Subject: 70-059

I just took TCP/IP and found it much easier than it's hyped up to be.

I had virually no questions on switches regarding nbtstat, netstat,

etc; others who had taken this exam told me there were lots of these.

My score was 862. I used Transcender and Exam Cram. Both are excellent

resources. The Transcender exams are almost identical to the real

test. And many of the Transcender questions are on the real test, word

for word, including people's names.

Tips:

You don't necessarily have to memorize the subnetting chart if you know

the formulas:

2n-2 = number of subnets for a mask

2z-2 = number of hosts per subnet.

example:

If your mask is 255, then that is 11111111 in binary (n=8 - count the

1's). Use the formula 2n-2 and you get your number of subnets, which is

254 (2x2x2x2x2x2x2x2 (256) - 2).

If your mask is 254, then that is 11111110 in binary (n=7 - count the

1's). Use the formula 2n-2 and you get your number of subnets, which is

126.

If your mask is 252, then that is 11111100 in binary (n=6 - count the

1's). Use the formula 2n-2 and you get your number of subnets, which is

62.

etc...

If your mask is 255, then that is 11111111 in binary. Use the formula

2z-2, where z represents the 0's left after you've borrowed the 1's from

your hosts for the mask, to get your number of hosts per subnet. For a

Class A address, you have 24 bits to alter. if your second octet is

255, then you have altered 8 of the 24 bits (by borrowing 8 from the

host id), which looks like this: 11111111.11111111.00000000.00000000.

Therefore, z=16. Use 2z-2 (2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2 (65,536) -

2), which is 65,534.

A Class B address would be 254 (2x2x2x2x2x2x2x2 (256) - 2).

You will be allowed to use the computer's calculator (scientific). If

you type in the decimal value of the subnet mask and click on Bin, it

will give you the binary equvalent. If you know the formulas above, you

can quickly figure out the number of subnets/hosts per subnet. There

will be approximately 6 of these on the test.

Also, you can't figure the number of hosts per subnet, if they don't

give you the Address Class. Even if they don't tell you which class it

is, if they give you the network IP address, you should be able to

figure out which Address Class it is.

Remember that a p-node makes a request directly to the WINS server, a

b-node broadcasts to the local network, an h-node directs p-node

request, then b-node (WINS server then broadcast), and an m-node

performs the opposite of h-node.

Know when to use WINS vs. DNS like the back of your hand. Know how to

troubleshoot HOST files, LMHOST files (wrong syntax issues in both) and

why a person can ping an IP address but not the host id (or the other

way around) when the IP address is correct and the host ID is in the

person's HOST file correctly.

Know when an error can be associated with wrong gateway vs. subnet mask

vs. Host IP address.

I suggest using at least one of the above resources.

Good Luck.