, if your client uses PXE compliant network adapter or is a NetPC, you can use it as a RIS client. If your computer is not a NetPC or has a PXE compliant network adapter installed, you can use rbfg.exe in the Command Prompt to create the remote boot disk for the RIS client and the RIS client can use it to boot the computer.

 

Figure 1.2.1.1 Window 2000 Remote Boot Disk Generator

 The RIS server must contain the operating system image to be installed in either compact disc (CD) or Remote Installation Preparation wizard (RIPrep) image format.

 

Figure 1.2.1.2 Implement RIS in your network

Creating the CD image is quite simple, after you install the RIS service on a Windows 2000 Server, you can specify the location of the Windows 2000 Professional CD and copy it to a network share that can be accessed from the RIS client computers. If you want to deploy some applications with the RIS installation, you can install those applications on a sample Windows 2000 Professional computer. After that, you can run riprep.exe from the sample Windows 2000 Professional computer. It will load the RIPrep image of the sample Windows 2000 Professional computer to RIS server. Sometimes you also need to modify the riprep.sif file for the unattended installation.

 

1.2.2 Install Windows 2000 Professional by using the System Preparation Tool.

If you install hundreds of Windows 2000 Professional computers for your company, most computers might have the identical hardware configuration. The most efficient installation method should be disk duplication using third party tools such as Ghost.

Every computer must have a unique security ID (SID) in the network. If you duplicate the hard disks, it also duplicate the SID so all the computers in your network might have the same SID. To solve this kind of the problem, the System Preparation Tool sysprep.exe is introduced.

 

To install the System Preparation Tool sysprep.exe in your computer, create a directory Deploy folder such as c:\deploy in your computer. Browse the Windows 2000 Retail CD and find the deploy.cab in the support/tools directory. Double click the deploy.cab file and extract all the files to the deploy folder you have created. After that, use Windows Explorer to run the sysprep.exe program and you will see the following warning dialog box.

 

Figure 1.2.2.1 Windows 2000 System Preparation Tool

After performing the initial setup steps on a single system, you can run SysPrep.exe to prepare the sample hard disk for cloning. After that, you can use the disk duplication tools such as Ghost to clone the sample hard disk. SysPrep assigns a unique SID to each target computer the first time the computer is rebooted. Sysprep also creates a Mini-Setup wizard which runs the first time the target user reboots the machine. This shortened GUI-mode setup takes five to six minutes instead of 45 to 60 minutes and prompts the end user only for required and user-specific information, such as the End User License Agreement, Product ID, and user name and company.

The syntax for the sysprep is: Sysprep [Sysprep.Inf] [-quiet] [-nosidgen] [-reboot] [-pnp]

Where:

-quiet  suppresses confirmation dialogs displayed to the user.

-nosidgen  informs Setupcl.exe not to generate new SIDs on the reboot. This switch can be used if you want to make a duplicated copy of your current hard disk.

-reboot  forces the computer to reboot instead of shutting down (useful during auditing only).

-pnp  The Mini-Setup wizard will automatically detect new or different plug and play devices on the destination computer and disable those that were used on the master computer but not found on the destination computer; however, you may want to use the -pnp switch to enable legacy device detection or to re-enumerate the devices on the destination computer. Using this switch adds a few extra minutes to the end-user's experience in order to do a complete hardware redetection.

HAL (Hardware Abstraction Layer) is the translation layer in Windows 2000 that resides between the kernel and I/O system and the hardware itself. A uniprocessor system and a multi-processor system use the different HALs.

Because we can use ¨Cpnp switch when using Sysprep, the master and destination computers don¡¯t need to be totally identical. They just need to use the same HAL and mass storage controllers (IDE or SCSI).  Those plug and play devices such as modems, sound cards, network cards, video cards, etc. do not have to be the same, but drivers for those devices should be available.

Sysprep.inf is a file created with Setup Manager that we will discuss later to customize the Mini-Setup routine that will run the first time the target user reboots the computer. Sysprep.inf can be used to automate the Mini-setup. To automate the Mini-setup, you can either put the Sysprep.inf into the  Sysprep folder on the system partition of the master computer, or save the Sysprep.inf to  a floppy disk, and insert the disk at the beginning of Mini-Setup.

 

1.2.3 Create unattended answer files by using Setup Manager to automate the installation of Windows 2000 Professional.

Microsoft might be happy for you to deploy several hundreds Windows 2000 Professional computers in your network. However it would be tedious work for the administrator who needs to respond to every prompt from every installation while Windows 2000 is being installed. Unattended installation saves a lot of time and effort if you have to install a large number of Windows 2000 Professional computers.

Before we discuss the unattended installations, let¡¯s review the general installation process for the Windows 2000 Professional. If you have the Windows NT or Windows 98 running, you can run winnt32.exe for the installation of Windows 2000 Professional. If you start the computer using the MS-DOS, you must run winnt.exe for the installation.

 

Unattended installation is the method by which the administrators install Windows 2000 in unattended mode. It might take a little more preparation to begin the unattended installation, but it saves your overall time because after the unattended installation, you do not need to response to any prompts when Windows 2000 is being installed. As we have learnt before, if you want to install Windows 2000, you can run winnt.exe or winnt32.exe on the command line.

 

To start Windows NT Setup in unattended mode (as an unattended installation), you must specify the following command from the i386 directory on the Windows 2000 Retail CD.

   WinNT[32] /u:<answer file> /s:<install source>

Where <answer file> contains the information to automate the installation process and <install source> specifies the source location of the Windows 2000 installation files. For example, i386 directory in Windows 2000 retail CD. You can also copy all files in i386 directory to the hard disk and specify the corresponding location in the hard drive. <Answer file> is referred to as the Unattend.txt file that we will discuss later.

If you compare the regular installation and unattended installation, you will find:

-          Both regular installation and unattended installation use winnt.exe or winnt32.exe

-          Unattended installation needs to specify the source location of the Windows 2000 installation files (or run winnt.exe/winnt32.exe from where the installation files located) and the answer file.

In fact, three files are important for the unattended installation. They are unattend.txt, UDF and sysdiff.exe.

 

The Unattend.txt file answers the prompts during the installation. Unattend.txt file is also called the Windows 2000 Professional Setup Script File.  A sample unattend.txt file is on the Windows 2000 Professional Retail CD and provides a solution for the simplest installation. Because different hardware requires different unattend.txt file, if you want to install Windows 2000 Professional on both the laptop and the desktop computers, at least two unattend.txt files are required.

An answer file usually consists of section headers, parameters, and values for those parameters. Most of the section headers are predefined (although some may be user-defined). It is not necessary to specify all the possible parameters and keys in the Unattend.txt file if the installation does not require them. The file format is as the following:

 [section1]

      ; Section contains keys and the corresponding values for those keys/parameters.

   ; keys and values are separated by "=" signs

   ; Values usually require double quotes "" around them

     key = value

 [section2]

   key = value

 

A more complicated unattend.txt file can be created using Setup Manager. Setup Manager (setupmgr.exe) is located in the \SUPPORT\DEPTOOLS\ directory on the Windows 2000 Professional retail CD.

Setup Manager can create the unattended installation file not only for the Windows 2000 unattended installation, but also for the Sysprep install and Remote Installation Services. (Figure 1.2.3.1)

 

Figure 1.2.3.1 Setup Manager

 

UDF (Uniqueness database file) is used to specify the difference between different installations, (such as the computer name and the user name). Although unattend.txt file can answer all the prompts during the installation, some information such as the computer name must be unique for each computer.

You can use /UDF option with either winnt.exe or winnt32.exe to specify the UDF file. Follow the below format:

/UDF:ID[,database_filename]

Winnt /s:d:\i386\ /u:unattend.txt /UDF:id1,udf1.txt

Means you install Windows 2000 Professional from d:\i386 directory and use unattend.txt as the unattended answer file. You also use udf1.txt as the UDF file, which contains a unique ID of id1.

 

Sysdiff is used to duplicate applications for installation. The System Difference tool enables you to distribute and install applications automatically during or after Windows 2000 setup.  This can significantly reduce deployment time and costs.  It can be used to record the changes made to your system when an application is installed; capture those changes in a "package" and then "apply" or install the package on another system during or after the setup process. Sysdiff takes a snapshot before and after installing the applications on a specific computer and creates a difference file that contains binary files for the applications. Moreover, sysdiff also initializes file and registry setting for applications.

 

Windiff is used for NT 3.51 for the unattended installation. Windows NT 4.0 and Windows 2000 replaces it with sysdiff.exe. Because this is the Windows 2000 Professional exam, understand that windiff as an answer is a trap set by Microsoft to trip you up on those unattended installation questions. The command Windiff is used in Windows 2000 Professional to compare two files or two directories.

 

Unattended installation using a bootable CD-ROM is somewhat different than using the winnt.exe or winnt32.exe command. In order to accomplish this task, the answer file (unattend.txt) must be renamed to Winnt.sif and saved to a floppy disk. Moreover, the answer file must include a [Data] section with the appropriate parameters. The UnattendedInstall value must be set to yes. The MSDosInitiated value must be set to no. The AutoPartition value must be set to 1. If any of these parameters are set with incorrect values, then Setup either will not continue or will prompt the user for input. Setting OEM Preinstall to yes instructs Setup to copy the contents of the OEM subfolders, which are located in the distribution folder, to the target computer.

The UDF file cannot be used with this method because you don¡¯t have a chance to specify the ID for the UDF file. The bootable CD-ROM method can be used only for the clean installation. To perform an upgrade, you must run winnt32.exe on the operating system that needs to be upgraded.

 

1.3 Upgrade from a previous version of Windows to Windows 2000 Professional

Microsoft will be happy if you upgrade your Windows 98, Windows 98 and Windows 2000 workstation computers that use compatible hardware to Windows 2000 Professional. Windows 2000 supports a new directory service called Active Directory, which was not supported in Windows 95/98/NT computers. If your computer uses incompatible or insufficient hardware, you can install the Windows 2000 directory service client on those systems and take advantage of the functionality of active directory services.

 

Before installing the directory service client on Windows 95/98/NT, you must have Internet Explorer 4.01 or later installed. In a nutshell, if your computer meets the Windows 2000 requirements, you can upgrade your computer. Otherwise, you can install directory service client on your computer.  To install the directory service client, insert the Windows 2000 Server retail CD, and run the Dsclient.exe command in the clients\win9x directory. Most client computers running older version of Windows can be upgraded to Windows 2000 directly. However, Microsoft is a lazy guy that did not provide the direct upgrade from Windows 2000 3.1 and Windows 2000 3.5 computers.

 

To upgrade Windows 2000 3.1 and Windows 2000 3.5 computers to Windows 2000, you must upgrade Windows 2000 3.1/3.5 to Windows 2000 3.51 or Windows 2000 4.0 first, then upgrade to Windows 2000.  However for Windows 95, Windows 98, Windows 2000 3.51 and Windows 2000 4.0, you can upgrade them to Windows 2000 directly.

 

To see if your hardware and software meet the Windows 2000 upgrade requirements, you¡¯d better generate a compatibility report before performing the upgrade. To generate a compatibility report, you can either

1.       Run winnt32 /checkupgradeonly

This command checks your computer for upgrade compatibility with Windows 2000. For Windows 95 or Windows 98 upgrades, Setup creates a report named Upgrade.txt in the Windows installation folder. For Windows NT 3.51 or 4.0 upgrades, it saves the report to the Winnt32.log in the installation folder.

2.       Run chkupgrd.exe

You can download it from http://www.microsoft.com/windows/downloads/default.asp

Most old applications can run on the Windows 2000 system. There are three kinds of the applications are not compatible with Windows 2000, you need remove them before the Windows 2000 installation.

1.       Third party networking protocol and client software that doesn¡¯t have upgrade in the i386\winntupg folder.

2.       All anti-virus applications. That¡¯s because Windows 2000 upgraded NTFS version 4.0 to NTFS version 5.0. Most anti-virus applications are tied up with the file system.

3.       Customized power management tools. That¡¯s because Windows 2000 uses new technologies such as ACPI (Advanced Configuration and Power Interface) and APM (Advanced Power Management) for power management.

 

1.4 Deploy service packs.

Service pack is a software patch that is applied to the operating system. Before Windows 2000, you must deploy Windows operating system and service packs separately. Sometimes you need to reinstall the components previously installed. However Windows 2000 supports service-pack slipstreaming, which means the service pack can be intergraded with the Windows 2000 installation files.

The service packs can be downloaded from Microsoft¡¯s website. You can run Windows Update from start menu or Internet Explorer to connect to the website that you can download the service packs.

To slipstream a service pack, copy Windows 2000 installation files to a network share. Run update /slip to replace the Windows 2000 installation files. After that, you can connect the client computers to the network share to perform the installation.

To install service pack after installing Windows 2000, you can run update.exe to replace the existing Windows 2000 files with the appropriate files from the service pack.

Home Page | Greetings | Web Design Notes | Computer Help Notes | Visit Calgary | Contact Tom

Web Pages and HTML Sources are ©Copyright 2002 T. McLaughlin & Associates.
All rights reserved. All ©copyrights, ®registered trademarks,™ trademarks,
used on these pages belong to their respective owners.