Overall, I thought this exam was perhaps the most straight forward of all the Microsoft exams that I have taken (NT 4 WS, NT 4 SRV, NT 4 ENT). Most all of the questions were related to material that was adequately covered in the student workbook. I only had one question that I would consider to be a truly bad question.
The question posed a scenerio in which a new DNS server was being added to a network that consisted of one other DNS server. The new DNS server needed to provide fault tolerance for the Primary yet could not produce any zone transfers. Which type of server should be installed? Referencing the Student Workbook, page 259, you'll notice that a secondary server is recommended for each zone for the purpose of redundancy, or fault tolerance. However, a secondary server must initiate a zone transfer. Choices for answers were Secondary, Caching, and two other completely invalid choices. I chose Caching because the wording of the question led me to believe that the zone transfer requirement was an absolute, though this was not explicitly stated. I assume I was correct based on my score (I got 100% for all sections except Troubleshooting where I scored 93%).
Another DNS question gave a situation in which a single NT server was being used as both an FTP server and a WWW server. It was desired to connect to the server as either "ftp.acme.com" or "www.acme.com". What type of entry was required in the DNS database to facilitate this? The correct answer was a CNAME record. It seems like most everyone has had a similar question so make sure you are familiar with the types of various DNS records.
There were many questions in which you either directly, or indirectly, needed to understand the difference between WINS and DNS. One WINS specific question gave a situation in which the user could contact an IP but could not contact the NetBios name. Other users were operating fine. The two computers were on different subnets. Basically the computer needed to have WINS configured, but the answer related it to the fact that the user was operating as a "B-node" so make sure you know your node types.
I had quite a few questions on the various utilities. None of the questions required detailed knowledge such as command syntax. Rather, you must simply be aware of what each command does and when to use it. All of these questions were very straight forward. Either you know it or you don't so make sure you know it.
I had about 4 or 5 utiltity questions where Netstat, Performance Monitor, and Network Monitor all appeared as choices. You'll definately want to know what each of these tools does and does not do. A few key points are that Netstat reports statistics from the moment the server was initialized. Performance Monitor allows you to view charts, log data for future analyzis, and export data. Network Monitor allows you to view traffic and analyze packets in real time.
At about question number 51, I was
beginning to think I would make it through the exam with only 1 "scenario"
based question. Then I was hit with a four question scenario for my
final four questions. This scenario included Unix and Windows
workstations on three subnets. An NT machine with three NICs was to
be used as the router between the subnets. There were two requirements
and three optional goals. You were tasked with facilitating centralized
assignment of IP addresses, cross-subnet browsing, Host name resolution,
and enabling the Unix and Windows workstations to use a couple of TCP/IP
applications (ftp and telnet) between each other. Since I had four
questions dealing with the same scenario I made a quick chart like this:
| Result | Question 1 | Question 2 | Question 3 | Question 4 |
| Required 1 | ||||
| Required 2 | ||||
| Optional 1 | ||||
| Optional 2 | ||||
| Optional 3 |
The chart helped me keep track of where my thoughts were going. This chart is exactly like the one I did on the exam. Notice that the proposed solution got slightly more correct with each question. It was definately helpful to be able to go back and forth through the questions. For instance, one of the requirements was that the NT server be able to route IP. The first proposed solution had IP being installed on the server. At first I read this to meet the requirement. The second proposed solution, however, listed both installing IP and enabling IP forwarding. After reading this I realized the the first proposed solution would not meet the routing requirement because it didn't including the IP forwarding.
Speaking of routing, I did get a good question on routing tables.
Your task is to update the routing tables such that packets can be routed from computer A to computer D. Four possible choices are given such as:
1. To Computer B:
Route
Add 167.232.193.0 255.255.252.0 167.232.19.2
To Computer C:
Route
Add 167.120.81.0 255.255.240.0 167.232.19.1
This really wasn't a difficult question. You just need to make sure you understand which interfaces you need to configure for each static router. The example answer given above is correct. Note that the entry on computer B points to the interface on computer C and the entry for computer C points to the inteface on computer B.
There really weren't a lot of questions regarding subnetting, though enough that you should still be familiar with the subnet tables. I did write down the tables beforehad and though I had plenty of time left it was still helpful to have. The calculate, by the way, is not available until you actually start the exam so make sure you either memorize the tables or are comfortable doing the exponential calculations necessary to create the tables.
That about covers most of what I thought were key issues. Study smart, practice some sample questions, and take about 15 minutes before starting your exam to relax and do a few breathing exercises to calm your nerves. Before you know it you'll be a MCSE.
Good luck!