Subnetting question.

[kw7]

I'm studying CCNA now and I come across a question I could not figure out....
The question is from the Sybex book by Todd Lammie. In which page 155.....
For a class A address with subnet mask 255.255.240.0, it said that the first subnet is 10.1.0.0....
I just wonder why shouldn't it be 10.0.16.0????
BTW, could anyone tell me how could I find any info the the 'Errata' on this book? I tried sybex's website, but no clue...
Thanks for the help.

 

[spidey07]

sounds like a mistake to me

 

[kw7]

My mistrake or the book's mistake?

 

[ScottMac]

OK, I'm gonna try this without confusing myself.

10.0.0.0 with a 255.255.240.0 mask.

20 bits of subnet mask, 12 bits of host address. If you go by "the rules," not convention, then the first subnet would be 10.0.0.0 (all zeros in the host portion,subnet 0000). The first available host address would be 10.0.0.1, and the broadcast would be 10.0.15.255 (all ones in the host portion).

The next subnet would be 10.0.16.0 (all zeros in the host adress, subnet 0001), the first host address of the second subnet would be 10.0.16.01, and the broadcast would be 10.0.31.255.

The third subnet would be 10.0.32.0 (all zeros in the host portion,subnet 0010), the first host address would be 10.0.32.1, the broadcast would be 10.0.47.255 (all ones).

The fourth would be 10.0.48.0, 10.1.48.1, 10.0.63.255 (subnet 0011)

Fifth would be 10.0.64.0, 10.0.64.1, 10.0.79.255 (subnet 0100)

and so on.

Since there are four bits left in the last subnet octet, the subnet will always increment by 16. If there were two bits, it would increment by four...etc.

The addresses by convention may not be acceptable. By convention, all zeros are the "net address" and all ones in the host portion are the segment broadcast address. The convention favors the old UNIX System V convention.....BSD used to use all zeros as a broadcast. So, though they are legit addresses "by the rules," conventional practice tosses out all zeros and all ones as host addresses.

If you wnated to ues the 10.0.0.1/20 in a Cisco router, you also have to enable "Use Zero Subnets"

Gawd I hope I got that right......it's been a rough day. Now I have to go sharpen my pencil again......

FWIW

Scott

 

[kw7]

Hey thanks for all the helps.

 

[spidey07]

pat yourself on the back scottmac, you are SO ON!

This starts the whole "can I use the all ones subnet or the all zeros subnet argument".

Safe answer = no, don't do it
Real world answer = yes

cheers!

 

[ScottMac]

I was sweating it there for a while...

-Scott

 

[nweaver]

Congrats Scottmac, That was pretty good. An easy way to figure subnet ranges is to make a step value. 2^x=step, where x is the number of subnet bits in the shared octet. You ignore any 255, but a 255.255.224.0, you could figure a step value of 2^3=8, so you would go 10.0.0.0 to 10.255.7.255, next subnet is 10.255.8.0 to 10.0.15.255

easier like this
10.0.0.0
add 255 for full octet of subnet and 8 for shared octet
10.255.8.0 add step of 8
10.255.16.0 +8
and then you can go subtract one from the next network address to get the previous Broadcast
10.0.0.0-10.255.7.255
10.255.8.0-10.255.15.255

Anyway, this is a quick tip for figuring valid ranges and Broadcast Addresses.

BTW, we should start a perm. Cisco tread for CCNA's in training... I'm in the same boat

BTW, 11:40 PM, usually go to bed at 9:30, so be gentle

 

[nweaver]

Er, Ignore that step value formula, correct one is 256-Subnet Mask =step Value

256-254=2

10.0.0.0
10.2.0.0
10.4.0.0
10.8.0.0
etc