This is Part 2 of the final blog post in this series. In Part 1 , we discussed a strategy for the exam-taking process. Now we’ll talk about a few snags that you might have to overcome during the exam. Each of the remaining sections presents a specific type of problem and provides recommendations on what you should do if you encounter it.
Although most Microsoft exam scenarios appear to be detail-oriented, they are usually not precise; that is, they often don’t explicitly state many relevant considerations. I don’t know for sure whether this is done on purpose or happens naturally. The important thing, however, is what you should do about it. When some minor but necessary details appear to be missing from a scenario, I suggest that you make assumptions based on common sense and your practical experience.
Here is an example. Active Directory Domain Services (AD DS) relies on DNS. The recommended and most practical way of using these two technologies is to configure all or most domain controllers as DNS servers hosting AD DS–integrated DNS zones. Generally, you can use standard zones, including those hosted on DNS servers running third-party operating systems (OSs). But if you have ever tried to install the first domain controller in a new forest while the corresponding DNS zone is being hosted on a stand-alone server, you should know how inconvenient, impractical, and unnatural that setup is. Suppose you then want to migrate the standard zone to the domain controller and make the zone AD DS–integrated. To be able to migrate the zone to the domain controller, you need to install the DNS Server role on the domain controller. But the moment you do that, an AD DS–integrated zone for the corresponding domain is automatically created and the domain controller is configured to use it, all without your consent or even a hint of a warning. Thus you’ve ended up with two independent zones for the same domain. To proceed with your original plan, you will have to delete the AD DS–integrated zone only to re-create it properly mere minutes later. In real life, you definitely do not want to be caught up in this hassle.
The most important implication from this example is that domain controllers are usually also DNS servers and the zones that they host are usually AD DS–integrated. If you encounter a scenario in which a DNS server is named DC1, DC2, or something similar, you might not be able to properly accomplish the task in the scenario unless you assume that the DNS server is also a domain controller. Or if a scenario says that a domain controller is also a DNS server but does not say anything about its configuration, chances are it hosts an AD DS–integrated zone for the local domain.
It is often useful to remember the default settings for various features, installations, or configurations. In the Microsoft world, the defaults are usually chosen intelligently so that an application or feature will work just fine with minimal manual configuration. If an important piece of information is not explicitly provided in a scenario, it is usually reasonable to assume that the default settings are in effect. For example, by default, all Windows Server 2012 domain controllers are configured as DNS servers and global catalog servers. Unless a scenario clearly indicates differently, it is reasonable to assume that all domain controllers in the scenario are indeed DNS servers and global catalog servers.
“Everything in moderation,” as the saying goes; so be careful not to overuse any given technique. If something has worked fine in one item, it does not necessarily mean that the same approach will work in another item. For example, the assumption that the default configuration is in use might work for you in an item where the answer is to provide a Windows Server 2012 DVD in order to switch a Server Core installation into full-GUI mode (because an original Server Core installation does not include the binaries for full-GUI mode by default), but can work against you in another item if you assume that a Windows Server 2012 computer runs Server Core just because this is the default mode. Using a GUI is so much more convenient than typing long, convoluted, and cryptic command lines; consequently, an administrator is likely to install a full-GUI server even if he or she is planning to manage it mostly by using PowerShell (PS). So, if a scenario does not say that a particular server is running in one mode or another, it is not sufficiently safe for you to assume that the default mode is in effect.
Here is another example where the default configuration does not necessarily work. Suppose a scenario states that a computer runs the original version of Windows Server 2008. PS is not installed in this OS by default. However, you should not reject a choice just because it suggests that the correct action be performed by using PS. Unless another choice suggests that the same correct action be performed by using a different method, you should assume that PS must have been installed, even though not by default.
Microsoft has rather strict guidelines on the use of technical terms. They periodically publish a book named Microsoft Manual of Style, in which they explain how technical writers should describe various features and objects related to Microsoft software products. However, even Microsoft exam writers make mistakes and use incorrect terminology sometimes. Here are a few examples that will help you know what to look out for.
The first example has to do with hard disks. Windows Server 2012 introduces a new model for arranging hard disks. The disks that meet certain requirements can be arranged into storage pools. First, those disks automatically appear in a special storage pool named Primordial, and then we can assign them to custom storage pools. In those storage pools, we can create special virtual hard disks (VHDs) that are known as storage spaces. A storage space appears in the Disk Management console as a physical disk. To be able to store data on storage spaces, we need to create volumes similarly to how we create volumes on regular hard disks. In some Microsoft exams, you might encounter items where storage pools are explicitly or implicitly referred to as storage spaces or where VHDs or other disks are treated as if they were volumes or vice versa.
The second example has to do with the graphical tool that presents files, folders, printers, and other objects. In the very first versions of Windows, the tool was known as File Manager. Starting with Windows 95, this tool was renamed Windows Explorer; this is the name all IT professionals and most users working with Windows have been using for the last couple of decades. In Windows Server 2012 and Windows 8, it has been renamed File Explorer. All the versions of this tool have been created by Microsoft; so it’s up to Microsoft software architects how they want to name their tool. But Microsoft exam writers, on occasion, seem to be a bit rusty; they often continue to use the name Windows Explorer without regard to the versions of the OSs running on the computers in question. Usually, it doesn’t matter, because we “know” (sort of) what they mean when they interchange the terms. But sometimes it might be important. If we are expected to perform some action on a computer by using Windows Explorer, are we supposed to conclude that the computer runs Windows Server 2008 or Windows 7? Or if the tool is referred to as File Explorer, does it always mean that the computer runs Windows Server 2012 or Windows 8? Different OSs support different features, and tasks in different OSs are performed by using different tools and different procedures.
The final example has to do with folders. Files can be arranged in containers known as folders. The folders can be shared, and the shares can be published in AD DS. Folders, shared folders (or simply shares), and shared folder AD DS objects are three distinct types of objects. It is not uncommon for exam items to routinely use folders or shares to refer to all three types of objects. Usually, it’s not too hard for us to figure out what exactly the exam writers meant to say, but that is not always the case.
What should you do if—perhaps I should say when—you come across a terminological discrepancy in an exam item? Here is my suggested rule of thumb for this issue: Cut some slack to Microsoft exam writers. They are humans, just like us, not divine creatures; they are neither omniscient nor infallible. When you are looking at an item that contains what appears to be a terminological error, I suggest that you imagine the exam writer as a buddy of yours who has a habit of interjecting “You know what I mean?” every couple of sentences. Most of the time, this paradigm will help you. However, always keep your mind open to other possibilities as well. Use every bit of information available in the item and all the logic you can muster to figure out what exactly the writer meant to say.
Many exam item scenarios present various technical problems that you are required to resolve. For you to be able to correctly identify a problem, it should be correctly described. However, some scenarios incorrectly describe the effects caused by the underlying problems.
Suppose a scenario states that you implement a certain configuration on a domain controller and appears to imply that AD DS replication starts failing as a result of your action. However, in reality, the action mentioned in the scenario cannot cause replication to fail. Rather, the configuration that you purportedly implemented simply cannot be implemented; if you attempted it, it would result in an error or would be ignored and nothing would actually change. Such an item cannot be resolved in a technically correct manner. Then what should you do? Thoroughly analyze all the available information, both explicitly provided in the scenario and implied by the question and choices, and try to figure out the exam writer’s intent. It could be that you are expected to simply force AD DS replication, and all you need to do is choose a suitable tool. Or you might be expected to choose a tool for reversing the changes that you have purportedly made. Yet another possibility might be that you are supposed to select the type of backup that you should restore in order to reverse your inappropriate action and a tool or method for doing so.
There can be no rule of thumb for such situations, because each technical problem can be misrepresented differently. Just stay calm, and don’t immediately jump to what might appear as an obvious conclusion at the moment. Take a deep breath, and do a proper analysis. And always remember: there is a live person, perhaps just like you, behind each item. Imagine that it was indeed you. What would your intent be if you wrote that item?
Sometimes you may come across technically flawed exam items; that is, items that just don’t make technical sense because the exam writer appears to have made a mistake. What you should do depends on the exact nature of the problem. Some of these issues are quite easy to address, but some might present a real challenge.
One common type of error happens when the exams initially designed for the original version of Windows Server 2012 (or 2008, for that matter) are updated to include material for the R2 version. Microsoft does not use the phrase original version to refer to the original version of Windows Server 2012 (or 2008). Instead, they say simply “Windows Server 2012” (or 2008) without any modifiers. When they update their exams for the R2 version, they add new items to cover new features and functionalities and, in the existing items, they often simply add the R2 version designator to the name of the OS and do not perform a proper technical review. As a result, some items become invalid; something that worked fine for the original version no longer works for R2. For example, the version of Active Directory Federation Services (AD FS) in the original version of Windows Server 2012 is v2.1, whereas Windows Server 2012 R2 comes with AD FS 3.0. There are major architectural differences between the two versions of AD FS: v2.1 is built on top of Internet Information Services (IIS), whereas v3.0 is built directly on top of HTTP.SYS. Consequently, items originally written for AD FS 2.1 and based on the involvement of IIS are invalidated when R2 is added to the OS’s name. This issue is not limited to AD FS; it can be about any feature that functions differently or is named differently in the original version of the OS than in the R2 version. This type of problem is usually relatively easy to resolve: just disregard the R2 version designator and pretend that the scenario is about a computer running the original version of the OS.
Another type of error could be the result of using inappropriate terminology; some “intuitively apparent” phrases can be deceptive and, if used in an exam item, might lead you in a wrong direction. Try to get into the exam writer’s shoes: try to figure out the writer’s personality, and based on that, try to second-guess the writer’s intent.
For example, if a scenario says that you need to assign a drive letter to a VHD, then it’s a clear sign that the writer was simply a bit inaccurate and meant to say that you needed to assign a letter to a volume residing on the VHD.
However, consider another example, in which the writer might be smart and might also be trying to trick you. Suppose a scenario does not explicitly state the exact version of the OS running on a server but instead mentions a feature that is available in only the R2 version. There might be one choice clearly intended for the R2 version of the OS and another choice that is typical for the original version but that can work in the R2 version as well. Take a good look at those choices. If both are valid, then the writer is just smart and is not trying to trick you; select the choice intended for the R2 version. If one of those choices is a bit inaccurate, somewhat sloppy, uses technical jargon instead of the proper terminology, or lacks a mandatory parameter in an otherwise feasible command, then the writer is not only smart but is also trying to trick you. In that case, select the choice that is 100 percent valid, even if it is not the one specifically intended for the R2 version.
Some wrong exam items appear to be the result of mistakes in technical documentation. Maybe the writer didn’t bother to personally test a scenario and just wrote it along the lines presented in an article from TechNet Library. The way to resolve such items is, again, to second-guess the exam writer. It can be quite easy or extremely difficult. If you are familiar with the TechNet article whose mistake the exam writer appears to have replicated, if you know about the mistake in that article, and if there are no choices with technically correct answers, then it is easy; all you need to do is pretend you never noticed the mistake in the documentation. But if one choice is based on the mistake in the documentation and another choice provides a technically correct solution, you can never know whether the exam writer failed to verify the scenario or, on the contrary, the writer was smart enough to catch the mistake and decided to have fun at your expense. My advice for such situations is this: Always choose the technically correct solution if there is one, and resort to second-guessing the writer’s intent only as a last resort.
Don’t get too worked up about the issues covered in this blog post. Usually, most of the items in an exam are sufficiently technically sound and only a small percentage of the items have the potential to cause trouble. Prepare for the exam properly: choose the right tools and materials, study, and play with the software in a lab until you can clearly envision all the dialog boxes and can type all the right commands without consulting a reference. Schedule the exam only a day or two in advance, when you feel you are in the best shape for it as you will ever be.