Confusion regarding zero subnet on exam

/usr

On page 361 of Odom's ICND1 book, he mentions that the default for the exam is to allow the zero subnet and broadcast as usable IP addresses unless the question explicitly states otherwise.

However, I am a bit confused. He says to use the 2s formula to calculate the number of possible subnets if the question allows the zero subnet (default).

However, in the practice questions, as well as on subnettingquestions.com, you use the 2s formula when it asks you to calculate the number of usable subnets and the answers are always right.

But they are still clearly defined as being the broadcast and zero subnet when you have to calculate the first and last usable IP addresses, so technically aren't they still reserved? What's the difference between this and the 2s - 2 formula when the addresses are explicitly reserved?

Maybe I've been staring at this stuff too long or I'm reading into it too much (again).

Could anyone explain this better to me? What am I missing? I know I'm overlooking something simple.

dynamik

Here's a quick link: http://tinyurl.com/3m7o94

And here's a more detailed one from Cisco: https://www.cisco.com/en/US/tech/tk648/tk361/technologies_tech_note09186a0080093f18.shtml

wbosher

From what I understand the subnet zero applies to the whole zero and broadcast subnet range, not just those two subnets. For example:

An network address of 192.168.1.0 broken into 8 subnets. Mask will be 255.255.255.224 or /27.

If ip subnet zero is enabled, the only addresses you can't use are the first and last addresses of each range, if ip subnet zero is not enabled you can't use the entire first and last range.

Hope this makes sense.

/usr

For example, this question.

"What is the last valid host on the subnetwork 172.25.16.0/20"

The answer is 172.25.31.254

From my understanding, this means the zero subnet is not usable since 172.25.31.255 is not a usable IP address and the correct answer is 172.25.31.254.

If this question were asked as is on the exam, would you say the answer is 172.25.31.255, since if nothing is mentioned the default is that 172.25.31.0 and 172.25.31.255 are usable IP addresses?

Am I understanding this correctly?

Nevermind, I'm a fool!

I didn't think the subnet number and broadcast address were ever allowed, that's why I was getting confused.

So if the zero subnet is not allowed, then no address from either entire subnet (all 1's and all 0's) is allowed, correct?

It's referring to the subnet as a whole...not to the first and last IP on a given subnet.

NOW am I understanding it correctly?

wbosher

That's correct. Using 172.25.16.0/20 the first range is 172.25.16.0 - 172.25.31.255, the last range is 172.25.240.0 - 172.25.255.255. If subnet zero is allowed, you still can't use the first and last address in each range, but the addresses in between are ok. If subnet zero is not allowed, you can't use the entire first and last range. Hope this makes sense, I'm starting to confuse myself.

Diminutive

ip subnet zero is a red herring, ignore it. No ip subnet zero OTOH. avoid the first possible subnet and the last

dynamik

/usr wrote:

Nevermind, I'm a fool!

I didn't think the subnet number and broadcast address were ever allowed, that's why I was getting confused.

So if the zero subnet is not allowed, then no address from either entire subnet (all 1's and all 0's) is allowed, correct?

It's referring to the subnet as a whole...not to the first and last IP on a given subnet.

NOW am I understanding it correctly?

Yes, the issue is whether you can use the all-zero or all-one subnets. The lowest address is always the network address and the highest is always the broadcast (are therefore are not assignable to hosts), regardless of whether or not you can use those two subnets. Sorry I didn't give you a better explanation right off the bat; I didn't completely understand what you were asking.

/usr

I appreciate the link anyway, it's what helped me realize the difference.

Thanks!

AlexMR

Op,

Having taken both exams (ICNDs) recently and passing them and having them fresh in the memory (last week), I have to strongly suggest you that if you are doing a SIM, you should check the running-config to see if the ip subnet zero command has been typed and proceed accordingly.

If it's a question and it's not specified, then you have to assume the subnet zero is usable, as this is the standard since...well, it is the default in cisco routers for the whole time Ive been studying .