Windows NT is a very robust operating system. I have found that it will give many wonderful hours of service with minimal attention, once you have it set up properly. This chapter is devoted to that tricky subject of getting NT installed and configured
properly. Don't get me wrong; we are not talking about magic or dumb luck here. We are talking about engineering discipline and basic problem-solving skills. It often takes patience, as you have to try multiple combinations and wait many minutes in between
tries as your server reboots itself so that you can see the results of your actions. Eventually, you will have experience, which can help you to avoid most of the more common problems.
Some of you may feel disgruntled. Why should it be so hard to install an operating system? Why can't Microsoft make it so that it is easy? The complexity lies in the wide variety of products in the PC world. Everywhere you look there are a number of
computer vendors out there making every product imaginable. They do a lot of good things to push technology forward and keep prices down. However, sometimes their innovations call for some changes requiring alterations to the operating-system device
drivers that interface them to the operating system.
Although change can be good, it does pose a problem for those of us who are just trying to get and keep our computers operational. These vendors usually come out with drivers for the mass-produced operating systems such as Windows 3.1 or Windows 95.
However, these vendors have not always come out with Windows NT drivers. Part of this is due to the fact that Windows NT has never had the millions of users that Windows 3.1 or Windows 95 has. The other part came from the fact that the Windows NT drivers
are somewhat different from the Windows 3.1 drivers that vendors were used to and had programmed with.
The recent rise in Windows NT popularity may help to ease this lack-of-drivers problem. However, in the near term, you have to deal with the fact that Windows NT does not have drivers for every imaginable device that you may install in your computer. So
how do you get around this? The answer comes in the form of the hardware-compatibility list provided on the Microsoft Web page and some sound planning. This chapter is devoted to a final review of the plans that you have made and then takes you through the
installation process. I have divided this chapter into the following sections:
The key to a successful installation is having a sound plan. You probably have guessed my feelings on choosing compatible hardware (yes, I have mentioned it many times so far). There are also a few other things that you should consider when you look at
your plans that will help you along the way. This section is devoted to providing a final checklist and an example of an installation plan for your consideration. The topics that I thought to include are as follows:
Let me start off with the hardware list. You already know to verify that there are software drivers for the version of Windows NT that you are working with for all of your hardware devices. A general suggestion when configuring a server is to stick with
the more common hardware items, especially in the server environment. You do not typically have the large number of servers that you do with workstations. Therefore, saving a few hundred dollars by picking out less commonly used components for a server
will not save you all that much money. However, it could cost you a lot of time and money when it comes to getting these uncommon components to work with the Windows NT server. You do not have to be extravagant and only choose the most expensive
components. Just try and pick parts that are very common in the NT server environment (for example, Adaptec SCSI controllers and US Robotics modems) if you can at all justify the costs. Honestly, I am not working as a representative for the bigger, more
expensive product vendors. I am just relaying my experiences with trying to integrate no-name products into an NT-based server.
The next task in front of you is to develop your software list. This list is composed of the optional Windows NT components (for example, the Internet Information Server), other BackOffice products (such as Exchange Server) and third-party products that
will be installed on your server. You will have to choose which disk drives to install the products on. This can be a significant factor in overall system performance—for example, splitting indexes and tables on different disk drives can provide
significant performance improvements for database management systems—so you need to consider this carefully in consultation with other administrators (especially the database administrator) and perhaps even users. This list should reflect options that
are to be selected with the products and also the order in which the products need to be installed.
Your next big decision (and this is a big one) is the security configuration of your new server. As I mentioned earlier, you cannot just switch between a workgroup server and a domain server. You would have to reinstall the operating system and do a lot
of configuration work to get yourself back to a working system. The factors in favor of choosing the domain environment are easier maintenance, if you have a number of servers, and also the fact that many BackOffice components are prodding you in the
direction of domains. It could also be important to choose the domain architecture if you want to use some of the remote-administration tools that are available for Windows 95 and Windows NT workstations (one of which might be located on your desk and
thereby save you from having to do all of your work in the computer room).
Your next decision is what your network configuration will be. This includes the basics as to which protocols you will be loading and how each of these protocols is configured. If you are using TCP/IP, you will have to obtain a list of legal IP
addresses for your network. If you are planning to be connected to the Internet, you will have to obtain an official set of IP addresses from a NIC. If you are just using a local network, then you or your network-operations staff will have to decide on the
list of addresses. It can cause a number of strange conflicts if you have two PCs that are trying to use the same IP address, so spend some time sorting this out up front. Also, you have to decide on the optional network services (WINS servers, DHCP
servers, and so on) that you will be installing on your computer.
Next, you should plan out which resources you will be sharing with your user community. This starts with the list of resources and which share name they are going to be using. Typical shared resources include directories (whose name changes to
"folders" under NT 4.0), printers, Remote Access Service permissions, and access to other applications (such as an Oracle or SQL Server database). You should also spend some time now deciding which groups and users will be allowed to access these
shared resources, the hours of their availability, and so on.
Your next task is to decide what your user and group structure will be. It can start with a simple mapping of the groups. You can build on this with a more detailed explanation showing what the resource access privileges of the groups are. It is
important to take some time up front to think this process out. It can be a challenge to change the group structures around after the fact; your users usually will have to log off and log back on to gain access to new group privileges, so this can be a
pain if you are in a busy production environment. Getting your naming conventions (for example, your building designation or which department your work in as being part of the name) documented before implementing your network is much easier than after the
fact.
A big task in the installation procedure is choosing how you will install NT. In the next section, I will present the various alternatives for installing Windows NT. Although you have the one basic idea that the Windows NT software is on a CD-ROM and
you want to put it on your hard disk drive, there are several different ways of accomplishing this task. The method that you choose depends on which format you received the NT system in and the starting configuration of your server.
Finally, I would argue that it is useful in most cases to build a simple checklist of tasks and options before you begin the actual installation process. This does not have to be fancy or even word processed (although I type faster than I write so I
would typically use a word processor). It should be a means that can jog your memory to ensure that all of the details are taken care of. It is also a form of discipline to ensure that you think of everything in enough detail to put it down on paper (some
people think that they know what to do until they actually have to do it or write it down). So what would I consider a checklist to be? Figure 10.1 shows a sample checklist that I would consider a basic starting point (you can get fancier and more
detailed, but this is a good beginning).
FIGURE 10.1. Sample installation checklist.
These are some of the basics to be considered before performing a general NT Server installation. In the installation chapters for the other BackOffice components, you will find additional pointers and planning steps that you will want to incorporate
into this server installation plan for servers that contain these components. Third-party products (such as Oracle databases) also have their own installation needs, which must be factored into plans for their servers. The key point that I cannot stress
enough is that you need to do a little planning up front to ensure that you have enough space on all of your disk drives, you have arranged data directories to meet your user needs and optimize performance, and a host of other issues. It can be very
difficult to change the configuration of a server after it is turned over to the users.
This section actually has two purposes. I first want to go over the alternative installation procedures that are available to you with NT Server. The different paths have advantages and disadvantages, so you should be aware of what you are doing. The
next topic is the installation of Windows NT Server 4.0 using one of these scenarios. Will this be the exact scenario for you? Probably not exactly. Your exact path will vary slightly depending on the version of NT that you are installing and the
configuration of your machine (both hardware and the network/optional products that you select). However, it gets across the basic ideas and will be useful when you go to perform your own installations. The chapters that have led up to this chapter were
designed to prepare you to answer the questions that will come up (for example, which network protocols to use) in your specific installations.
So what are these installation options that you will be faced with? You may ask, "Why isn't being a system administrator ever simple?" Actually, the installation options are not all that bad, and they depend on two key factors: which
installation media you received and the current condition of your server. Windows NT comes on CD-ROM media. Microsoft has given up on the idea of trying to distribute a huge box of diskettes that contain all of the contents of this rather robust operating
system. It is not practical and I, for one, would not have the patience to change all of those diskettes. You can get a CD-ROM drive for a reasonable price these days, and it is worth it when you go to perform a massive installation such as NT. The only
question is whether you received a set of diskettes with your distribution. If you received only the CD, then you will want to make a set of diskettes so that you can boot your server from these NT diskettes in case you run into any problems.
The second variable that will determine your choice of installation procedures is the current state of your server. If you are starting out with a blank computer, you will need to start from a boot diskette, format the hard disks, and then install the
operating system. If, however, you are upgrading an existing NT server installation, all you want to do is install the new software. A slight variation on this theme would be a complete reinstallation of the operating system, which is required when you
convert a workgroup server to a domain controller.
Figure 10.2 shows the installation options that I will be discussing. You might be able to think up some other scenarios, but the basic ones that I will be covering are the following:
FIGURE 10.2. Windows NT Server installation options.
The first option, installation from DOS without NT diskettes, covers the situation in which you have a new server with empty hard disk drives.
To install from DOS without NT disks:
The first task that you might be faced with is preparing a bootable DOS diskette that has CD-ROM support. This can be a bit tricky if you received the machine with Windows 95 or NT workstation installed. Because these operating systems have their own
CD-ROM drivers, you do not load the old CD-ROM drivers using the autoexec.bat file (which you could use to determine the interrupt settings and device-driver files you would need for your boot diskette). Anyway, you can usually find all of the device
drivers and put together a diskette that can boot and access the CD-ROM. Test this diskette out thoroughly for its ability to access the CD drive before you begin the installation procedure. There are also a couple of key DOS utilities that I like to put
on these boot diskettes:
Perhaps I am living in the past, but I like to have one of these boot diskettes for my computers. They boot a really simple operating system that I can use to access the computer. There is one complication, though. These DOS diskettes do not have
drivers to read NTFS partitions. For this, you will need NT-based diskettes, which you have either received as part of your installation package or can make from your CD. Most of the box sets that I have received contain the diskettes, whereas
distributions such as those of the Microsoft Developer Network usually contain only the CD.
Anyway, after you have the computer booted into DOS mode, you have to set up your disk partitions (using fdisk) and format your drives, if this has not already been completed by the manufacturer. This basic process prepares the drives to receive data.
Note that these disks are formatted with the FAT file system. You will be given the option of upgrading from FAT to NTFS later in the installation process. When the drives are ready to receive data, you have to complete the following steps:
All of these tasks are kicked off by running the winnt.exe file that is located in the appropriate directory on the Windows NT Server CD-ROM (\i386 for the Intel-based PCs). The first task is to prepare the diskettes that can be used to boot the NT
operating system. This is not absolutely necessary for the installation process. You can use the command line winnt -b, which will install the boot files from the operating system directly to the fixed disk drive. However, if you ever run into problems
with your fixed disk drives, you will want to have a set of bootable NT diskettes available to enable you to try and repair your operating-system installation. Of course, you may have a set of these diskettes provided to you by Microsoft, in which case you
would use the Installation From DOS with NT Diskettes described in the next section. By the way, you can get a list of all the options available from the winnt.exe program by typing winnt /? at the command line.
The next task is to unload the operating-system installation files from the CD-ROM to the hard disk drive. These files are stored on the CD in a compressed format. Therefore, it takes a while for the computer to copy the files to the hard disk drive and
uncompress them. This will make the next step go a little bit faster, but it can still take quite some time, especially for Windows NT 4.0. These files are copied to a temporary holding directory that is usually titled something like $WIN_NT$.~LS.
The third task is to reboot the system under the Windows NT operating system (at least the subset used for Windows NT installation). When you are running NT (either as loaded from diskettes or from your hard disk drive), you will be given a series of
panels that let you select from a number of installation options. Key items in this configuration step include specifying where you wish to locate Windows NT and whether you want to use FAT or NTFS file systems (see the File System section in Chapter 7, "Windows NT Server Overview," for more information). After you have specified all of the options and have loaded, you will be asked to reboot the system for final configuration work.
The final task after this reboot is to perform a few more configuration options. Many of these options are prepared by wizards, which are software routines that examine your system to determine the hardware that is installed or other
configuration options that make sense. This wizard is not as detailed or extensive as the Windows 95 installation script, but it is getting close. After this process is completed, you will be asked to reboot the system once again. After you complete this
reboot, however, you should have a functioning Windows NT operating system ready for all of the postinstallation configuration work such as adding users and loading applications.
A few notes on this process are in order. Loading NT from DOS may seen like a rather primitive process, especially when you can just boot from the diskettes provided by Microsoft (if you have them). It has the advantage of being the start-from-zero
approach—which is needed if you lack boot diskettes or if you need to partition and format your hard disk drives. This is also the process you may want to use if the computer already has Microsoft Windows installed on it.
A final note on this installation process: the software that Microsoft provides with Windows NT enables you to start multiple operating systems on the same computer. This is an interesting "free" piece of software that is most often useful in
the workstation environment. For example, you may want to run either Windows NT or Windows 3.1, depending on the task that you need to perform. Anyway, this software provides you with a menu upon startup that lets you select which operating system (or
version of an operating system when you have multiple versions on the same system) you wish to run. It has a countdown that automatically loads the default operating system you designate after a specified period of time (useful when you are recovering from
power outages or automatically reboot your system). The following is a sample of the boot.ini file (a hidden file located on your C:\ drive in the root directory) that is used to control this process:
[boot loader] timeout=5 default=c:\ [operating systems] C:\="Microsoft Windows 95" multi(0)disk(0)rdisk(1)partition(1)\WINNT="Windows NT Server Version 4.00" multi(0)disk(0)rdisk(1)partition(1)\WINNT="Windows NT Server Version 4.00 [VGA mode]" /basevideo /sos
This process is similar to the one already described, but you use the NT boot diskettes that came with your CD or that you made for a previous installation of NT Server. The key difference here is that if you have the Microsoft distribution diskettes,
you do not have to make boot diskettes. This can save a bit of time as you have to wait while the diskettes are formatted and then loaded to nearly full capacity. The Microsoft-supplied diskettes enable you to load software from the CD to the hard disk,
whereas the diskettes you make from the CD assume that this process has already been completed.
If you have a previous version of Windows NT installed on your computer, you can save some time during the installation process. You can run the winnt32.exe file that is found in the appropriate installation directory on your Windows NT Server CD (for
example, \i386). This executable enables you to make boot diskettes and load software from the CD to the hard disk drive at the same time.
The actual decision about whether to upgrade or to overwrite an existing installation is made after you reboot the computer and choose the directory in which to load NT. Windows NT scans the directory to see if there is an existing Windows NT
installation in that directory. If there is, you will be given the option of upgrading this installation or performing a new installation. I have run into problems when trying to use this upgrade process to revert to a previous version of NT. In these
cases, I have had to wipe out the directory and perform a fresh installation. However, I have had good luck on the upgrade process.
The key difference between upgrades and fresh installations is whether the configuration information stored in the registry and other locations is preserved. This is very important because most applications need their configuration parameters to
function properly. It can take quite a bit of time to reinstall all of your dependent applications. I would always use the upgrade option whenever possible. You may want to try out the upgrade on a development or test server to verify that the benefits
outweigh the costs of the upgrade.
This process is basically similar to the one described in the previous section. Again, the key difference is that you wipe out the registry on a reinstallation and therefore lose all of your application settings. This may be necessary in cases where you
have a really messed-up registry that you cannot seem to reset. It is also necessary to convert a workgroup server to a domain controller. It is a serious operation that can take quite a bit of time, especially if you are running a number of BackOffice and
third-party applications on your server.
What I want to do next is cover a sample installation process for Windows NT Server. I have chosen to cover the installation from DOS without NT diskettes. It is a superset of the steps that you would have to perform on most of the other installation
paths. I have also chosen the NT 4.0 process because that is the current version on the market, even though there are still a lot of 3.51 users. The installation process is fairly similar; if you understand one, you will understand the basics of the other.
Once again, you will find that the screens you receive will be based on the options you select and the configuration of your system. That is one of the great things about the installation-wizard concept. Windows NT senses which equipment you have in your
system and presents you with only the screens that are necessary. It also presents additional screens based on previous options selected rather than displaying them all of the time. The bottom line is that if you get a screen as part of the Windows NT
installation process, you probably need to review what it is telling you (in the case where NT has sensed something and wants a confirmation) or enter some data.
The first phase that you need to go through involves the preparation of your system to load NT. You can think of it as starting in the planning processes described in the last chapter and the beginning of this chapter. It also includes setting up the
hardware if your system has not been completely assembled at the factory. It may also involve partitioning and formatting your hard disk drives if the factory did not take care of this task or you are not fond of the way that the factory has set things up.
In either case, the end result of this phase is that you have a working computer with disk drives that are ready to receive the NT operating-system files.
To prepare your system to loat NT:
The next phase in the installation process is the DOS-based installation process (see Figure 10.3). You now need to insert the Windows NT Server CD into you CD-ROM drive and run the winnt.exe program. The first screen that you get asks you where the
Windows NT distribution files are located. Even though you ran the winnt.exe program from the CD drive, you have to put in its path again (for example, e:\i386 for Intel-based PCs that use E:\ as their CD drive). In this DOS-based portion of installation,
you hit either the F3 function key (if you want to exit the installation procedure) or Enter (to continue to the next screen).
FIGURE 10.3. The DOS-based portion of the installation process.
Next you will see a series of three screens that ask you to insert the three diskettes you will be using to boot your Windows NT system in the next phase. These diskettes need to be completely empty and formatted. You will get an error message if you
try to use diskettes that already have some data on them. These diskettes are to be labeled as follows:
One of the neat things that Microsoft did when designing this portion of the installation process was to write the diskettes in reverse order. This causes you to make the setup boot disk last. Thus, you have the first disk that will be used when you
reboot the system in the drive at the end of the diskette-creation process. I guess the little things impress someone like me.
After you have created that last diskette and press Enter to continue, you are taken to a screen that shows you the progress as you copy files from the CD to your fixed disk drive. This process may seem as if it takes a long time, especially with
version 4 of NT (which has so much more to it). What the system is doing is taking the compressed files from the CD, uncompressing them, and then placing them in a temporary directory on the server's fixed disk. This will make the later installation work
go much more quickly, so try to be patient. Also, I have found that the percent indicator does not move in a linear fashion. It seems to move more quickly at the beginning of the process than at the end. Apparently the files at the end of the distribution
take longer to uncompress than those at the beginning.
At the end of this installation, you get another screen that tells you to ensure that the boot diskette is in drive A:\ and press enter to restart your computer. This is the first reboot that you will go through in your installation. It also marks the
dividing line between the DOS-based portion of installation and that conducted under the Windows NT environment (see Figure 10.4). This is a key point in the installation process. After you leave this point, you will be dependent on NT device drivers and
not those that existed in your previous DOS and/or Windows environment. I have had several cases where I did not have a working driver for NT for a network card or CD-ROM drive and lost functionality for these components after this point.
FIGURE 10.4. The NT-based software installation process.
After you reboot your system, Windows NT will start loading from the setup boot diskette. When the system is done with that diskette, you will be prompted to insert setup disk #2. It will continue to load for a while. During this time, it will load the
Windows NT Executive, the hardware abstraction layer, and some of the device drivers. You are then presented with the blue startup screen you will soon become familiar with. This is a text-based screen that indicates NT is starting up. It will indicate the
version and build of NT being loaded for your reference.
After NT is running (at least in the limited form loaded from diskettes), you are given a welcome screen. Here you can choose between express and custom setup. I typically choose the express setup because it still gives a fair degree of control during
this process and is quicker. Anyway, you can press Enter to continue with express setup or c to perform a custom setup.
The next screen indicates that NT is searching for mass storage devices. This is primarily focused on SCSI controllers/drives and CD-ROM drives. If you have standard IDE fixed disk drives, they will not show up on this list, but you will have access to
them. I guess Microsoft just assumes that you will have IDE fixed disk drives, so they are not treated as anything special. Next you are given a list of these devices for your review, and you can add devices if need be. Otherwise, you can press Enter to
continue on with the process.
The next screen shows you the disk partitions available for Windows NT installation. You can adjust these partitions and select the one that has adequate space. I usually want to put NT on a partition that has at least 200–300M free. It may not
take all of that space now, but operating systems continue to grow in size. Also, some less sophisticated applications install themselves on the same disk drive as the operating system without giving you the option to move them. If you run out of space on
the operating-system drive, you can run into some really nasty problems as the operating system tries to allocate temporary or log files and cannot.
Your next decision is whether to leave the file system intact or convert it to the NTFS file system. FAT was designed in a simpler era. There are certain things, such as having file ownership and access permissions, that can be done only under NTFS.
Also, it would be easy for some nefarious person to boot your computer with a DOS diskette and access any information that you have on FAT drives. Finally, FAT drives are somewhat more sensitive to being corrupted than NTFS drives. I typically use NTFS for
server drives. If you choose to convert the file system to NTFS, it will take a little bit of time for this conversion to take place.
Your next choice is the name of the directory that will contain Windows NT. If you are trying to perform an upgrade, you will want to ensure that you specify the name of the current Windows NT operating-system directory. Otherwise, you can accept the
defaults or pick a name that suits your taste. I typically accept the defaults.
At the next step, NT performs an "exhaustive" check of the drives. It actually does not take that long. After this check is completed, NT will copy the files from the temporary staging directory to the permanent NT operating-system directory.
When this is completed, you are presented with a screen informing you that this portion of the installation process is completed. You will be prompted to reboot your computer.
You are now ready to enter the final installation phase (see Figure 10.5). In this phase, you will set up the detailed hardware- and software-configuration parameters that your operating system will use. This is where you choose which protocols to use,
what your security environment will be, and all of those other choices that you have been waiting to set ever since your wrote up your installation checklist.
FIGURE 10.5. The final NT configuration process.
The first screen that you are presented with is designed to satisfy all of the lawyers in the crowd. Yes, to continue with the installation process, you have to agree to the Microsoft license agreement that says you will not pirate software or use the
operating system in ways that violate this agreement. After you select yes, the operating system copies a few more files to your disk drive.
When this is completed, you start the setup wizard. This is similar to the Windows 95 wizard concept in that the software looks at your configuration and previous responses and presents you with a series of questions (with a fair amount of explanatory
text) that guide you through the installation process. The first step in this process displays the stages of the installation process:
One of the nice features about these wizards is that you have Next and Previous buttons. When you run into a problem, you can usually go back one or more screens and reset some options to try and solve that problem. For example, when you get to the
screen that prompts you for an IP address and you realize that you do not have one, you can go back and de-select TCP/IP networking to avoid getting the IP address screen (which you cannot leave unless you put an address in). You can then get the address
after you complete the installation and add TCP/IP using Control Panel.
The information-gathering section presents screens that ask for the following information:
The next phase determines your Windows NT networking configuration. Although you could have a standalone Windows NT server, I'm not sure why you would do this. The typical server configuration has you connected to a local area network and/or telephone
network. To set up networking, you will go through the following sequence (note that the exact list of panels is strongly dependent on the options that you select in previous panels, but this gives you the general idea):
FIGURE 10.6. TCP/IP Properties page.
This completes the network setup phase. Your final phase is referred to as finishing setup. You have just a few more details to clean up before you can start your new server. The steps in this phase are also determined by the options that you have
previously selected. Typically they include the following:
After this, Windows performs a number of steps including copying more files, updating shortcuts, removing the temporary installation files, and saving its configuration. The configuration save can take quite a bit of time and may appear to be stuck at
certain points. Give it a liberal amount of time before you conclude that there is a problem. It just takes time, and you are at a very sensitive juncture here. You are almost ready to start up, but you need to save all of the settings that you have just
entered.
Finally, you will be prompted to restart your computer. This reboot should take you to a working Windows NT server. There are still a few steps left that will be discussed in the next couple of sections, but your NT installation is complete. Now might
be a good time to consider taking a system backup. You may encounter problems installing other BackOffice components or third-party products. It may be useful to return to this known working configuration if problems get really bad.
There are a number of problems that you can encounter and things that you need to keep in mind while installing Windows NT. I want to mention a few of the ones that come to mind based on my experience:
After you have completed your installation, there are still a number of tasks that need to be accomplished. You have to create user accounts, shared directories, and all of the other things that are needed to make this server useful to your user
community. This can often take a lot of time (especially if there are a lot of user accounts). Therefore, you also need to factor this time into your schedule.
Another demanding task can be installing the other BackOffice components and any third-party applications that will be running on this server. Each of these will have its own installation procedure. They may also require a fair amount of configuration
work. It is usually best to complete the installation of these products before you create user accounts. That way, if they interface with Windows NT's security system, you can provide access to these applications for your users when you create the accounts
as opposed to after the fact.
I would also argue that you should spend a reasonable amount of time verifying that everything is functioning well with your new server. Think of this as taking your server for a test drive before you let your users go at it. I recommend that you test
the functions in detail (for example, try to copy a file located on a remote server rather than just bringing up Network Neighborhood to see if you see any remote computers). This can take some time, but it is easier to make the time now rather than after
you have turned the system over to the users and it is in production.
This chapter has covered the Windows NT Server installation process. It is a rather complicated process that has a number of options to it. The exact set of screens varies with the version that you install and the configuration options that you select. However, the basic process follows one of several defined paths that are based on the current condition of your server and the media you have received. The key to being able to work through this process for your particular situation is a good general knowledge of the operating system and a plan that you work out in advance based on your reading of the installation notes and understanding your user's needs.
