[roger_smith]

I had a good time learning this today during my CCNA self-study, and I thought I"d share it.

So everyone understands what an IP address is, I assume. In case you don't, an IP address (IPv4 specifically) is a set of 4 numbers divided by a decimal, and usually looks like

192.168.10.1

Familiar, right?

An IP address is a logical address used to identify a computer on a network. This is different from a physical address in that the physical address of a computer (aka MAC address) is burned into the ROM of the network adapter.

Anyway, IP addresses identify 2 things in them, the network address and the host address. Using the example above we have the network 192.168.10 and the host address is 1.

IP addresses have 3 main classes they use, A, B, and C. The class of the IP address dictates which part of the address is the network ID and which part identifies the host. Let's look at another example.

10.10.142.183

In this case, the 10.x.x.x IP is a Class A IP address. The first octet identifies the network, and the remaining 3 octets specify the individual hosts on the network.

172.16.223.86

This is a Class B address, the first 2 octets identify the network, the remaining 2 identify the host.

IP Class Ranges first octet
A 1-126
B 128-191
C 192-223

So, using the table above, which class IP is the following?

14.28.72.183

Did you say Class A? Then you're right.

How about this one?

183.82.173.69

Class B.

You get the idea here, right?


Can you count binary? If not, you're about to learn, because that's how IP addresses exist.

Quick and dirty binary Math:

A single Byte of information is made up of 8 bits. A bit is a 1 or a 0, a yes or no, a electrical charge, or a lack of electrical. So, a Byte might be something like:

00101010

Now, what if I told that that binary string I wrote represented as a decimal number would be 42? What if I also told you I can figure that out in my head fairly quickly?

I'll show you how.

Above each bit (one or zero), imagine a table that looks like this:

128|64|32|16|8|4|2|1

In the case of binary, a 1 is true, and a 0 is false. Let's plug some numbers into the table we made:

128|64|32|16|8|4|2|1
1 | 0 | 0| 1 | 1| 1|1| 0

Now, we add up each number that has a 1 under it, so

128+16+8+4+2 = 158

Do that 3 more times and you have an IP address!

That's enough for now, this is a bit of a ramble. If you want to know how the Class A, B and C are dictated, let me know. That was the whole point I wanted to make in this thread, but I feel like it's gotten too long.

[Nil]

I think it's worth noting that this is known as classful networking and I'm pretty sure has been obsolete since the mid 90's. Also, class A would technically be from 0-127.

[Aeolian]

Informative & was a good read.

I think you should add what IP actually stands for. (Internet Protocol)

Thanks for the share.

[roger_smith]

I think it's worth noting that this is known as classful networking and I'm pretty sure has been obsolete since the mid 90's. Also, class A would technically be from 0-127.

127 range is reserved for loopback and not typically included in the class listing

http://www.vlsm-calc.net/ipclasses.php


While CIDR is basically law of the land these days, you have to understand this foundation to understand subnetting and understand how routers see IP addresses. That was the main point I didn't get to yet and likely will when I'm less tired.

[Nil]

I think it's worth noting that this is known as classful networking and I'm pretty sure has been obsolete since the mid 90's. Also, class A would technically be from 0-127.

127 range is reserved for loopback and not typically included in the class listing

http://www.vlsm-calc.net/ipclasses.php


While CIDR is basically law of the land these days, you have to understand this foundation to understand subnetting and understand how routers see IP addresses. That was the main point I didn't get to yet and likely will when I'm less tired.

That's why I said technically. If you go off how the classes are organized by the prefixes, then 127 would fall into class A, although you're right, loopback and shit. And I don't see any reason on how understanding classful before CIDR would help in understanding the current architecture of the of the network. Maybe purely from a historical standpoint it is worth being aware of.

[roger_smith]

I think it's worth noting that this is known as classful networking and I'm pretty sure has been obsolete since the mid 90's. Also, class A would technically be from 0-127.

127 range is reserved for loopback and not typically included in the class listing

http://www.vlsm-calc.net/ipclasses.php


While CIDR is basically law of the land these days, you have to understand this foundation to understand subnetting and understand how routers see IP addresses. That was the main point I didn't get to yet and likely will when I'm less tired.

That's why I said technically. If you go off how the classes are organized by the prefixes, then 127 would fall into class A, although you're right, loopback and shit. And I don't see any reason on how understanding classful before CIDR would help in understanding the current architecture of the of the network. Maybe purely from a historical standpoint it is worth being aware of.

Classful networking still exists, it's the default subnet mask. CIDR requires an understanding of the bits so you can understand how a to subnet a network range.

https://learningnetwork.cisco.com/thread/34420

Class A would be a /8 if you didn't modify the subnet mask, Class B is a /16, and Class C is a /24. the /X number represents the number of bits in the "on" position of the subnet mask.

So default subnet mask for a class A address is 255.0.0.0, which is 11111111.00000000.00000000.00000000. You have 8 bits w/ a 1. A /16 would be 11111111.11111111.00000000.00000000, or 255.255.0.0.
If you had a Class B address with a subnet mask of 255.255.128.0, that would look like 11111111.11111111.10000000.00000000. This would also be represented in CIDR notation as a /17, because there are 17 bits in the "on" position.

[Nil]

127 range is reserved for loopback and not typically included in the class listing

http://www.vlsm-calc.net/ipclasses.php


While CIDR is basically law of the land these days, you have to understand this foundation to understand subnetting and understand how routers see IP addresses. That was the main point I didn't get to yet and likely will when I'm less tired.

That's why I said technically. If you go off how the classes are organized by the prefixes, then 127 would fall into class A, although you're right, loopback and shit. And I don't see any reason on how understanding classful before CIDR would help in understanding the current architecture of the of the network. Maybe purely from a historical standpoint it is worth being aware of.

Classful networking still exists, it's the default subnet mask. CIDR requires an understanding of the bits so you can understand how a to subnet a network range.

https://learningnetwork.cisco.com/thread/34420

Class A would be a /8 if you didn't modify the subnet mask, Class B is a /16, and Class C is a /24. the /X number represents the number of bits in the "on" position of the subnet mask.

So default subnet mask for a class A address is 255.0.0.0, which is 11111111.00000000.00000000.00000000. You have 8 bits w/ a 1. A /16 would be 11111111.11111111.00000000.00000000, or 255.255.0.0.
If you had a Class B address with a subnet mask of 255.255.128.0, that would look like 11111111.11111111.10000000.00000000. This would also be represented in CIDR notation as a /17, because there are 17 bits in the "on" position.

One isn't required to understand the other. In fact, it would probably be better if off to do away with classifying anything with a class. Firstly, most people in the industry will laugh. Secondly, working backwards from a CIDR address won't always correlate with the correct class (A, B, C) based on the mask and just serves to confuse people. You're saying it still exists, but that's just an unneeded correlation between an explicitly stated subnet mask to a particular class, and in no way gives any kind of importance to the classes in the current architecture.

[roger_smith]

That's why I said technically. If you go off how the classes are organized by the prefixes, then 127 would fall into class A, although you're right, loopback and shit. And I don't see any reason on how understanding classful before CIDR would help in understanding the current architecture of the of the network. Maybe purely from a historical standpoint it is worth being aware of.

Classful networking still exists, it's the default subnet mask. CIDR requires an understanding of the bits so you can understand how a to subnet a network range.

https://learningnetwork.cisco.com/thread/34420

Class A would be a /8 if you didn't modify the subnet mask, Class B is a /16, and Class C is a /24. the /X number represents the number of bits in the "on" position of the subnet mask.

So default subnet mask for a class A address is 255.0.0.0, which is 11111111.00000000.00000000.00000000. You have 8 bits w/ a 1. A /16 would be 11111111.11111111.00000000.00000000, or 255.255.0.0.
If you had a Class B address with a subnet mask of 255.255.128.0, that would look like 11111111.11111111.10000000.00000000. This would also be represented in CIDR notation as a /17, because there are 17 bits in the "on" position.

One isn't required to understand the other. In fact, it would probably be better if off to do away with classifying anything with a class. Firstly, most people in the industry will laugh. Secondly, working backwards from a CIDR won't always correlate with the correct class (A, B, C) and just serves to confuse people.  You're saying it still exists, but that's just an unneeded correlation between an explicitly stated subnet mask to a particular class, and in no way gives any kind of importance to the classes in the current architecture.

Well I guess all I can say is that you may be more well versed in networking than I am. I intended this to be a post about some basic IP stuff, and it's still that, and still accurate.

This page seems to more or less reaffirm what I've said and how I understand it as of this point in time.

https://ciscoskills.net/2011/06/01/class...s-routing/


I guess I don't understand why they would dedicate an entire chapter to understanding the information I outlined in the original post if it wasn't somehow relevant or important foundational information. Perhaps the author was paid by the word and needed a little more money for a down payment on something :p

[Nil]

Classful networking still exists, it's the default subnet mask. CIDR requires an understanding of the bits so you can understand how a to subnet a network range.

https://learningnetwork.cisco.com/thread/34420

Class A would be a /8 if you didn't modify the subnet mask, Class B is a /16, and Class C is a /24. the /X number represents the number of bits in the "on" position of the subnet mask.

So default subnet mask for a class A address is 255.0.0.0, which is 11111111.00000000.00000000.00000000. You have 8 bits w/ a 1. A /16 would be 11111111.11111111.00000000.00000000, or 255.255.0.0.
If you had a Class B address with a subnet mask of 255.255.128.0, that would look like 11111111.11111111.10000000.00000000. This would also be represented in CIDR notation as a /17, because there are 17 bits in the "on" position.

One isn't required to understand the other. In fact, it would probably be better if off to do away with classifying anything with a class. Firstly, most people in the industry will laugh. Secondly, working backwards from a CIDR won't always correlate with the correct class (A, B, C) and just serves to confuse people.  You're saying it still exists, but that's just an unneeded correlation between an explicitly stated subnet mask to a particular class, and in no way gives any kind of importance to the classes in the current architecture.

Well I guess all I can say is that you may be more well versed in networking than I am. I intended this to be a post about some basic IP stuff, and it's still that, and still accurate.

This page seems to more or less reaffirm what I've said and how I understand it as of this point in time.

https://ciscoskills.net/2011/06/01/class...s-routing/


I guess I don't understand why they would dedicate an entire chapter to understanding the information I outlined in the original post if it wasn't somehow relevant or important foundational information. Perhaps the author was paid by the word and needed a little more money for a down payment on something :p

No worries. I thought just a disclaimer would be nice in the OP for newcomers. I still think it's important to know things from a historical perspective regardless if we can agree if it helps currently

And Cisco has always been known to test (probably a little to much) on classful addressing and they like to overplay it's importance. I don't know much about Cisco, but they could have tons of old equipment and expect you to be working with older protocols. Or they wanted to add more questions to their exam. Who knows.

[mothered]

I recall going through something similar to this around 10 years ago.

In terms of the basics, a well written tutorial.
Good job.