Classless routing protocol

[chinpohpang851]

I saw frm the book stated Classful routing protocol doesn't support VLSM but a later example it uses VLSM model with the following classful address in the figure. aren't those classful addresses? Then why appeared in VLSM model....

172.16.9.1/30

172.16.2.1/23

172.16.5.1/24

[Masoud Pourshabanian]

Hello,

Those are not classful addresses.

Classful IPs

1- Class A: IP starts with 0-127 with the subnet of /8

2- Class B: IP starts with 128-191 with the subnet of /16

3- Class C: IP starts with 192-223 with the subnet of /24

4- Class D

5- Class E

IPs in your example are not considered classfull. They will be classfull with the subnet of 16 ( class B)

Hope it helps,

Masoud

[chinpohpang851]

I don't get it. The book said classful contains 3 parts Network; subnet, host and classless only 2 parts network n host.

172.16.9.1/30  = class B  /16 only

172.16.0000 1001.0000 0001/30  this IP got 14 bits subnets and 2 bits Host, so is classful right?

172.16.9.1/16 = class B; 16 bits host, no subnets  so is classless correct?

[Masoud Pourshabanian]

Just look at the first octet and subnet mask

when 172 is considered a class B IP, if its subnet is 16

172.16.9.1/16 is a class B IP because it starts with 172 and its sunbet is 16. a classful address has only network and host.  

Network :172.16          you can have 2^16 network 

Host       :0000 1001.0000 0001   2^16 host

172.16.9.1/30 it is not a classful IP. because its subnet mask is 30

Classless has network, subnet and host  sections.

Network.172.16           16 bit for network  (2^16 network)

Subnet.0000 1001.0000 00      14 bit sunbet    (2^14 subnet)           ( /30 is  network + subnet)

Host   :01     2 bit host                (2^2=4 --> only 4 host)

More examples:

192.168.1.1/24 is classful  ( first octet 192 and subnet mast 24)

192.168.1.5/30 is classless.( if it starts with 192, subnet mask must be 24 to be considered classful)

10.1.1.1/8 is classful

10.1.1.1/24 is classless

Masoud

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

Perhaps a quick review of classful vs. classless addressing would be helpful.

(IPv4) Network addresses are divided into a network portion and a host portion.  The problem doing that though, is how do we know where the dividing line is?

Well, one way, is to have an address mask that provides this information.  Another way, is to have blocks of address space imply where the division is by their binary value.  The latter is classful addressing.

For IPv4, if the first bit is a zero, it's a class A address (/8), if the first bit is one and the 2nd bit is zero, it's a class B address (/16), and if the first bit is one, the second bit is one, and the 3rd bit is zero, it's a class C address (/24).  (More details can be found here: https://en.wikipedia.org/wiki/Classful_network)

The principle reason having classful addressing, a routing protocol doesn't need to advertise the mask along with the route.  The original design also allows determination of class A, B or C by looking at just a few of the first bits (important on very old computers that had to shift/rotate a register through a carry bit to determine how a bit was set).

Subnetting is taking a classful route and subdividing its host portion.  There are "rules" for how such are used with classful routing protocols (see also Rick's post).

To your OP, the three addresses you list, as listed, are not native classful addresses, because for all three, the classful address would be a /16.  They could be subnetted classful addresses, again though, there are "rules" for how they are used with a classful routing protocol.

My guess is, your book was showing how subnetted classful addresses might be used with a classful routing protocol.  If so, the book should try to explain the "rules".

[Masoud Pourshabanian]

I agree with you. It is confusing. there is still some confusion for me. VLSM introduced first or classless routing protocol?

If VLSM first, the owner of a class A IP had to make a static route inside the network since routing protocols did not support advertising classless IPs?

[Richard Burts]

While I agree that there is some confusion about this I would suggest that this is really a discussion about history. In today's environment everyone does classless routing using VLSM. I do not think that anyone does real classful routing with Fixed Length Subnet Masking other than people experimenting in a lab while they learn the fundamentals of networking.

As far as the relationship between VLSM and classless routing I would say that they developed at the same time. You could hardly develop a network where subnets within the same network were different sizes (VLSM) without having a routing protocol that would advertise the subnet length. And you would not develop a routing protocol that advertised subnet length if there were not networks to advertise where subnets had different lengths. It is not that one came first and the other was later. VLSM and classless routing developed together at the same time.

HTH

Rick

[Masoud Pourshabanian]

Thanks Richard.

I thought the same way. I think Cisco is still using the Classful concept in books because RIP v1 and IGRP are still being used in some networks. Newer IOSs do not support RIP v1 and IGRP anymore. I guess in the future the concept of classful will fade.

Masoud

[Joseph W. Doherty]

Regarding history - this web page has a nice summary:

http://www.dscentral.in/2011/07/14/understanding-ip-address-and-subnet-mask/#.VjEDWyvdfSg

For the OP this might help in understanding IPv4 addresses:

http://www.cisco.com/c/en/us/support/docs/ip/routing-information-protocol-rip/13788-3.html

Lastly, as Rick originally noted, it's possible to use subnetted classful addresses, but I noted there are "rules".  One situation you want to avoid, or deal specially with, is "Discontiguous Networks", see:

http://www.cisco.com/c/en/us/support/docs/ip/interior-gateway-routing-protocol-igrp/13724-55.html

[Ganesh Hariharan]

I saw frm the book stated Classful routing protocol doesn't support VLSM but a later example it uses VLSM model with the following classful address in the figure. aren't those classful addresses? Then why appeared in VLSM model....

172.16.9.1/30

172.16.2.1/23

172.16.5.1/24

Hi,

Classful networks use the 'classful' subnet mask according to the leading bits in the first block of the IP address as mentioned in masoud post.

Classless IP addressing means you can use any subnetmask you want, even assigning partial blocks. For example subnet 172.16.0.0 is a class B network.

Classless routing allows me to break up this IP address into more useful segments.                                  

172.16.1.0/24               

172.16.2.0/24               

172.16.3.0/24              

172.16.4.0/24              

172.16.5.0/24               

Hope it Helps.

-GI

Rate if it Helps

[Richard Burts]

I suggest that we need to take a different approach in trying to answer this question. We need to recognize the distinction between classful addressing and classful routing. Several of the responses have offered explanations of classful addressing. But my reading of the original post is that it was asking about classful routing.  Classful routing protocols do support subnetting, but put a restriction on the use of subnets that once a subnet size is chosen for a particular network that all subnets of that network must be the same size.

The original post asks about three addresses/subnets which clearly are all in the same class B network. So they are indeed examples of VLSM. These three subnets could easily be used in a classless routing protocol (such as OSPF or EIGRP). But they would be invalid in a classful routing protocol (such as RIPv1 or IGRP). Since we do not know what book this was and do not know exactly what it was saying it is hard to explain how those addresses got involved in a discussion of a classful routing protocol.

HTH

Rick

[chinpohpang851]

Is from the book CCNA ICND2 Official exam guide. it stated Classful has 3 parts network, subnet, host. So, which 1 is true?

[Masoud Pourshabanian]

I just understood your confusion. Classful concept was used before but it is not used anymore.

When classful was used, IP address had two parts:network and subnet.

We do not use classful today, but if we look at today's IP address from classful perspective, we have network, subnet and host.

Instead of using three parts, we only consider two parts for IP now: Prefix and host.

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

"Addresses", for "Classful", Is true, but perhaps not 100%.

It's been too long, so I don't recall whether classful, in RFCs, makes subnet an "official" portion of a classful address.

First, keep in mind, the subnet portion of the address is optional, and second it's a subdivision of the host portion.

Again, subnetting a classful address, you might say, is a bit of a clever "trick".  Which is also why I noted there are a bunch of "rules" when you do so.

Subnetting a classful address, I always thought, was one of the more confusing parts of networking, until you really, really understood what it's all about.