RE: 70-293 MS training book has serious Errors! Please comment

Can't edit my original post.
I meant the book says Method 1 and not Method 2 (typo error) when I am
talking about first and last subnets being invalid.

"Graham Allen" wrote:

I feel I have to post this because of how bad the official MS 70-293 training
book is.
This is the book in question
I know MS don't directly write this book but it is licensed under MS press
and they should vet it at least!

Firstly let me say that I am already pretty knowledgeable on the 2003 OS. I
am going through this book to brush up on my skills in prep for the exam.
In particular these errors are relating to subnetting. I have only got into
the early part of the book and already I have found loads of mistakes.
This post may be a bit difficult to read but I have done my best to clear
things up where I can so my apologies if you find it hard to follow.

Let me explain the problem. On page 2-32 you have some practice questions.

The first one gives you a network subnet of with several
questions. The ones that are important (and answers wrong for that matter)

The First subnet?
The last Subnet?”

Now, I have always been frustrated with the vagueness of MS Exam questions
because usually you need a bit more information as you can answer the
questions with different answers and all will be technically correct. MS
however wants you to answer them a specific way and usually expect you to
make assumptions.

Let’s look at the first question:
Ok we work out that the subnet mask is This means that each
subnet increments by 32 (256-224 subnet mask).
Depending on your experience/training the first subnet can be:
Network ID:
Broadcast ID:
IP Range: –
Note: This is the answer I came up with!

Or if you count the first and last subnets as invalid ranges (which we will
come onto later on) then your first subnet is as follows:
Network ID:
Broadcast ID:
IP Range: –
Note: This is the “correct” answer in the MS training material.

So straight away you have 2 different answers! MS make no explanation in the
chapter about which method should be used. The second method conforms to RFC
specification but MS version of TCP/IP allows you to start the first subnet
on Oh well...Let’s just hope we get it right in a lucky guess!
Anyway we have the training material right?? So we can see which is the
correct method at least for the MS exam?? Wrong! Based on the answer in the
book this means we are using the second method which states that the first
and last subnet ranges are invalid. In the second question which asks what
the last subnet is, again you have 2 options again:

Method 1 (where you don’t count the first and last subnets as being invalid)
Network ID:
Broadcast ID:
IP Range: –

Method 2 (where you count the first and last range is invalid)
Network ID:
Broadcast ID:
IP Range: –
Now according to the book then the last subnet should be what is in method 2
but it isn’t. The answer says the results of method 2!! So not only is the
question vague in the first place but it now is applying one rule on question
and a completely different rule on the second question!!! So what is the
correct one? How am I meant to have any faith in the MS exam answers based on
Now if you thought that was bad then see the following forum post. The
person doing the explaining is basically saying the same as me but at the
bottom he has provided a link that takes you to an MS support page stating
the corrections and amendments for this book. This problem isn’t listed! I am
paying good money for this training material which is riddled with errors and
the best MS can do is update them on a site and not even reference this in
the book!! It was by chance that I found this and investigated further!
Here is the link to the excellent post I found

Don’t worry about the link, it is cached copy from the Google servers.
Mysteriously the original post has been deleted..hmm. I think MS may have
taken it down. If you are scared of the URL then go to Google and type this
in “Subnetting in 293 in MCSE Exam”. The first post that comes up click it
and it will say the post doesn’t exist. Go back then click the “cached” link
by the green URL. This will take you to the URL above and a cached copy of
the original post a Google server.

Let’s do some more analysis on this...I found a total of 18 questions
relating to subnetting in this section of the book 2-32. 6 of them are wrong,
that is a third. That means that if I took the exam with just these questions
and answered everyone correctly a third of them would be wrong which would be
an instant fail!! This is unacceptable and I think this is something that
needs fixing!
I will list all the questions now and why they are wrong.

3 questions on 2-32 in the same format as the examples I gave above just
using different IP subnets. Of course all of these are wrong because one rule
is applied to the first subnet and a different one to the last subnet.

The next set of questions are based on a case scenario. It is a massive
load of text but I will just mention the facts that are relevant to the
incorrect answers:
• You have been assigned a private IP subnet of
• You have 4 LAN’s and this subnet must be further subnetted so that each
LAN has 1 (and only 1) subnet.
• The maximum amount of host’s on a LAN is 100.
Question 5 Page 2-52
Which IP range classes can you not use when selecting a network address?
(choose all that apply)
Class A
Class B
Class C
Class D

I chose class D only but the book says Class C and D.
Firstly we know a class D is a multicast so it can’t be used. Now let’s look
at why I see Class C as being valid.
Again we have to make assumptions...sigh! Based on the fact that the book
says C and D are not valid but A and B are I presume they are referring to
the network address range they gave you earlier and the question
implies you can choose an IP range of your choice. The reason I say this is
because if we use then this is a class B network therefore all
answers would not be valid except the Class B answer. But because they say
class A is valid then they can’t be saying you must stick with the address address they brought up earlier. So here comes the next
problem...if you are allowed to use any IP range then all of them would be
valid except for class D which is the answer I gave?!?!
I thought about this for a while and tried to come up with a solution that
would fit with the MS answers and here is what I got.
The only reason I can see them saying a class C is invalid is if you base it
on the fact that each subnet must be able to accommodate 100 PC’s (as stated
in case scenario). Again we have to make another assumption though for this
to work. This assumes that the question means you must subnet the network
however. A class C network allows for 254 hosts therefore this would be valid
so I assume you must subnet it. If you subnet it by just 1 bit you narrow it
down to 128 hosts. Now this is still valid but using 1 bit only allows us to
have 2 subnets of the class C network (as I said we must assume that it wants
us to subnet because a normal class C network would valid). Because we only
have 2 subnets (and now a third assumption) these 2 subnets are invalid
because you can’t use the first and last subnets of a network (using method 2
in above examples). Still with me?? Anyway as you can see there is so much
ambiguity and assumptions you must make to get this answer correct!!

Question 6 Page 2-52
For each answer you selected (in above question) explain why?
Now because I only selected Class D I would get this wrong because it wants
me to explain why a class C is wrong.
Note: The answer the book gives for the Class C not being valid states it
can’t support 100 PC’s. This further supports my idea that the question
(previously) wants you to subnet the network because a normal class C network
not subnetted would suffice.

Question 7 Page 2-52
Assuming you will use a network in the class B private address range what is
the max number of subnet ID bits you can use and still have a sufficient
number of host ID bits to support the comps on each of your networks?
(This is talking about the network address) My answer was 9
bits as 7 bits are left for the host ID leaving you with a max 126 hosts per
subnet. My answer was the same as the book. The answer is correct but this
affects the next set of questions which is why I have listed it here (bear
with me!).

Question 8 Page 2-52
Using the network specified earlier (What network previously?? The in the case scenario or the one I have just subnetted in the
previous question) how many subnet ID bits are you using for your network
See what I mean about being vague?? I thought this question was talking
about my network which I have just subnetted in the previous question. It
turns out it wasn’t.
This reference was to the that is listed in the case scenario.
It could mean one or the other. I mean let’s think about normal conversation
flow. If I am talking to you about a object/person/car (for example John)
then I start talking about Frank then I say “he said he went home” who do you
think I’m talking about?? I know I would think that Frank went home but
applying this analogy to the question in the book it would mean I was talking
about John again now without first telling you I was. It doesn’t matter
really what you think, my point is that it is too vague a question to say
“the previous network specified earlier” when you now have 2 different

Question 9 Page 2-52
What subnet mask must you use for the unregistered LAN’s on your network?
Answer (in book):
Now what network is this referring to?? The in the case
scenario (question 8) or the generic class B network it asked you to subnet
in question 7?in each answer you gave for the two separate questions above
result in different subnet masks.
I thought it was talking about question 7 so with this it would mean we have
16 for the class B network and borrow 9 bits to subnet it (as asked in
question 7). This leaves us with 7 bits for the host thus your subnet mask
would be which according to the book is wrong.
But it was talking about which would be correct to get the subnet mask.

Question 10 Page 2-52
List the IP address ranges for the first 4 subnets created from your
unregistered network address?
I got this wrong of course because my subnet mask was which
meant I calculated the subnets wrong.

As you can imagine it took me a while to write all of this and since then I
have moved further into the book and found more errors which again have gone
From what I understand the whole networking part of the book was originally
in another book called “Network Essentials” about 10 years ago and this was a
separate exam by MS. They decided to phase out this exam and integrate the
knowledge and exam questions into this exam. It looks like it hasn’t been
updated in 10 years either though!! Here is an example which is clearly wrong
(but may have been right 10 years ago):
You have a IP address, what could be the cause of this?
It turns out the answer was a network cable was unplugged.
This is no longer correct...especially in XP and 2003 (which is what the
exam is about!!). Now it says the media is disconnected and your IP shows as About 10 years ago though NT maybe Win98 or even Win2k did show this
(but I can’t remember to be exact).

This is a joke and the training material needs a serious overhaul. The test
questions you get in these books and other books come directly from MS
because as I have seen them similar in the exams. No exam questions should be
vague. If there are clearly different ways to answer a question then MS
should give more details, at least hinting how they want it answered. You
could quite easily fail because of the lack of details given in the questions.
I am sure I am not the first to see this and I would like to hear other
people’s opinions on this.

Thanks reading.


Relevant Pages

  • 70-293 MS training book has serious Errors! Please comment
    ... The first one gives you a network subnet of with several ... I have always been frustrated with the vagueness of MS Exam questions ... Ok we work out that the subnet mask is ...
  • Re: ICS questions and confusion
    ... >>>It doesn't HAVE to be on a different subnet, ... but that requires that the ICS host become a bridge. ... >> ICS is a software based NAT router, and routers work best when the ... >> network already had the required address" is confusing. ...
  • Re: How to set my MAC address
    ... because the VLAN isolation fails between the wifi side and the wired ... ports that I started experimenting with the 2nd subnet feature. ... there is no option to isolate the wireless from the VLAN. ... give out the password of my own wifi network, ...
  • Re: CIDR Explanation
    ... slash notation that's becoming more common. ... CIDR or Classless Internet Domain Routing is VERY similar to subnetting and is ... This two part address is built of a network address and a host address. ... TCP/IP Classes allow for a 'default' subnet mask to identify the network from ...
  • Re: Make a DHCP server using Fedora - Help
    ... The dhcpd configuration is highly network-dependent. ... What is your network topology? ... please write a subnet declaration ... Do you want dynamic updating of your local dns server? ...