1

This is from a fake test before the real exam. It's a course which goes through Cisco academy - a course in networking.

  1. Use the address space 10.10.16.0/24 to design an addressing scheme. a) Split the address space into two equal-sized networks by subnetting it again.

I consider the question a bit unclear, for instance "subnetting it again". I guess they mean VLSM?

I have really tried to subnetting with VLSM this packettracer network but I always fail.

It's a network 10.10.16.0/24 that is to be subnetted effectively into two equal networks, that is not wasting any addresses. (VLSM). There are one router that is to be configured with the new parameters which should be applied to the router interface g0/0 and g0/1. There are two switches and two computers. See the image.

enter image description here

I have tried to first borrow 3 bits from 10.10.16.0/24 taking the new network 10.10.16.32 and subnetting that again taking two bits so I lastly get a 10.10.16.40/29 network.

This is now I got network 1 and I calculated network 2 to 10.10.16.48/29.

So - then packettracer wants me to take the first and last hostaddress for the router interface and the pc-computer. I did so, and yes, the PC:s can ping the default gateway and each other. But I fail completely according to the answer.

Improve this question
9
  • There is some missing information here or you are unsure of the question. When you say "two equal networks," do you mean divide the entire /24 into two networks? If you say "not wasting any addresses," how many host addresses do you need for each subnet?
    – Ron Trunk
    Dec 2, 2014 at 18:16
  • I edited the question. Yes - two equalsized network - they want me to reserve the first and last usable host-address so I guess 4 "hosts" (ip-addresses) are required.
    – Björn Hallström
    Dec 2, 2014 at 18:37
  • Ok so if you break a /24 into two equal subnets what would the new mask be?
    – Ron Trunk
    Dec 2, 2014 at 18:47
  • I borrowed 3 bits at a beginning from the last octet - then advancing to a new mask of 27. Then I take one of these new networks taking another 2 bits getting a mask of 29.
    – Björn Hallström
    Dec 2, 2014 at 19:06
  • Why do you do that? Why 3 bits? You only need two nets
    – Ron Trunk
    Dec 2, 2014 at 19:08

1 Answer 1

Reset to default
3

Subnetting will always be a "2n" bit-wise operation. Subnets fall on specific bit boundaries, no matter what. You can't subnet on arbitrary boundaries or addresses. Just think, "always equal proportions." Not any different than cutting a pie into equal proportions.

I always use a /24 as a "base" to start from. Again, this is a 2n (ie multiply or divide by 2) operation, so you go from a single subnet, to two halves, or four quarters, or eighths, sixteenths, etc.

If you split a /24 into two halves, each one of those halves will be a /25 subnet.

If you split a /24 into four quarters, each one of those quarters is a /26 subnet.

If you split a /24 into eighths, each eighth is a /27 subnet.

The converse is true when going in the opposite direction.

If you combine two /24's into a single subnet, you end up with a /23.

If you combine four /24's into a single subnet, you end up with a /22.

If you combine eight /24's into a single subnet, you end up with a /21.

To answer your question, if you split 10.10.16.0/24 into two equal-sized subnets, you end up with:

1. 10.10.16.0/25
2. 10.10.16.128/25

The first and last usable from each subnet would be:

1. 10.10.16.1/25 and 10.10.16.126/25
2. 10.10.16.129/25 and 10.10.16.254/25

Hope this helps.

Improve this answer
13
  • Thanks I'll try this at once and I think @Ron is correct that "I am overthinking" this. In the Answer the label for this part is VLSM - that means Variable Length Subnet Masking. But if this is the answer how could it fall under VLSM? this is the reason why I have gone to far "thinking".
    – Björn Hallström
    Dec 2, 2014 at 19:36
  • VLSM is a term that was only coined to account for routing protocols that could carry the subnet mask in routing updates, and when classful addressing/routing was being phased out. Before VLSM, every single subnet had to have the same subnet mask in a network. After classless routing and addressing happened, the term VLSM was introduced to indicate that that one had more freedom when choosing the size of subnets based on need (variable length subnet mask). In other words, using a /24 for a point to point link was a waste, but was a requirement with classful routing.
    – John Jensen
    Dec 2, 2014 at 19:38
  • thanks a lot! it works!!!! But the last usable host for NW1 is 126 (127 is broadcast)
    – Björn Hallström
    Dec 2, 2014 at 19:53
  • Right you are. Good catch. :-)
    – John Jensen
    Dec 2, 2014 at 19:54
  • Ok - so the term VLSM is like more "freedom" to subnet, i.e. IF neccessary - do subnet a subnet.
    – Björn Hallström
    Dec 2, 2014 at 19:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.