# Subnetting Question ?

Apr 7th, 2008

Hi,

You have a class B ip address with a 12 bit subnet. How many subnets and hosts are available ?

a.-4096 subnets;16 hosts

b.-16 subnets;4096 hosts

c.-4094 subnets;14 hosts

d.-14 subnets;4094 hosts

I have selected option 'd' but the right answer selected is opction 'c',

any explanation is welcome

Overall Rating: 0 (0 ratings)

## Replies

scottmac Mon, 04/07/2008 - 20:03
• Green, 3000 points or more

Since they have you starting with a classfull address in the Class B range, that implies a natural mask of 16 bits

11111111.11111111.00000000.00000000

(aka 255.255.0.0 or /16)

When you add 12 subnet bits (additional network bits beyond the natural mask, that's 2^12 - 2 (network and broadcast) 2^12 power is 4096, subtract 2 (network and broadcast) that's 4094 SUBNETS (added to the 64K NETWORKS defined by the natural mask)

So, taking the class b network bits (16 of 'em), adding 12 more Subnet bits give us 28 bits:

11111111.11111111.11111111.1111000

That means there's 4 bits left for host addresses. 2^4 - 2 is 14 (2 to the fourth power is 16, subtract two - network and broadcast, for 14 hosts addresses available for each subnet.

Hope this helps

Good Luck,

Scott

Hi All

Can I just ask if this question came from a old/out dated source?

I would contest that none of these answers are correct at this point in time.

It is my understanding that for the last number of years we no longer subtract 2 when working out the number of available subnets and this was certainly the case in the Cisco certifications I have taken in the last 2 years.

Therefore the number of available subnets would be 4096 and the number of available host addresses would be 14, as we still subtract 2 for the subnet number and the broadcast address. As this answer is not available, then IMHO there is no correct answer listed for this question.

Now stepping back a number of years when it was still the practice not to use the first and last subnet then yes, answer C would be correct and the formula shown by ScottMac is how you arrive at this answer.

Just my thoughts on this.

Best Regards,

Michael

celticfc2007 Tue, 04/08/2008 - 06:19

Current study materials for CCENT (ICND1 exam) clarify when to subtract 2 or not:

Snip taken from Cisco ICND1 book:

**Classful routing protocol is used.

**RIP v1 or IGRP as the routing protocol.

**the no ip subnet zero command is configured. =Then subtract two.

*Classless routing protocol.

*RIP v2, EIGRP, OSPF as the routing protocol.

*ip subnet zero command is configured.

*VLSM is used.

*no other clues.

=Do not subtract two.

I watched something the other day on this issue, the instructor stated that Cisco wants CCNA candidates to know when to subtract two or not. From what I understand the older material was always subtract two.

But looking at the answers for the question, whether you can subtract the zero and broadcast subnet or not, your only gonna get 14 hosts, and only one answer has 14 hosts. I understand your point about no correct answer, but from the inconsistency in subnetting questions from different vendors I've seen, the only choice is to find the best answer.

Respectfully,

Jason

aaronfb717 Wed, 04/09/2008 - 05:20

Correct me if I am worng, but the -2 is done because the network address and the broadcast address of an IP range cannot be used correct?

If so, in what instances can the network broadcast address be assigned to a host?

Hi Arron

There is currently no instance where a network/subnet number or broadcast address should be assigned to a host.

My point in my previous post was that when I sat my CCNA (prior to the change/upgrade) we were instructed that we no longer had to subtract 2 when working out the number of subnets available as we could (and do in the real world) use the first and last subnet.

So it was my contention that none of the answers given to the question in the original post were correct in this day and age and I questioned the validity of the material.

However a number of years ago it was the correct procedure not to use the first and last subnet, so 2 was subtracted as per the resoponse by Scott.

Best Regards,

Michael

aaronfb717 Wed, 04/09/2008 - 08:28

The "-2" is done on the host side, not the subnet side correct?

If you have 16 ip addresses for a particular subnet range, you still only have 14 usable IP addresses since you can not asign the broadcast IP or the network IP address to a host.

Is my thinking correct here?

Hi Aaron

Yes you are correct. A /28 subnet yields 16 IP addresses in total. However only 14 of these are usable as host IP addresses.

Best Regards,

Michael

scottmac Wed, 04/09/2008 - 14:07
• Green, 3000 points or more

You need to follow the context of the question.

Watch for things 'How many TOTAL' versus 'How many USABLE'

Look for clues like whether they indicate that "no subnet zero" is in the config (since "subnet zero" is now the default)

They can ask a similar question a dozen different ways, and some answers will require that you subtract 2 and others won't.

The test is trying to get you to analyze the situation and make the best choice of the answers provided. In some cases, you gotta hold your nose and make the choice, whether you agree with it or not.

That is why "studying for the test" versus "learning the information" is important ... "real life networking" does not have to be squashed into a five answer multiple choice or a drag & drop.

If you know the material, there is no way they can throw you a ringer on the test ... none. The worst it gets is poor verbiage, ambiguous questions, flaky premise ... nothing that changes from not knowing.

Good Luck

Scott

Hi All

I agree with ScottMac to a point. Often when going through prep questions in your study material you will come across ambiguous questions like the one in the original post. In this case it is important that you understand the formula to use to get the answer, which Scott showed in his response to the question.

Thankfully I have not experienced any such ambiguity on any of the actual Cisco certifications I have taken so far and I could work out the answer required with no "best guess" required.

I would still maintain that there was no correct answer given in the selcetion of answers posted with the question.

Even taken the post from Celticfc2007 into account, where he writes

"Snip taken from Cisco ICND1 book:

**Classful routing protocol is used.

**RIP v1 or IGRP as the routing protocol.

**the no ip subnet zero command is configured. =Then subtract two. "

The default configuration of all routers for quite some time now is for "ip subnet zero" to be enabled, so the first and last subnet can be used, even with RIP v1 and IGRP. This helped to slow down the rate at which IPv4 IP addresses were beign depleted.

I believe that if the question wanted you to subtract 2 from the number of subnets available, then it would/should have mentioned that support for subnet zero had been disabled, either in the text of the question or by showing a snippet of a router configuration showing "no ip subnet zero" as per the snippet from the ICND 1 book. And I believe that an actual Cisco certification exam would show this.

As there was no mention of subnet zero support, I would therefore assume the default behaviour (ip subnet zero support is enabled) and would NOT subtract 2 when working out the number of subnets.

So for me the answer would have been 4096 subnets with 14 IP addresses available to hosts on each subnet. The first and last IP address on each subnet being reserved for the subnet number and subnet broadcast address respectivly.

It really irks me that there are still so many texts available for CCNA which still continue to include ambiguous questions like the one posted, which only serves to confuse the students who use these texts as certification aids.

Best Regards,

Michael

celticfc2007 Fri, 04/11/2008 - 01:31