#mindtap

Module Introduction

Summarize the different ways that Windows Server 2019 can be used within an on-premises or cloud environment

Explain the purpose and function of Windows virtual machines and containers

Outline the key features of Windows Server 2019

Identify the differences between Windows Server 2019 editions

Discuss the considerations necessary to plan for a Windows Server 2019 installation

Describe the concepts and processes used to perform a Windows Server 2019 installation

Outline common post-installation configuration tasks for Windows Server 2019

Identify the different virtualization configurations that can be used to explore Windows Server 2019 within an IT lab environment

Microsoft Windows Server systems serve a critical role in nearly all organizations today, from small businesses to large, multinational corporations. Windows Server 2019, Microsoft’s newest server platform, offers even more roles for servers, better security, easier server management, new desktop features, and more reliable computing than its predecessors.

This book is intended to give you a solid grounding in how to install, administer, and support Windows Server 2019 in a wide variety of different environments. In this module, you learn the different ways that Windows Server can be implemented within an organization, in addition to the key features and editions of Windows Server 2019. Additionally, you’ll learn how to plan for and install Windows Server 2019, as well as perform post-installation configuration tasks.

Many of the Windows Server 2019 features and supporting topics introduced within this module are properly covered in more depth throughout this book. They are introduced within this module so that you can select the correct Windows Server 2019 edition and installation features to support your environment.

1-1Using Windows Server 2019 within an Organization

Since the introduction of the IBM PC in 1981, Microsoft operating systems have been at the forefront of personal and business computing. By 1990, over three quarters of all personal computers (PCs) ran the Microsoft MS-DOS or Windows operating system on IBM-compatible hardware, and the term PC became synonymous with Microsoft.

During the rise of computer networks and the Internet in the 1990s, most organizations started using PCs to connect to other computers across a computer network to obtain access to shared resources such as files, databases, and printers. The PC connecting to the shared resource was called the client, and the computer sharing the resource was called the server. While any operating system can function as a server if it shares resources with other computers, specialized operating systems geared toward resource sharing on the network also became common during the 1990s. Microsoft released their Windows NT Server operating system in 1993 and it quickly became the standard server operating system that client PCs connected to within organizations. By 2000, Windows NT Server was renamed “Windows Server,” with new versions released every three to five years. In 2018, Windows Server operating systems were installed on over 80% of all servers located within organizations worldwide.

Servers that are located within organizations are referred to as on-premises servers. Today, many servers exist within data centers outside the organization and are accessed via the Internet. These servers are referred to as cloud servers and the collective of servers publicly available on the Internet is referred to as the cloud.

Server hardware has a different form factor compared to desktop PCs. Nearly all servers within an organization are housed within a rackmount case that is mounted alongside other servers on a vertical server storage rack. Consequently, we call these servers rackmount servers.

Each rackmount server in the rack may contain a different operating system (or multiple operating systems if virtualization software is used) and will connect to a shared monitor/keyboard/mouse. This shared monitor/keyboard/mouse often folds away into the rack for storage and is necessary for initial configuration tasks such as server installation. All other server administration is normally performed remotely from a PC running remote administration tools.

Most racks also contain one or more storage area network (SAN) devices, which provide a large amount of hard disk or solid state disk (SSD) storage for the servers within the rack, as well as one or more uninterruptible power supply (UPS) devices, which provide backup battery power to servers and SANs within the rack in the event of a power loss.

The minimum height of a rackmount server is 1.75 inches. Servers of this size are called 1U servers. (The letter “U” is short for unit.) Most 1U servers have up to two hard drives (or SSDs) and up to two processors. Other rackmount servers take up more than one spot on the rack and have a height that is a multiple of a 1U server. For example, a 2U server is twice as high as a 1U server and often contains up to four processors and eight hard disks (or SSDs). Rackmount servers rarely exceed 4U, but SAN devices are often 4U or more.

Figure 1-1 shows a sample server rack configuration that hosts three 1U servers (Web server, file server, and firewall server), two 2U servers (database server and email server), a 2U UPS, a 4U SAN, and a management station with a shared monitor/keyboard/mouse.

Figure 1-1A sample server rack

A single rackmount server may contain multiple smaller, modular servers. In this case, the modular servers are called blade servers.

If Non-Volatile Memory Express (NVMe) SSDs are used, then the number of SSDs within a rackmount server or SAN may be substantially higher due to the small physical size of NVMe devices.

Main content

1-1aUnderstanding Windows Server Virtualization

Virtualization is the process of running more than one operating system at the same time on a single computer. It has been used in various capacities since the dawn of computing in the 1960s. To implement virtualization, you must use software that allows the hardware to host multiple operating systems. This software, called a hypervisor, serves to handle simultaneous requests for underlying hardware efficiently. Type 2 hypervisors are designed to run on top of an existing workstation operating system (referred to as the host operating system). All additional operating systems (called guest operating systems or virtual machines) must access the hardware through both the hypervisor and underlying host operating system. Type 2 hypervisors are common today for software testing and development. For example, a software developer can test a specific application or Web app on a variety of operating systems without requiring separate computers. Many college technology courses today also take advantage of Type 2 hypervisors to run multiple operating systems within a classroom or lab environment.

Common Type 2 hypervisors include VMWare Workstation, Oracle VirtualBox, and Parallels Workstation.

By the mid-2000s, a typical server closet or data center contained many individual rackmount servers. To maintain security and stability, each rackmount server contained a single (or small number of) separate server software applications. One rackmount server might host Web server software, while another might host file sharing services, and so on. Unfortunately, most of these server software applications only used a small fraction of the actual rackmount server hardware, and supplying power and cooling to the large number of rackmount servers was expensive. To solve these problems, many IT administrators turned to server virtualization, but with a Type 1 hypervisor to ensure that each virtual machine runs as efficiently as possible. A Type 1 hypervisor interacts with the hardware directly, and contains a small operating system to manage the hypervisor configuration and virtual machines. Figure 1-2 shows the difference between Type 1 and Type 2 hypervisors.

Figure 1-2Comparing Type 1 and Type 2 hypervisors

Microsoft Hyper-V is a Type 1 hypervisor. Other common Type 1 hypervisors used today include VMWare ESX/ESXi and Linux KVM.

Hyper-V is available on Windows Server 2012 and later, as well as on Windows 8 and later (Professional and Enterprise editions).

Nearly all hypervisors today require that your processor supports hypervisor acceleration; this feature is referred to as Intel VT (for Intel) or AMD-V (for AMD). Most hypervisors, including Hyper-V, also require that your processor supports Second Level Address Translation (SLAT) extensions.

Regardless of the hypervisor used, all virtual machines store their configuration within a small configuration file specific to the hypervisor. Meanwhile, they store the actual operating system data for the virtual machine in a virtual hard disk file. When you create a virtual machine, you must choose the size of this virtual hard disk file. You must also choose between allocating a fixed space for the virtual hard disk file when it is created (called thick provisioning) or dynamically allocating space as the virtual machine needs it (called thin provisioning). For example, if you create a 250 GB fixed-sized virtual disk, then 250 GB is reserved on the storage device immediately. Thin provisioning creates a small virtual disk file that grows up to 250 GB as the virtual machine stores more data. Thin provisioning is often preferred for server virtualization as it conserves space on the underlying server storage hardware.

Hyper-V virtual hard disk files have a .vhdx extension, VMWare virtual hard disk files have a .vmdk extension, Oracle VirtualBox virtual hard disks have a .vdi extension, and KVM virtual hard disks have either a .qcow2 extension or omit the extension altogether.

Microsoft supports a wide range of operating systems for installation within Hyper-V virtual machines, including Windows 7 and later, Windows Server 2008 and later, Linux, and FreeBSD UNIX. Additionally, Hyper-V can emulate slower, legacy hardware for older operating systems (called a Generation 1 virtual machine) instead of modern hardware for newer operating systems (called a Generation 2 virtual machine).

Hyper-V also supports virtual machine checkpoints (often called snapshots in other hypervisors). If you take a checkpoint of a virtual machine, it creates a second virtual hard disk file that stores any changes to the operating system after the time the checkpoint was taken. This is useful before testing a risky software configuration; if the software configuration fails, the checkpoint can be used to roll back the operating system to the state in which it was before the software configuration was applied.

Most on-premises and cloud operating systems today are virtual machines, and the virtual hard disk files that contain each virtual machine operating system are often hosted on a SAN within the organization or cloud data center. This configuration reduces the number of storage devices needed within the rackmount servers on the rack, which in turn reduces the space needed to host the server hardware. Consequently, most servers today that run a hypervisor and virtual machines are 1U, allowing the rack to accommodate more rackmount servers in the same space. For example, virtualization could be used to consolidate the five sample server operating systems shown in Figure 1-1 into two 1U servers running Hyper-V and five virtual machines, as shown in Figure 1-3.

Figure 1-3A sample server rack that utilizes virtual machines

Unlike most other hypervisors today, Hyper-V allows you to run other virtual machines within an existing virtual machine. This feature, called nested virtualization, is shown in Figure 1-4. For nested virtualization to work, both management operating systems illustrated in Figure 1-4 must run either Windows 10, Windows Server 2016, or Windows Server 2019.

Figure 1-4Nested virtualization using Hyper-V

Nested virtualization gives cloud data centers the ability to implement a more complex virtualization structure that suits their needs. Additionally, nested virtualization gives software developers and IT administrators the ability to implement a complex virtualization structure on their Windows 10 PCs for learning and testing purposes. For example, you can install a Windows Server 2019 Hyper-V virtual machine on a Windows 10 PC, and then install additional Hyper-V virtual machines within the Windows Server 2019 virtual machine.

Nested virtualization must be enabled in the underlying Hyper-V using a Windows PowerShell command (Windows Powershell is discussed in the next section). For example, to ensure that the virtual machine named VM1 can install and use Hyper-V to create additional virtual machines, you can power off VM1 and run the following command within Windows PowerShell on the management operating system:

After you power on VM1, you will be able to install and configure Hyper-V on VM1 to host additional virtual machines. To ensure that virtual machines created by VM1 can access the network connected to the underlying Hyper-V hypervisor (to access the Internet, for example), you also need to run the following command within Windows PowerShell on the management operating system to allow MAC address spoofing:

1-1bUnderstanding Windows Containers

Although virtualization makes more efficient use of server hardware, each virtual machine running on a hypervisor is a complete operating system that must be managed and secured like any other operating system running exclusively on server hardware.

Unlike virtual machines, containers do not have a complete operating system. Instead, a container is a subset of an operating system composed of one or more Web apps and the supporting operating system files needed by those Web apps only. As a result, containers must be run on an existing operating system that has container software installed, as shown in Figure 1-5.

Figure 1-5Using containers to run multiple Web apps

When you run a container, the enclosed Web apps are executed in a way that is isolated from Web apps running within other containers and the underlying operating system; this type of execution is often called sandboxing. To allow each Web app to be uniquely identified on the network, each container functions as a virtual operating system with a unique name and IP address.

Although separate virtual machines could instead be used to run Web apps on the same computer in an isolated fashion, containers are much smaller and use far fewer underlying system resources as a result. This makes containers well-suited for cloud environments, where resource efficiency and scalability are important for controlling data center costs.

Say, for example, that you create an IoT device that can be controlled remotely from a Web app running on a cloud server, and that you plan on selling thousands of these devices to customers. In this case, you don’t need to create a large, complex Web app that is designed to connect to thousands of devices simultaneously, and that is hosted within a large virtual machine on a cloud server. Instead, you can create a small, simple Web app that can connect to a single device, and run that Web app within a container that can be run thousands of times on a cloud server. When a customer connects to their device, a new container is run on the cloud server to start a unique copy of the Web app for that customer’s device. Similarly, when a customer disconnects from their device, the cloud server stops running the customer’s container to free up system resources.

The most common container software used to implement containers on operating systems today is Docker, and the underlying component within Windows Server 2016 and later that allows you to install and use Docker is called Windows Containers.

Docker is also available for many non-Windows operating systems, including Linux, UNIX, and macOS.

The core component of an operating system that executes all other components of the operating system is called the kernel. Containers do not contain a kernel, and thus must rely on the kernel in the underlying operating system to execute Web apps that they host. This means that the three Web apps shown in Figure 1-5 must be written for the Windows operating system and run within a Windows container if they are to use an underlying Windows operating system kernel for execution. It also means that the underlying operating system kernel is a single point of failure; too many containers on a single underlying operating system may slow down the performance of the kernel or cause it to crash. Furthermore, one container could potentially access another container running on the same underlying kernel if a security loophole were exploited.

To solve these problems, Hyper-V can be used alongside containers to provide a separate copy of the underlying kernel to each container. These Hyper-V containers use the functionality of Hyper-V alongside Windows containers to provide additional performance and security features to Web apps that are run within containers on Windows Server. Figure 1-6 illustrates the difference between Windows containers and Hyper-V containers.

Figure 1-6Comparing Windows containers and Hyper-V containers

We will examine the configuration of Docker, Windows containers, and Hyper-V containers within Module 10.

1-2Windows Server 2019 Features

Windows Server 2019 offers many features that make it a solid server and network operating system. This section examines the key areas of Windows Server 2019 that deserve special focus, including:

Active Directory

Security

Volume and filesystem features

Performance and reliability

Administration tools

Small footprint installation options

Hybrid cloud features

Linux application support

Each of these features is introduced in the sections that follow. You’ll learn more about these features as you continue through this book.

1-2aActive Directory

When you install any Windows client or Windows Server operating system, your computer is part of a logical grouping of computers on the network called a workgroup that is given the default name of WORKGROUP. Workgroups implement peer-to-peer networking as each computer (or peer) on the network maintains its own shared resources, users, groups, and security. When connecting to a shared resource on a computer that is part of a workgroup, you are first prompted to supply a valid user name and password on that computer that has permissions to access the resource (a process called authentication). If other users on the network want to access those same resources, they must also authenticate with a valid user name and password.

Unfortunately, this means that each user within a workgroup is responsible for the security of their own resources and must know how to set the proper permissions and security. Moreover, client computers are often not designed to handle a growing load of resource requests from other computers on the network. While workgroups are common in home and small office environments of fewer than 10 computers, they aren’t well suited for most organizations that frequently access resources on servers and other computers on the network.

To coordinate resource access, users, and security on networks today, you normally configure each client and server computer to join a domain instead of a workgroup. A domain is a logical grouping of computers that authenticate to a central database of users stored on special servers called domain controllers. When users log into a computer that is joined to a domain, their user names and passwords are authenticated on the nearest domain controller, which maintains a central database of users and passwords on the network. Once authenticated, the user receives a token from the domain controller that follows them around the network and automatically proves their identity to other domain-joined servers and clients. Those servers and clients will then allow the user to access resources that specifically grant them access. Because users only need to authenticate once to a domain controller to prove their identity to all domain members, this feature is called single sign-on.

The software components that provide for this functionality are collectively called Active Directory, and Windows Server can easily be configured as an Active Directory domain controller to provide single sign-on for other computers that are joined to a domain.

In addition to single sign-on for simplified resource access, Active Directory contains many other services and components that can be used to centrally manage and secure the computers that are joined to the domain. When you join a computer to a domain, groups that contain IT administrator user accounts (e.g., Domain Admins) are given administrative access to your computer. This allows IT administrators to easily configure any computer within the domain. Group Policy can also be used to configure operating system settings, security, and software for different computers and users in the domain, and Active Directory Certificate Services can be used to automate the configuration of deployment of encryption certificates to domain computers and users. Moreover, nearly all Microsoft server products (such as Microsoft Exchange Server and Microsoft SQL Server), as well as many third-party server products (such as Spiceworks) integrate with Active Directory for ease of management.

A single Windows Server computer can host several different roles on the network simultaneously. For example, a single Windows Server 2019 computer could be a domain controller, file server, print server, database server, and email server. Depending on the hardware capabilities, the server can handle hundreds of users at once, providing fast response when delivering the shared resource, and less network congestion when multiple workstations access that resource.

You can also purchase services hosted within Microsoft’s Azure cloud to provide Active Directory services to your organization. This is called Azure Active Directory and allows easier management of domains that span multiple locations around the world. Windows Server 2019 provides easier integration between Active Directory and Azure Active Directory.

1-2bSecurity

Windows Server 2019 is built to be even more secure than previous Windows Server systems. One important approach built into Windows Server 2019 is implementing security by default. When you install Windows Server 2019, add a feature, or install a Windows component, an essential level of security is automatically implemented. This helps to ensure that no backdoors are left open for an attacker.

Windows Server 2019 also has more protection against malware. Microsoft Defender monitors the server for malware and automatically stops processes that are known to be dangerous. If Microsoft Defender Advanced Threat Protection (ATP) is installed, any processes or files that are not known to be dangerous but look suspicious are automatically sent to servers within Microsoft Azure, a cloud computing platform, for deeper inspection. The servers in the Azure cloud use deep learning and sandboxing techniques to determine if a process or file is a new type of malware, and automatically updates ATP on all Windows Server 2019 systems worldwide with new information if it is. This allows Windows Server 2019 to more effectively combat serious security threats, such as zero-day attacks and root kits.

Microsoft Defender ATP requires an additional license beyond Windows Server 2019. Microsoft offers a free 12-month trial which you can access by searching on Microsoft Azure.

To ensure that malicious users cannot access the virtual hard disk files used by Hyper-V virtual machines, Microsoft added BitLocker encryption support for these files starting in Windows Server 2016. This feature is called shielded virtual machines, and originally required that the virtual machine ran Windows Server 2012 or later (virtual machines running Linux were not supported). With Windows Server 2019, you can now use shielded virtual machines to encrypt virtual machines running Linux, and the data transfer between virtual machines can now also be encrypted.

Windows Server 2019 also places more emphasis on security within the Internet Information Services (IIS) Web server software. Additional modules enable IIS to have a lower attack surface (vulnerable openings exposed to network attackers and malicious software), as well as allow modular security updates to quickly repair vulnerabilities that are found in the future.

Windows Server 2019 additionally includes many basic security features, such as:

File and folder permissions

Security policies

Encryption of data

Event auditing

Various authentication methods

Server management and monitoring tools

Main content

1-2cVolume and Filesystem Features

All servers store data on volumes (e.g., C:\) that are formatted with a filesystem. Like Windows Server 2016, Windows Server 2019 supports both the traditional New Technology File System (NTFS) available since Windows NT Server, as well as the new Resilient File System (ReFS).

NTFS is a very mature filesystem in terms of stability and features; it has built-in support for file and folder permissions, compression, Encrypting File System (EFS) encryption for individual files and folders, and user quotas that can be used to limit the space users are allowed to consume. It also supports data deduplication (or data dedup for short), which saves space by allowing duplicate files on a volume to be stored once on the physical storage device. In Windows Server 2019, you can now create NTFS large disk volumes of up to 8 PB (Windows Server 2016 only supports volumes of up to 256 TB).

NTFS is also a journaling file system, which means that it tracks changes to files and keeps a record of these changes in a separate log file. Journaling can be important, for example, when the computer crashes due to a power failure in the middle of updating or changing files. The logged journal information makes it possible to restore a file to its original condition prior to the power failure. Journaling enhances both the security and reliability of a system.

Starting with Windows Server 2008, NTFS volumes are self-healing. This means that when software encounters a damaged disk area, NTFS can heal the area without having to take down the server. NTFS self-heals by generating a “worker thread” that repairs data from the damaged area. The data is not available to the software until the worker thread completes its work. The next time the software wants to access that data, it is available for use. Prior to Windows Server 2008, a damaged disk area meant that you had to reboot the server into Safe Mode (offline to users) and run the chkdsk utility to rebuild the data in the damaged area.

NTFS also works very well in a virtualized environment because of the way that it works with file system cache. File system cache is an area designated in physical computer memory that is used in Windows operating systems to help speed up reading and writing to hard disk. The size of the allocated cache for NTFS operations can affect the speed of reading and writing, which in turn affects how long users have to wait on the server. In Hyper-V virtual machines, file system cache is dynamically allocated so that as memory is released by other applications or operating system processes, that memory is automatically used to increase the size of the NTFS file system cache for faster read and write operations.

ReFS was first offered with Windows Server 2012, but it is still regarded as a filesystem in development. It was originally developed as a filesystem that could fix filesystem writing errors automatically, eliminating the need for utilities such as chkdsk. While some NTFS features, such as EFS encryption, are not supported on ReFS, it excels over NTFS for use within Storage Spaces. Storage Spaces is a feature first introduced in Windows Server 2016 that allows you to build large, fault-tolerant volumes that span multiple physical storage devices. The third-generation version of ReFS (ReFS v3) was made available in Windows Server 2019 and comes with additional performance and feature improvements, including support for data deduplication and dramatically improved performance with Storage Spaces.

Windows Server 2016 introduced the ability to replicate data between two different servers seamlessly—a feature called Storage Replicas. Storage Replicas now have better performance in Windows Server 2019 and are supported in more editions (discussed in the next section). Microsoft also released a new Storage Migration Service in Windows Server 2019 that simplifies the moving of data to newer systems, as well as to systems in the Azure cloud.

1-2dPerformance and Reliability

Several features make Windows Server 2019 reliable and powerful, including privileged mode, protected processes, multitasking, multithreading, processor scalability, and server clustering.

Privileged Mode and Protected Processes

The Windows kernel runs in privileged mode, which protects it from problems created by a malfunctioning program or process. In addition to privileged mode, Microsoft uses protected processes in Windows Server 2012 and later to improve stability and performance. A process is a computer program or portion of a program that is currently running. One large program might start several smaller programs or processes. A protected process cannot be influenced by a user or other processes on the system. Key processes on Windows Server 2019, including those that perform system maintenance or update a database, run as protected processes and cannot be interrupted prematurely.

Multitasking and Multithreading

Windows Server 2019 and other recent Windows systems take full advantage of the multitasking and multithreading capabilities of modern computers. Multitasking is the ability to run two or more programs at the same time. For example, Microsoft Word can print a document at the same time that a Microsoft Excel spreadsheet can calculate the sum of a column of numbers. Multithreading is the capability of programs written to run several program code blocks (called threads) at the same time. For instance, a Microsoft Access database query runs a thread to pull data out of the database, while another thread generates a subtotal of data already obtained.

The multitasking in Windows Server 2019 is called preemptive multitasking. This means each program runs in an area of memory separate from areas used by other programs. Early versions of Windows used cooperative multitasking, in which programs shared the same memory area. The advantage of preemptive multitasking is that it reduces the risk of one program interfering with the smooth running of another program, thus increasing reliability.

Processor Scalability

One of the reasons why Windows Server has become such a versatile and powerful operating system is that it can be easily scaled upward in processor capacity. As more capacity is needed, more processors can be added.

A physical processor is plugged into a processor socket on the motherboard of the computer. Windows Server 2019 can support up to 64 sockets for individual physical processors. Furthermore, one physical processor can house several logical processors. Each logical processor is called a core and can run its own executable threads. A physical processor might consist of eight cores, for example, that enable it to function as eight processors in one. Windows Server 2019 supports an unlimited number of logical processors, providing a large amount of scalability for data center environments.

Additionally, when a computer is functioning as a virtual server, each virtual machine can be set up to use logical processors as virtual processors. In this case, a virtual processor is a logical processor that functions for the use of a specific virtual machine.

Server Clustering

Clustering is the ability to increase the access to server resources and provide fail-safe services by linking two or more discrete computer systems so they appear to function as one, as illustrated in Figure 1-7. You can configure a server cluster to provide both increased speed (such that user access is spread evenly across each server in the cluster) and fault tolerance (if one server fails, the other servers in the cluster can still respond to client requests). Alternatively, you can configure a server cluster for fault tolerance only. In this configuration, one server actively responds to client requests, while another one only responds to client requests if the other one fails; a process called failover. Consequently, server clusters are often used for frequently accessed and mission-critical services, such as databases. You can also configure Hyper-V virtual machines to work within a server cluster, or use a server cluster to allow a large amount of disk storage across several different servers to be made available to users (a feature called Storage Spaces Direct), with failover for disk storage as well.

Figure 1-7A server cluster

The power of clustering is only as good as the tools used to configure it. Windows Server 2019 offers tools to:

Create a cluster configuration and test to ensure it is set up to accomplish the tasks for which it is intended

Migrate configuration settings from one cluster to another

Troubleshoot cluster problems

Set up and optimize the storage used in a cluster

Secure a cluster and enable it to use new network capabilities

1-2eAdministration Tools

Windows Server has several different administration tools that you can use to manage the servers and services on the network. The major administration tools used with Windows Server 2019 include Server Manager, Windows PowerShell, and the Windows Admin Center.

Server Manager

The Server Manager tool was originally introduced in Windows Server 2008, with each subsequent version of Windows Server adding even more enhancements. While Server Manager allows you to monitor and manage the configuration of the local server it is running on (as shown in Figure 1-8), it can also be used to centrally manage multiple servers on the network if you choose to add additional servers to the Server Manager interface. When you first log into a Windows Server with administrator privileges, Server Manager automatically starts by default, but can also be started from the Server Manager icon that is pinned to the Windows Start menu. In addition to performing basic server configuration following a Windows Server installation, Server Manager is often used to:

View computer configuration information

Add, remove, and configure server roles, features, and system properties

Configure networking and security (including a firewall)

Configure Remote Desktop

Troubleshoot service and performance problems

Configure and manage storage

Configure and manage Active Directory objects

Figure 1-8Configuring local server properties within Server Manager

While Server Manager is typically used to configure newer components and services introduced in Windows Server 2012 and later (e.g., Storage Spaces), most server configuration is still performed using Microsoft Management Console (MMC) snap-in tools that run outside of Server Manager and are specialized for a specific component or service (e.g., the Group Policy Management MMC snap-in tool). Luckily, you can easily access these tools within Server Manager by selecting the Tools menu and choosing the correct MMC snap-in tool from the list.

Windows PowerShell

Microsoft’s first operating system, MS-DOS, did not contain a graphical desktop. Instead, MS-DOS users had to type commands within a command line interface called a shell. This MS-DOS shell has since been available in every graphical Windows operating system, and is still available within Windows Server 2019 if you execute the Command Prompt app or run the cmd.exe command from the Windows Start menu. Unfortunately, the MS-DOS shell doesn’t offer the same standardization, rich command set, or advanced scripting functionality that UNIX and Linux shells offer. As a result, Microsoft created a new shell in 2006 called Windows PowerShell that provided these features to computers running Windows and Windows Server.

Windows PowerShell is available for Windows Server 2003 and later, as well as Windows XP Service Pack 2 and later.

Commands within Windows PowerShell are called cmdlets and have an action-object (or verb-noun) structure. For example, the Get-WmiObject cmdlet shown in Figure 1-9 is used to obtain computer system information from the Windows Management Instrumentation (WMI) object called win32_computersystem.

Figure 1-9Running the Get-WmiObject cmdlet with Windows PowerShell

As shown in the Windows PowerShell title bar in Figure 1-9, Windows PowerShell is run as an Administrator account. While you can start Windows PowerShell by selecting the Windows PowerShell pinned icon from the Windows Start menu, many system configuration tasks that you perform require that you open Windows PowerShell with administrative privileges. To do this, you can right-click the Windows PowerShell pinned icon on the Windows Start menu and choose More, Run as administrator.

Nearly any task (administrative or otherwise) can be performed by cmdlets within Windows PowerShell. When you install additional software, cmdlets are often added to the system that can be used to manage the newly installed software. In fact, some tasks do not have a graphical configuration option and must be performed using cmdlets within Windows PowerShell. If you are logged into an Active Directory domain as a user with administrative privileges in the domain, you can also use Windows PowerShell to perform configuration tasks on other computers in the domain. Common tasks that are performed by administrators within Windows PowerShell include:

Working with files and folders

Monitoring and managing disk storage

Configuring network settings and troubleshooting network connectivity

Installing and managing software applications and server roles

Viewing configuration information for auditing and inventory purposes

Managing services and processes

Restarting multiple computers within a domain environment

Managing Active Directory users and groups

Windows PowerShell is also a powerful scripting language. You can also place cmdlets within text files (that have a .ps1 extension) alongside control structures that modify how the cmdlets are executed. These PowerShell scripts can then be executed to perform a series of tasks that can be re-executed periodically in the future as necessary.

Many system administration-related PowerShell scripts are available on the Internet. You can often download and repurpose these scripts to save time when performing system administration.

Windows Admin Center

New to Windows Server 2019 is a Web-based management tool called the Windows Admin Center. The Windows Admin Center is not installed on Windows Server 2019 by default, but can be downloaded from https://aka.ms/WindowsAdminCenter and installed afterward. Following installation, you can use a Web browser on any networked computer (only Chrome and Edge Chromium are supported at the time of this writing) to connect to the Windows Admin Center. For example, to connect to the Windows Admin Center installed on a Windows Server 2019 computer named win-v9cnv5rm2l6, you can enter the URL https://win-v9cnv5rm2l6 in your Web browser, and log in as a user with administrative privileges to obtain access to the Overview page shown in Figure 1-10.

Figure 1-10The Windows Admin Center tool

Microsoft also allows you to install the Windows Admin Center on a Windows 10 or Windows Server 2016 computer.

Like Server Manager, the Windows Admin Center can be used to manage other Windows Server systems (Windows Server 2008 with limited functionality, and Windows Server 2012 or later with full functionality). However, the Windows Admin Center cannot perform all Windows Server configuration tasks at the time of this writing (e.g., Active Directory configuration). Some of the administrative tasks that you can perform in the Windows Admin Center include:

Performing system updates

Displaying resources and resource utilization

Managing encryption certificates

Managing hardware devices

Viewing system events

Managing files and server storage

Configuring firewall and network settings

Managing installed software, processes, and services

Configuring local user accounts (not Active Directory)

Editing the Windows registry

Scheduling tasks

Managing Hyper-V virtual machines

Managing clusters

Managing Azure cloud integration and services

Obtaining a Windows PowerShell console or Remote Desktop connection

1-2fSmall Footprint Installation Options

You have the ability to install Windows Server 2019 with a minimal set of services, features, and functionality; this is commonly called a small footprint server installation. Because small footprint installations have fewer network services that other computers can interact with, there are fewer ways that a malicious user can use to gain unauthorized access to the server. In other words, small footprint installations have a smaller attack surface, and are more secure as a result. Additionally, small footprint installations use far less storage, memory, and processor resources on a server, and can be installed within a virtual machine or used as a container. This makes them more suitable for cloud environments, where thousands of servers are used to respond to client requests across the Internet.

Windows Server 2019 has two small footprint installation options: Server Core and Nano Server. Server Core retains much of the core Windows Server operating system but removes most of the graphical frameworks and processes. As a result, you typically configure Server Core using MS-DOS commands, PowerShell cmdlets, or configure remotely from a computer running Server Manager or the Windows Admin Center. To install Server Core, you can choose the appropriate option for your Windows Server edition from the installation wizard. When you boot your Server Core following installation, you will see a graphical login screen. When you log into this screen, you will obtain an MS-DOS command prompt that allows you to start Windows PowerShell (powershell.exe), or run MS-DOS commands such as the Windows Server Configuration Wizard (sconfig.cmd) shown in Figure 1-11. The Windows Server Configuration Wizard is a quick way to configure server settings (network, firewall, computer name, domain, and so on) or perform server administration tasks (reboot, install updates, and so on) from the command line.

Figure 1-11The Windows Server Configuration Wizard within Server Core

Many Microsoft and third-party software packages cannot be installed on Server Core because they rely on graphical .Net components that are not available in Server Core. However, you can download and install the Server Core App Compatibility Feature on Demand (FOD) framework to add many of these .Net components and allow more software to work on Server Core. The FOD framework also adds several graphical management tools back into Server Core, including

Microsoft Management Console (MMC)

Event Viewer

Performance Monitor

Resource Monitor

Device Manager

File Explorer

Disk Management

Failover Cluster Manager

Hyper-V Manager

Nano Server is an even smaller footprint server than Server Core. Without additional software, a base Nano Server consumes less than 500MB of disk space. Nano Server still has fundamental elements, such as the .Net Framework used to develop and execute applications, but it has few other components. As a result, only four server roles are supported on Nano Server:

DNS server (discussed later in this module)

DHCP server (discussed later in this module)

Web server (running custom Web apps)

File server

Consequently, Nano Server is ideal for use within cloud environments that require the hosting of Web apps and supporting services only. Although it is possible to install Nano Server directly on the hardware of a server, it is designed to be installed within a virtual machine or used as a container. If used as a container, Nano Server does not contain a Windows Server kernel. Moreover, most administration of Nano Server must be done from a remote computer using a subset of the Windows PowerShell cmdlets found in other Windows Server systems.

To install Nano Server, you use an existing Windows Server to generate a virtual hard disk file that contains Nano Server for use within a virtual machine, or you can download a pre-made Nano Server container from the Internet, as discussed in Module 10.

1-2gHybrid Cloud Features

With Windows Server 2019, Microsoft is adding more integration between on-premises Windows Servers and Windows Servers and services that run within the Azure cloud. This type of integration is often referred to as hybrid cloud. The Windows Server Azure Network Adapter allows you to easily connect your on-premises Windows Server systems with other virtualized Windows Servers and services in the Azure cloud using a secure connection. You can also use Azure Backup to back up important information to storage in the Azure cloud, Azure Update Management to centrally manage the updates for on-premises computers and virtual machines running in Azure, as well as Azure Site Recovery to ensure that key services that you run on-premises are made automatically available in the Azure cloud if you encounter a failure of an on-premises server. Moreover, hybrid cloud features can be configured and monitored using the Windows Admin Center.

If your organization deploys a large number of Windows Containers, Windows Server 2019 now has support for Kubernetes, which is a software product that can coordinate the execution and management of both on-premises containers, as well as containers hosted within the Azure cloud.

1-2hLinux Application Support

Most Web apps that run in the cloud run within Linux containers. As a result, Microsoft has made it easier for Web app developers to create and test Linux apps on their Windows 10 PCs using the Windows Subsystem for Linux (WSL). The WSL provides a Linux kernel interface for Linux apps that allows them to execute directly on the Windows kernel within a Linux operating system environment (including a virtual Linux filesystem and directory structure). In short, WSL allows developers to create and test Linux apps on their local PC before deploying those apps to a Linux virtual machine or container in the cloud. Windows Server 2019 now supports WSL, which means that you can run Linux containers on Windows Server, alongside Windows Containers.

1-3Windows Server 2019 Editions

Servers have a wide variety of uses designed to match the needs of a wide variety of different environments, from home offices and small businesses to international corporations and large data centers. Because one size does not fit all needs, Microsoft offers different versions of Windows Server 2019 called editions. Each edition is built on the same foundation but offers unique capabilities to suit the size and needs of an organization. Before you deploy Windows Server 2019, it’s important to first select the correct edition for your needs. In this section, we’ll examine and compare the features of different editions of Windows Server 2019. The major Windows Server 2019 editions include:

Windows Server 2019 Essentials Edition

Windows Server 2019 Standard Edition

Windows Server 2019 Datacenter Edition

However, there are two additional Windows Server 2019 editions, each customized for a specific use:

Windows Storage Server 2019

Microsoft Hyper-V Server 2019

1-3aWindows Server 2019 Essentials Edition

For a business or organization with up to 25 users, Windows Server 2019 Essentials Edition is a good, cost-effective option for providing most Microsoft server services to users. Windows Server 2019 Essentials Edition also comes with hardware and connection limits; you can’t install it on a server with more than 2 processor sockets or more than 64 GB of memory, and only 50 concurrent remote access connections are allowed.

Windows Server 2019 Essentials Edition cannot join an existing Active Directory domain, but can host a single, small Active Directory domain with a single domain controller. Furthermore, Windows Server 2019 Essentials Edition provides most but not all server roles. Most notably, it does not provide the Hyper-V role for hosting and managing virtual machines. However, Windows Server 2019 Essentials Edition can be installed as a virtual machine on an existing hypervisor, such as Hyper-V (included in higher editions).

Microsoft focuses on specific features in Windows Server 2019 Essentials Edition (most of which are also available on higher editions) to make it particularly attractive to small businesses and organizations. These features allow for the following:

User groups can be created to manage clients and client access to Microsoft Office 365.

Backups and restores can use file history information for each user instead of only for each device.

Size and growth of a server folder can be managed through a space quota.

Installation can be on a standalone physical server or as a virtual machine, which allows you to run Windows Server on an existing hypervisor.

Server Health Reports are automatically installed to be available at the time Essentials Edition is installed.

Mobile devices can be managed using Dashboard, which is a tool within Server Manager for simplified management of the server.

The BranchCache file sharing feature is available to allow fast data access to other Windows and Windows Server computers running an offsite location.

Table 1-1 compares the features found in Windows Server 2019 Essentials Edition to those in other major Windows Server 2019 editions.

Table 1-1

1-3bWindows Server 2019 Standard Edition

Windows Server 2019 Standard Edition is designed to meet the everyday needs of most businesses and organizations. It provides file and print services, secure Internet connectivity, centralized management of users, and centralized management of applications and network resources. It is built on technology from previous editions of Windows Server but includes many new features and security enhancements. Additionally, many features that existed in previous Windows Server versions have been further optimized for performance.

A company might use Windows Server 2019 Standard Edition to host its accounting and payroll software, provide network resources such as shared files and printers, deliver email functionality, or even manage users’ access to cloud application software, such as Office 365. Some other key features in Windows Server 2019 Standard Edition include:

A modern desktop user interface that matches that of the latest Windows 10 builds, including the ability to use virtual desktops

An improved Windows Defender that configured and enabled by default for malware detection and prevention

Easier configuration, management, and security options for applications, files, networking, and Active Directory (including the ability to clone domain controllers)

Desired State Configuration, which can be used to quickly configure multiple servers using a template file that lists required software and configuration items

Intelligent storage features for volumes that use multiple physical storage devices. This includes storage tiers (which automatically moves frequently-used data to faster storage devices, such as SSDs) and storage pinning (which allows you to force specific data to be stored on a particular device).

The ability to use Storage Replicas (which previously required Datacenter Edition in Windows Server 2016)

The ability to create up to two Hyper-V virtual machines under the default Windows Server license

The ability to create an unlimited number of Windows Containers, and up to two Hyper-V containers

When it comes to purchasing a license for Windows Server 2019 Standard Edition, Microsoft charges based on the number of processor cores that you run it on. In addition, each user connection requires a Client Access License (CAL). Microsoft gives you options to either purchase a set number of CALs per server, or a CAL for each client that accesses the servers within your organization.

1-3cWindows Server 2019 Datacenter Edition

Windows Server 2019 Datacenter Edition is designed for environments with mission-critical applications, very large databases, very large virtualization requirements, cloud computing needs, and information access requiring high availability. It shares the same hardware support, features, and licensing requirements as Windows Server 2019 Standard Edition but allows you to create an unlimited number of virtual machines under the default license, as well as an unlimited number of containers (including Hyper-V containers). Because Windows Server 2019 Datacenter Edition is typically used to host a larger number of virtual machines, it also comes with additional Software Defined Networking (SDN) features, such as the Network Controller role that can be used to monitor and manage virtual networks used by large numbers of virtual machines.

1-3dWindows Storage Server 2019

Windows Storage Server 2019 cannot be purchased directly from Microsoft. Instead, original equipment manufacturers (OEMs) that create storage solutions for organizations can offer Windows Storage Server 2019 on the server-based products they sell. Windows Storage Server 2019 turns a server into a central storage center for data in an organization and takes advantage of the storage utilities offered in Windows Server 2019. To learn more about the storage services, see Module 7, Configuring and Managing Data Storage.

1-3eMicrosoft Hyper-V Server 2019

Microsoft Hyper-V Server 2019 is the Hyper-V hypervisor. It also installs a small footprint version of Windows Server that contains a PowerShell interface for creating and managing virtual machines. While Microsoft Hyper-V Server 2019 is free to download and use in your environment, each Windows Server virtual machine that you run on it requires a valid license (Linux virtual machines, however, do not).

1-4Preparing for a Windows Server 2019 Installation

Just as a trip goes better with advanced planning, so does an operating system installation. You are likely to work with the operating system for some time, so it makes sense to get off on a solid footing to avoid problems later.

The first step in planning the installation of any operating system is to determine the hardware requirements. Most operating systems come with a list of minimum hardware requirements which must be met for the operating system to run and often a list of recommended requirements. For Windows Server 2019, these requirements are listed in Table 1-2.

Table 1-2

Minimum Hardware Requirements for Windows Server 2019

HardwareMinimum requirementsAdditional considerationsBasic Input/Output System (BIOS)Unified Extensible Firmware Interface (UEFI) 2.3.1c BIOS or higher for physical server installationTrusted Platform Module (TPM) is also required for secure boot and encryption features.Processor1.4 GHz 64-bit processor (includes support for NX, DEP, CMPXCHG16b, LAHF/SAHF, PrefetchW, EPT, or NPT)Processor clock speed, amount of processor cache, and number of processor cores should be considered based on planned usage. Hyper-V also requires processor virtualization extensions (Intel VT or AMD-V, with SLAT).Memory512 MB (2 GB for a server with the GUI desktop)Each virtual machine requires 800 MB for setup (although this can be scaled back after setup is complete). ECC memory is recommended for physical (non-virtualized) server installations.Storage32 GB32 GB is the minimum for Server Core, while 36 GB is the minimum for installing a full Windows Server.Network interface1 gigabit or faster Ethernet adapter that is compatible with PCI Express architecture and Pre-boot Execution Environment (PXE)Additional adapters are recommended if you support multiple virtual machines.Optical driveDVD drive (optional)While a DVD drive is needed for installations from DVD media, many administrators install from a USB drive today.DisplaySuper VGA at 1024 × 768 or higher resolutionMultiple servers can share one display via the use of a keyboard-video-mouse (KVM) switch.Interactive devicesKeyboard and pointing deviceMultiple servers can share a keyboard and pointing device via the use of a KVM switch box.

In general, you should always exceed the minimum hardware requirements listed for any operating system. Exceeding the minimum requirements also makes your server more scalable, allowing it to meet increased requirements as the organization grows. The amount that you plan to exceed should be determined according to the role the Windows Server 2019 system will play on the network, as well as the number of clients that you expect to connect to the server. For example, if your server will function as a file server, hosting things such as home folders and company-shared files and printers, additional storage space will be needed beyond what is required to simply run the operating system. The amount of storage space depends on the number of users that are going to store data on the server as well as the average size of the data per user.

Planning hardware for a specific server given its use within an organization is often called capacity planning. Here are some sample questions to consider when performing capacity planning for Windows Server 2019:

What role or roles will the server have in your organization? For example, is this server limited to file and printer sharing? Are you implementing a Web server, SQL database server, or email server? Will your server offer remote access or will it be a source of applications for users?

Do you need to deploy virtual machines, and if so, how many?

How many local and remote users do you need to support?

What kind of support do you need from the hardware vendor?

What redundancy features do you need to ensure the server continues running in the event of a hardware failure, such as a failed disk drive, power supply, or network interface?

What growth in server use and resources do you expect in the next 3 to 5 years?

These questions only provide a starting point for your planning. In a small business, much of the planning can be done with the help of the business owner. In medium and large businesses, the planning will likely require input from management, user departments, technical people, software providers, and hardware vendors. In either case, hardware should be planned based on the server role and projected growth, so as to exceed what is needed to:

Accommodate the clients that will access the server

Provide for extra software and services

Match data storage needs

In terms of processing speed, plan to pay particular attention to the number of processors, as well as their speed and number of cores. For example, a small business using one server with Essentials Edition may consider purchasing a 3 GHz processor with only two cores (two logical processors). A corporation that plans on running several large databases within several virtual machines may consider running Windows Server 2019 Datacenter Edition on a server that has two 4 GHz processors with 12 cores each, as well as a motherboard with two additional processor sockets to allow the server to scale up to a total of four processors in the future.

It’s also important to ensure that your server has enough memory for the applications that it will host, as well as free memory slots for future scalability. The small business mentioned earlier might start with 32 GB of memory with room to expand to 128 GB of memory, whereas the corporation mentioned earlier might start with 256 GB of memory with room to expand to 2 TB of memory.

The amount and type of storage is another important consideration. Most servers today come with multiple SSD storage devices for the operating system and core applications, configured to be fault tolerant using Redundant Array of Independent Disks (RAID) in the BIOS of the server. You can plan to add additional hard disks (large capacity, slower speed, lower cost) or SSDs (lower capacity, faster speed, higher cost) within the server to store the data used by your applications, or the virtual machine hard disk files used by your virtual machines. Alternatively, if your organization uses a SAN to store data and virtual machine hard disk files, you can simply ensure that your server can connect to the SAN. For iSCSI SANs this will require an additional iSCSI-capable network card, and for Fiber channel SANs this often requires a proprietary PCIe controller card. The size of each storage device is ultimately determined by your estimate of the demand for storage space on the server in the next year. It’s also important to ensure that the storage on your server can easily be expanded in the future, as organizations often consume more storage than what is originally planned for.

Before any final decisions are made in selecting hardware, you should check the hardware for the Certified for Windows Server 2019 sticker or consult the Windows Server Catalog, which is a list of compatible hardware devices for each version of Windows. You can access the Windows Server Catalog online at https://windowsservercatalog.com.