Sams
Scenario/Problem: For any operating system (OS), hardware must meet minimum requirements in order to run the OS, and in a production environment, your hardware needs to meet at least the optimal requirements. You need to assess whether the hardware your organization owns will meet or—better yet—exceed the minimum requirements for the Windows Server 2008 OS.
Solution: Determine what the minimum, recommended, and optimal requirements are for Windows Server 2008 and compare your findings with your hardware.
When determining whether you have the hardware requirements needed to install and run an OS, you need to focus on three hardware resources:
Memory:
- Minimum: 512MB
- Recommended: 1GB
- Optimal: 2GB
Processor:
- Minimum: 1Ghz
- Recommended: 2Ghz
- Optimal: 3Ghz
Disk space needed for system partition:
- Minimum: 10GB
- Recommended: 40GB
- Optimal: 80GB
Note – The recommendations take into consideration only what is needed to run the OS. You need to also determine whether any applications are going to run on the server and include any resources requirements for those applications.
Note – Keep in mind that if you increase your memory above 16GB, you will need to increase your disk space requirements to accommodate for the pagefile (if kept on a system partition), hibernation, and the dump file.
Note – You can use a tool called the Microsoft Assessment and Planning Toolkit (MAP) to inventory your servers and generate a report to help determine which servers will work for your Windows Server 2008 installations. At the time of this writing, the tool is located at https://technet.microsoft.com/en-us/library/bb977556.aspx.
Perform Other Pre-Installation Tasks
Scenario/Problem: When you know that your server memory, processor, and disk space meet the requirements for Windows Server 2008, you need to perform some other recommended tasks before you actually install the Windows Server 2008 OS. What are these other tasks?
Solution: The following is a list of the tasks that should be performed before the actual installation.
Check application compatibility
Disconnect the uninterruptible power supply (UPS)
Run the Windows Memory Diagnostic tool
Identify mass storage device drivers
Back up servers
Disable virus protection software
Prepare Active Directory
Some or all of these tasks are recommended, depending on the path of installation and whether this is a new installation or an upgrade from an existing OS; in addition, you need to perform the Active Directory prep only if you are going to promote your Windows Server 2008 machine to a domain controller and add it to an existing Windows 2000/2003 domain.
Now let’s take a closer look at each of these tasks.
Check Application Compatibility
Before you install Windows Server 2008, you must be sure that any third-party applications you plan to run on the server will be supported. One way you can do this is to contact the application vendor and get documentation on whether the application will run on Windows Server 2008. In a real-world environment, the documentation is very important because if things do not work as expected, you may be able to save your job by providing the documentation. (Obviously, you would have tested the application on a development server first.)
Another tool that you can use is the Microsoft Application Compatibility Toolkit 5.0 (ACT 5.0). This tool can be used to collect compatibility data about your environment into a centralized data store. Having this information can be essential when evaluating the risk involved with an OS upgrade.
Note – ACT 5.0 can also be used for lower-impact changes to your platform, such as a browser upgrade or a Windows Update release. Check the following site for more details: https://technet.microsoft.com/en-us/library/cc507852.aspx.
Disconnect the UPS
During the installation process, Windows Server 2008 attempts to detect devices attached to serial ports. If you have a UPS connected to a serial port, you may run into issues with the installation, so be sure to disconnect it until the installation is complete.
Run the Windows Memory Diagnostic Tool
You can use the Windows Memory Diagnostic tool to test the random access memory (RAM) on your server. At the time of this writing, you can download this tool and a guide from https://oca.microsoft.com/en/windiag.asp. After you download the tool, you can perform the following steps:
Run the downloaded file mtinst.exe to start the setup for the Windows Memory Diagnostic tool.
Choose Create Startup Disk to install Windows Memory Diagnostic onto a floppy disk or choose Save CD Image to Disk to use a CD-ROM to which you can boot the server.
Reboot the server to the disk you just created.
The server will boot to the Windows Memory Diagnostic tool interface and automatically start the first test. It will continue to run tests with the same settings until you exit or pause.
Note – To run a more thorough test on the memory, you can choose to run the extended test suite by pressing T while the Windows Memory Diagnostic tool is running. If you do this, it would be best to leave the tool to run overnight.
Identify Mass Storage Device Drivers
If a vendor has supplied a driver file for your storage device, now would be a good time to have that file stored on a floppy, a CD, a DVD, or a flash drive. You should store files either in the root directory or in a folder named according to the processor architecture. During the installation, you will have the opportunity to load this driver.
Back Up Servers
Backing up servers is standard procedure when making any platform changes. Make sure you have a good backup of any critical data. When performing an OS upgrade, it is a good idea to make sure you have a backup of the boot and system partitions as well as the system state data. An alternative way to back up this configuration data is to create a backup set for Automated System Recovery (ASR).
Note – You should consider this recommendation if you are planning to upgrade an existing OS.
Disable Virus Protection Software
Virus protection software can affect the speed of your upgrade. Every file that is copied to your server will need to be scanned.
Note – You should consider this recommendation if you are planning to upgrade an existing OS.
Prepare Active Directory
There are two steps in preparing the Active Directory service for a new Windows Server 2008 domain controller:
Prepare the forest.
Prepare the domain.
Note – You need to prepare Active Directory only if you are going to build a Windows Server 2008 domain controller that will be joined to an existing Windows 2000/2003 domain.
Let’s first go through the steps to prepare the forest:
Log on to the Schema Master of your existing domain with an account that is a member of either the Enterprise Administrators, Schema Administrators, or Domain Administrators group.
Copy the adprep directory from the sourcesadprep on the Windows Server 2008 installation CD to the schema master.
From a command prompt, navigate to the adprep folder you just copied. Then run adprep/forestprep.
For a read-only domain controller (RODC), run adprep/rodcprep.
Wait for the task to complete and replicate prior to running the second portion of the ADS preparation.
When you have waited for the changes to replicate, you can follow these steps to prepare the domain:
Log on to the infrastructure master of your existing domain with an account that is a member of the Domain Administrators group.
Copy the adprep directory from sourcesadprep on the Windows Server 2008 installation CD to the infrastructure master.
From a command prompt, navigate to the adprep folder you just copied and then run adprepdomainprepgpprep.
Wait for the task to complete and replicate.
Now that you have completed some or all of the pre-installation tasks, you can start with the installation of Windows Server 2008. But first you must decide which edition of Windows Server 2008 you need for your environment.
Decide What Edition of Windows Server 2008 to Install
Scenario/Problem: Many different editions of Windows Server 2008 are available. The various editions allow support on x86, x64, and Itanium processors and also allow for native high availability, load balancing, and virtualization. You need to review all the various editions and decide which one best fits your organization’s needs.
Solution: You need to take a close look at each of the available editions of Windows Server 2008 and evaluate them in terms of your organization’s infrastructure goals.
The available editions are as follows:
Windows Web Server 2008
Windows Server 2008 Standard
Windows Server 2008 Standard without Hyper-V
Windows Server 2008 Enterprise
Windows Server 2008 Enterprise without Hyper-V
Windows Server 2008 Datacenter
Windows Server 2008 Datacenter without Hyper-V
Windows HPC Server 2008
Windows Server 2008 for Itanium-Based Systems
Note – This chapter does not include descriptions of Windows Server 2008 Standard, Enterprise, and Datacenter without Hyper-V as these are identical to their corresponding counterparts with Hyper-V. However, editions without the Hyper-V role are available.
Windows Web Server 2008
The title really speaks for itself: This edition is built for a single purpose, as a web server. Windows Web Server 2008 comes with architectural enhancements included within IIS 7.0, ASP.NET, and Microsoft .NET Framework. This edition is used to deploy web pages, web sites, web applications, and web services.
Windows Web Server 2008 supports the following:
32GB RAM on 64-bit (4GB on 32-bit)
Four multicore processors
Windows Server 2008 Standard
Windows Server 2008 Standard is a robust server OS that includes the following features to improve functionality, security, management, and reduce infrastructure costs:
Web services
Hyper-V (hypervisor-based virtualization)
Terminal Services
Presentation virtualization
Application virtualization
Network Access Protection (NAP)
BitLocker
RODCs
Windows Service Hardening
Bidirectional Windows Firewall
Next-generation cryptography support
Server Manager
Windows Deployment Services
Windows PowerShell
Next-generation TCP/IP
Server Core
Windows Server 2008 Standard supports the following:
32GB RAM on 64-bit (4GB on 32-bit)
Four multicore processors
250 network access service connections (RRAS)
50 network policy server connections
250 terminal server connections
Hyper-V virtualization with one free instance
Windows Server 2008 Enterprise
Windows Server 2008 Enterprise adds high availability, the latest in security, and scalability to the Standard edition. The following are some of its features:
Failover clustering (up to 16 nodes)
Fault-tolerant memory synchronization
Cross-file replication
Licensing for up to four additional virtual server instances
Active Directory Federation Services (ADFS)
Advanced certificate services
Active Directory Domain Services (ADDS)
Windows Server 2008 Enterprise supports the following:
Eight processors
2TB RAM on 64-bit (64GB RAM on 32-bit)
Unlimited number of virtual private network (VPN) connections
Unlimited Network Access Service connections
Unlimited Network Policy Server connections
Windows Server 2008 Datacenter
This edition can be used for large-scale virtualization needs and added scalability for mission-critical applications in a large IT infrastructure. The following are some of the features of this edition:
Large-scale virtualization (Licensing allows you to add an unlimited number of virtual instances.)
Failover clustering
Dynamic hardware partitioning
Windows Server High Availability Program
Windows Server 2008 Datacenter supports the following:
2TB RAM on 64-bit(64GB on 32-bit)
64 x64 64-bit processors and 32 x86 32-bit processors
Unlimited virtual image use rights
Hyper-V based unlimited virtualization use
16-node failover clustering
Hot add/replace memory and processors on supported hardware
Fault-tolerant memory synchronization
Cross-file replication (DFS-R)
Unlimited Network Access Services connections (RRAS)
Unlimited Network Policy Server connections
65,535 terminal server connections
Advanced identity management
Windows HPC Server 2008
Used specifically for high-performance computing (HPC), this edition enables you to scale to thousands of processing cores. This is advantageous when you’re load balancing heavy workloads across multiple processors and need to manage and monitor your HPC environment for stability and health.
Windows Server 2008 for Itanium-Based Systems
Windows Server 2008 for Itanium-Based Systems allows you to run Windows Server 2008 on Itanium-based systems. Itanium-based processors have the ability to handle intensive computing needs of business-critical applications in an enterprise-level environment. An Itanium processor uses a whole new architecture, not just extending the 32-bit architecture to 64-bit, and it can thus be called a native 64-bit processor. Another feature of this processor is the Intel Explicitly Parallel Instruction Computing (EPIC) architecture, which improves performance of the processor through instruction-level parallelism, maximizing opportunities to execute instructions in parallel. Up to six instructions can be processed in parallel.
Windows Server 2008 for Itanium-Based Systems supports the following:
Dynamic hardware partitioning
Use of Itanium RAS (Reliability, Availability, and Scalability)
2TB RAM
64 Itanium processors or 64 cores
Hot add/replace of memory and processors
Eight-node failover clustering
Fault-tolerant memory synchronization
Licensing for unlimited virtual instances with a third-party virtualization product
Now you know what each edition of Windows Server 2008 has to offer. Say that you decide that you need to install the Standard edition. Let’s get started with the installation and see what your options are.
Install Windows Server 2008
Scenario/Problem: You have decided to install Windows Server 2008 Standard edition. You need to decide whether you are going to perform a manual installation or an unattended installation. There may be some servers that will need to be upgraded also.
Solution: The following sections look at the procedures for doing both a manual installation and an unattended installation. We will also consider what is involved in upgrading to Windows Server 2008 from an existing operating system.
Manual Installation
The Windows Server 2008 installation procedure has been streamlined. If you are familiar with the Windows 2003 Server installation, you may remember that during the installation, you were prompted to answer configuration questions. With Windows Server 2008, these prompts have been moved to the Initial Configuration Task Wizard, which appears when the installation is complete. The following is the only information you need to provide during the actual installation:
Language, currency, and keyboard layout information
A valid product key
Installation location
Which version of the operating system you are going to install (if no product key is entered)
Whether you are performing an upgrade or fresh installation
The complete setup for Windows Server 2008 requires only three stages:
Operating system setup, including key validation
Initial configuration tasks
Server Manager setup
Operating System Setup
Follow these steps to set up the OS:
Insert the installation CD and boot the server to the CD.
When you are prompted for language, time and currency, and keyboard format information, as shown in Figure 1.1, make the appropriate selections and click Next.
- Figure 1.1
Configuring language, time and currency, and keyboard information.
The Install Now option appears. If you are unsure of what hardware requirements are needed, you can click the link What to Know Before Installing Windows. You can also click the link to perform and repair the OS rather than perform a full installation.
Input your product key and check the box Automatically Activate Windows When I’m Online (see Figure 1.2). Click Next.
Figure 1.2
Providing a valid product key.
If you did not enter a product key in the previous window, you now have to choose which edition of Windows Server 2008 you will install and check the box I Have Selected an Edition of Windows That I Purchased (see Figure 1.3). If you did enter a product key, the installation program will be able to identify which edition of Windows Server 2008 you are going to install. Then click Next.
- Figure 1.3
Selecting the edition of Windows Server 2008 to install.
Note – Chapter 2, “Configure and Manage Server Core,” discusses in detail the installation of Windows Server 2008 Server Core.
Read the license terms and accept them by checking the box. Then click Next.
In the screen that now appears, you decide whether to perform and upgrade or a custom (advanced) installation of Windows. Because you booted from the installation CD, the Upgrade option is disabled (see Figure 1.4). Click Custom (Advanced).
Note – If you wanted to perform an upgrade, you would need to execute the installation procedure from within the original Windows OS.
On the next screen, decide where you want to install Windows and, if you have any third-party storage drivers, load them by clicking the Load Driver link (see Figure 1.5).
- Figure 1.4
The Upgrade option is disabled when you boot from the installation CD.
Figure 1.5
Loading third-party storage drivers and choosing where to install them.
Now the actual Windows installation takes place. You will see the progress of each step as it completes, by percentage. During the installation, the server will reboot multiple times. The installation will complete the following tasks:
Copying files
Expanding files
Installing features
Installing updates
Completing installation
When the installation is complete, change the administrator’s password before the first logon. When the password had been changed you are logged in to the OS. You have completed phase 1 of the installation.
Initial Configuration Tasks
Now that the OS install is complete and you have logged in to the Windows Server OS, the Initial Configuration Tasks Wizard appears (see Figure 1.6). There are three sections in this wizard:
Provide computer information
Update this server
Customize this server
The Initial Configuration Tasks Wizard.
So what configuration changes can you make in these different sections?
In the Provide Computer Information section, you can do the following:
Change the time zone.
Configure the network settings on your network interface card (NIC) interfaces. You can also assign static IP addresses, subnet masks, default gateways, and DNS/WINS server. In many environments, you will probably be teaming two NICs for a production data LAN (using third-party software) and have a separate NIC dedicated for backup data connected to a backup LAN. Alternatively, you can leave the setting to be automatically assigned by a DHCP server, assuming that you have a DHCP server configured.
Note – In a real-world environment, you will usually assign static IP addresses to infrastructure servers. If this is the case, you will need to have gathered this information along with valid IP addresses for the default gateway and for DNS and WINS servers prior to the installation, along with the new server name if you are held to a strict naming convention in your organization.
Supply a computer name for the server, along with domain or workgroup information.
You need to reboot the server for these changes to take effect.
In the Update This Server section, you can do the following:
Enable automatic updates and feedback.
Configure the download and installation of OS updates.
In the Customize This Server section, you can do the following:
Add the server role or multiple roles. When you select a role, a wizard takes you through the complete installation of that role. You can choose from the following roles:
Active Directory Certificate Services
Active Directory Domain Services
Active Directory Federation Services
Active Directory Lightweight Directory Services
Active Directory Rights Management Services
Application Server
DHCP Server
DNS Server
Fax Server
File Services
Network Policy and Access Services
Print Services
Terminal Services
UDDI Services
Web Server (IIS)
Windows Deployment Services
Add features. As with roles, when you select a feature, a wizard takes you through the installation of that feature. There are many features to choose from, as shown in Figure 1.7.
Selecting features you would like to install.
Note – With both the roles and features lists, if you highlight a role or feature, you see a description of each role or feature on the right side of the list. When you are selecting roles and features, keep in mind that you should install as few as possible or only items you are currently planning to use. If you install unnecessary roles and/or features, you will also install services and possibly open up ports that will not be used in production but will render the server less secure. Chapter 5, “Install and Configure Specific Server Roles,” provides a more detailed discussion of installing and configuring roles and features.
Enable Remote Desktop connections to the server.
Configure the OS firewall settings. By default, the firewall is enabled.
Now let’s move on to phase 3 of the installation.
Server Manager Setup
Server Manager (shown in Figure 1.8) gives you a complete overview of your server. When looking at the default details pane, you can see computer information, security information, and a summary of the roles and features installed. And at the bottom of the page, you see a resources and support section. On the left side of the window are many tools to help you add/remove and configure roles and features. You can also see options for diagnostics, configurations, and disk management. When you have completed your changes in Server Manager, your manual installation is complete.
Server Manager.
Note – We will take a closer look at Server Manager in Chapter 3, “Work with Server Manager.”
Unattended Installation
Now that you have completed the manual installation, let’s take a look at how you would go about performing an unattended installation. With Windows Server 2008, you use a unattend.xml file rather than an unattend.txt file; in fact, the unnattend.xml file also replaces the Sysprep.inf, Winbom.ini, and Cmdlines.txt files. The XML format has been adopted because it makes it easier to describe nested values, add new elements, and validate the answer file. You can open the unattend.xml file in Internet Explorer 5.5 and later to parse the .xml file and see if it is well formed. If the file is not formed correctly, Internet Explorer shows you where the errors are.
To run an unattended installation, you execute the setup.exe file with the unattend switch:
?C:>setup.exe /unattend:unattend.xml
The unattend.xml file contains the responses needed while running the setup.exe file. This file contains such information as computer name, acceptance of the End User License Agreement (EULA), installation disk information, and so on. You can also show or hide the user interface (UI) for each value that is set by using ShowUI flag = Yes/No. Let’s take a look at how the installation reacts when you use the ShowUI flag:
ShowUI flag = Yes and setting is specified in the unattend.xml file: Setup uses the setting specified in the unattend.xml file and shows the UI with this setting.
ShowUI flag = No and setting is specified in the unattend.xml file: Setup uses the setting specified in the unattend.xml file and does not show the UI.
ShowUI flag = Yes and the setting is not specified in the unattend.xml file: Setup shows the UI, with the default value, and the user can change this setting, if needed.
ShowUI flag = No and the setting is not specified in the unattend.xml file: Setup uses the default value and does not show the UI.
While performing an unattended installation over a network, the system installer must have access to the unattend.xml file. When the setup is started from removable media (CD or DVD), the setup program looks for the unattend.xml file in the following locations:
The current working directory
The root of the removable media where setup.exe was initiated
Other removable media, such as floppy disks, USB devices, or another CD or DVD
The syntax for the unattend.xml file is broken up into elements, and each element needs to be opened and then closed in the proper order (when nested). When this is achieved, it is a well-formed .xml file. There is only one root element,
The running of the unattend.xml file stops with an error message if any of the following is true:
The EULA has not been accepted
The product key is invalid
The install disk cannot be written to
An unattend.xml file with some syntax information.
Creating an unattend.xml file can be tricky, but when you have this file created, it can make your job much easier. There are some tools available on the web that can help you create these files. You can also get very creative by adding some scripting to your installations to automatically generate computer names that adhere to your naming convention as well as many other configuration options.
© Copyright Pearson Education. All rights reserved.
