Classless subnetting (variable length subnettmasks)

[tuliphead]

Hello guys!
Wonder if you could help me out a little.

I am stribing to understand variable length subnetting. I am using the syngress 291 study book. Been trough the classfull subnetting. It was easy. But I can´t say I am following every bit of the examples on variable length subnetting (classless?)

I understand that variable length subnetting (classless subnetting) is to split an ip-range into segments of different sizes. I thought it would be the most easy approach to start with subnetting a class C-nettwork (even if that means you could get some strange broadcast adresses and so on ... it is fewer numbers to relate to

I thought I would just paste in the example and make a comment on what I don´t understand. Any input on this would be appreciated:

Example of Subnetting a Class C Network:

In real world scenarios, you might need to create subnets that contain only a few IP addresses.This is done to logically isolate devices on separate networks. Examples of networks with a few devices include routers on a network backbone or a point-to-point WAN connection that needs only two addresses. In these cases, you want to create small subnets to avoid wasting IP addresses.This is done with Class C network addresses, which already use 24 bits to denote the network space.
As you subnet a Class C, the number of hosts per subnet will go down by about a factor of 2, quickly reducing the number of host addresses per subnet.The maximum number of hosts in a Class C network is 254. Each subdivision results in roughly half the number of host addresses: 128, 64, 32, 16, 8, 4, 2, 1, following weighted binary values. Let’s look at an example so you can better understand the mechanics of this process.
Rather than working on a random Class C network, we will continue the previous example, using one of the Class C-type network addresses you created to meet Requirement #3. Let’s take the very last unused network address, 129.69.255.0/24, and assume you will put your routers and WAN connections on these smaller subnets.

Requirement #1:
Create One Subnet with at Least 60 Host Addresses

The current configuration uses 24 bits for the network ID.You need at least 60 host addresses on this subnet.You need six bits for host address spaces.This means you can take two bits from the fourth octet for additional network subnetting.The beginning address is 129.69.255.0/24.You take two more bits and create 129.69.255.0/26.The subnet mask is 255.255.255.192.

Using this configuration, the possible network addresses are:

Subnett 1 > 129.69.255.0/26
Subnett 2 > 129.69.255.64/26
Subnett 3 > 129.69.255.128/26
Subnett 4 > 129.69.255.192/26

You have four subnets that can be used to meet this requirement and you will use the first one to meet this requirement.

// My own comment ... so far so good. This seems pretty logic ...

Requirement #2: Create at Least Five Subnets with Up to Six Host Addresses

You can take one of the four subnets created for Requirement #1 and use it to create subnets with up to six host addresses. Begin with 129.69.255.128/26, because you might want to use 129.69.64/26 for future expansion of your 60-host subnets. Subnetting 129.69.255.128/26 to create up to six host addresses requires that you keep three bits for host addresses (23^3 = 8), so you can take up to three more bits for network addresses.The results are shown in Table 2.10.

// I am following this as well.

Subnet for six host adresses:
129.69.255.128/29 10000001.01000101.11111111.10000000
129.69.255.136/29 10000001.01000101.11111111.10001000
129.69.255.144/29 10000001.01000101.11111111.10010000
129.69.255.152/29 10000001.01000101.11111111.10011000
129.69.255.160/29 10000001.01000101.11111111.10100000
129.69.255.168/29 10000001.01000101.11111111.10101000
129.69.255.176/29 10000001.01000101.11111111.10110000
129.69.255.184/29 10000001.01000101.11111111.10111000
… …
129.69.255.248/29 10000001.01000101.11111111.11111000

By subnetting the previous subnet, you can create 16 subnets that can have up to six host addresses per subnet.You’ve met the requirement and have subnets left for future expansion.The subnet mask for this configuration is 255.255.255.248.

// New comment ... from here I loose it. In the book the it seems like they use the last five bits (of the 29 bits used for nett-id) to create 32 subnetts of 2^3-2 ip-adresses. Still they are commenting above that I can create 16 subnets ???

In terms of logic (my logic at the moment) i understand it like this:
To meet requirement nr #1 I created four subnets (listed above). I used the first of those four subnets to meet the requirement of a subnet with at least 60 host adresses.

To meet requirement nr #2, I would choose to further split one of the other four subnets into smaller subnets of 2^3-2 host adresses. The author of the book suggests saving subnett 2 for future expansions. I can´t see why that should be a problem, so I move along to subnet nr #3 (129.69.255.128/26).

Simple math: 3 bits for hosts, 3 bits for subnett-id.

This should give me the following subnets (when subnetting subnet nr #3):

000 > 129.69.255.128 / 29
001 > 129.69.255.136 / 29
010 > 129.69.255.144 / 29
011 > 129.69.255.152 / 29
100 > 129.69.255.160 / 29
101 > 129.69.255.168 / 29
110 > 129.69.255.176 / 29
111 > 129.79.255.184 / 29

Here is the first thing I don´t understand. In the example from the syngress book, they use all 5 of the last bits to create 32 different combinations (2^5), while as I see it, bit 26 and bit 27 should be fixed (static) to represent the original subnet we are subnetting from (129.69.255.128/26) ? ... and the three bits added later on will give us eight new subnets within the legal range of subnett nr #3, without violating the adress space for subnett 4? When I look at the proposed subnetting scheme in the example above (syngress) it starts with:

129.69.255.128/29 10000001.01000101.11111111.10000000

and ends with:

129.69.255.248/29 10000001.01000101.11111111.11111000

withs leads me to the conlusion that they are subnetting both the third subnet (129.69.255.128/26) and the fourth (129.69.255.192/26) ???
If this is correct, I can not see why they would choose this approach.

Anyway, if my assumption is right, and the syngress book is wrong? My subnetting of the third subnet (129.69.255.128/26) would be complete with something like this:

Subnet-ID: 129.69.255.128 / 29 IP-range 129.69.255.129 - 129.69.255.134 Broadcast: 129.69.255.135
Subnet-ID: 129.69.255.136 / 29 IP-range 129.69.255.137 - 129.69.255.142 Broadcast: 129.69.255.143
Subnet-ID: 129.69.255.144 / 29 IP-range 129.69.255.145 - 129.69.255.150 Broadcast: 129.69.255.151
Subnet-ID: 129.69.255.152 / 29 IP-range 129.69.255.153 - 129.69.255.158 Broadcast: 129.69.255.159
Subnet-ID: 129.69.255.160 / 29 IP-range 129.69.255.161 - 129.69.255.166 Broadcast: 129.69.255.167
Subnet-ID: 129.69.255.128 / 29 IP-range 129.69.255.168 - 129.69.255.174 Broadcast: 129.69.255.175
Subnet-ID: 129.69.255.176 / 29 IP-range 129.69.255.177 - 129.69.255.182 Broadcast: 129.69.255.183
Subnet-ID: 129.69.255.184 / 29 IP-range 129.69.255.185 - 129.69.255.190 Broadcast: 129.69.255.191

This would meet the second requirement in their example. From 192 (last octet) we will be in the fourth subnet and as I see it this is no point?

So my question is: Am I totally fucked up ... when it comes to understanding this? Or are the syngress approach right ... and my approach wrong?

Please comment

 

[chieftan999]

I have always used the following approach to subnet masks and subnetting:-

1: How many subnets does the subnet mask produce?
2: How many valid hosts per subnet?
3: What are the valid subnets?
4: What are the valid hosts in each subnet?
5: What is the broadcast address of each subnet?

It is very important that you understand the powers of 2 at this point.
Here is how you determine the answers to the five questions:-

255.255.255.192 (First legal class C subnet)

11111111.11111111.11111111.11000000

(Always try and remember that the RFC's state that you cannot use all 0's and all 1's so you have to minus 2 off everything).

1: How many subnets? 2x - 2 = amount of subnets. x is the amount of masked bits, or the 1's.
For example - 11000000 is 2 to the power of 2 (2 bits switched on). So there are 2 subnets in this example.

2: How many hosts per subnet? 2x - 2 = amount of hosts per subnet. x is the amount of unmasked bits or 0's.
For example - 11000000 is 2 to the power of 6 (6 bits switched off).So there are 62 hosts per subnet in
this example.

3: What are the valid subnets? 256 - Subnet mask = base number. For example, 256 - 192 = 64.

4: What are the valid hosts? Valid hosts are the numbers between the subnets minus all 0's and all 1's.

5: What is the broadcast address for each subnet? Broadcast address is all host bits turned on, which should
be the number immediately preceeding the next subnet.