Class A/B subnetting question

Bearded Steve

I'm pretty OK with subnetting, but there's a part in Todd Lammles book regarding Subnetting of Class a/b networks that i don't fully understand. I seem to be missing something.

class B

as long as all the subnet bits on the third are
not all off, then subnet 0 in the fourth octet is valid. Also, as long as all the subnet bits in
the third octet are not all on, 192 is valid in the fourth octet as a subnet.

I don't get that, surely when you borrow bits from both the third and fourth octet, the subnet bits in the third are always on, i.e 255. So subnet zero is always valid. Same goes for subnet 192, which, according to the above statement, isn't valid if you have 255 in the third, which is AWAYS, so it's never valid.

According to the statement above. For 192 in the 4th to be valid, you would need a network mask with something other than 255 in the 3rd octet. Something like this? I've never seen anything like it

255.255.0.192

Same goes for class A

255.255.255.192

Valid subnets? Okay, now we need to add subnet numbers from the second, third, and
fourth octets. In the second and third, they can range from 0 to 255, as long as all subnet
bits in the second, third, and fourth octets are not all on or off at the same time. For the
fourth octet, it will be 256 – 192 = 64. And 0 will be valid as long as at least one other subnet
bit is turned on in the second or third octet. Also, 192 will be valid as long as all the bits
in the second and third octets are not turned on.

So, he says that subnet 0 is valid as long as at least one subnet bit in the second or third octet is on. I've never seen a class A mask where this isn't the case, there's always at least one bit turned on in the second or third, so subnet 0 is always valid in a class A sunbet mask is it not?

And the same again for subnet 192, he says that as long as the subnet bits in the second or third octet aren't all on, then 192 is valid. But when you borrow from the 4th octet, it only stands to reason that all the bits in the second and third are on, or else you wouldn't need to borrow from the 4th octet would you?

Please somebody expalin this to me.

EdTheLad

Steve, i understand subnetting but i have no clue what Todd is talking about in the quotes above.Probably you are using an old edition with a crap explanation.I used 4th edition and i found the subnetting very straight forward.

Bearded Steve

That is the 4th edition and is pretty much word for word.

It's confusing isn't it. To condense all of that, can anybody expalin to me when the zero subnet and the last network value (128,192,224 etc) are valid subnet values.

EdTheLad

Steve,your questions are very obscure, i think your reading to much into nothing.Why dont you read up on subnet-zero,there is a already a recent post explaining it.The zero subnet and the last network value (128,192,224 etc) are always valid,its your choice to use or not.

Example: If you have a class B network address 151.98.0.0/16
Imagine you want to create subnets with 60 users i.e. a block of 64
you will use a mask of 255.255.255.192
valid subnets 0,64,128,192
i.e. 151.98.0.0,151.98.0.64,151.98.0.128,151.98.0.128,151.98.0.192

Now probably what Todd is trying to explain is this.
For your original network 151.98.0.0/16, your subnet zero has an identical
address 151.98.0.0,the only difference is the mask.
For your original network 151.98.0.0/16 the broadcast address to
address all devices 151.98.255.255, on your last subnet the broadcast address is 151.98.255.255, the same how to differentiate?
both subnet and broadcast are the same as the major network which can casue you problems.
Anyway read about subnet-zero.

Bearded Steve

Steve,your questions are very obscure, i think your reading to much into nothing.

Yeah i think you're right. I guess my main problem was not understanding why the zero and last subnet were not commonly used. Your explanation has helped.

thanks