Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

This question already has an answer here:

Assuming CIDR and classless addressing, the CCNA books I'm reading say

  1. Pearson 640-822 official cert guide 2012, pgs 403-404: According to their method S = P - N (subnet, prefix, host bits respectively). In this case S = 8 - 24 < 0.
  2. CCNA guide 7th literally says in its subnettting chapter that the prefixes /8 through /15 can only be used with class A addresses.

So for example, a subnet 220.0.0.0 cannot have a net mask of 255.0.0.0. Does that really make sense and hold true in the real world? Can a network or subnet really be assigned an IP address with any prefix?

I'm practicing on a c2691 router and it takes the ip route 220.0.0.0 255.0.0.0 Serial 0/0 command no problem. If 220.0.0.0 is a subnet, then what is its network?

It also takes ip route 192.0.0.0 192.0.0.0 Serial 0/0. Here we have a "class C" subnet with a mask actually less than 255.0.0.0! Is that possible in the real world?

I just tried adding some routes to my Cisco router and the following output is possible. You can assign the same IP 4.0.0.0 more than once in the same route table. Interestingly, it seems that it won't say "subnetted" unless the route you added has mask at least 255.0.0.0.

R2(config)#do show ip route
...

     4.0.0.0/16 is subnetted, 1 subnets
S       4.0.0.0 [1/0] via 192.168.1.1
S*   198.252.206.0/24 [1/0] via 192.168.1.1
     10.0.0.0/24 is subnetted, 3 subnets
O       10.1.6.0 [110/128] via 10.1.4.2, 04:54:56, Serial0/1
C       10.1.5.0 is directly connected, Serial0/0
C       10.1.4.0 is directly connected, Serial0/1
C    192.168.1.0/24 is directly connected, FastEthernet0/0
S    2.0.0.0/7 [1/0] via 192.168.1.1
S    4.0.0.0/7 [1/0] via 192.168.1.1
S    4.0.0.0/6 [1/0] via 192.168.1.1
S    8.0.0.0/6 [1/0] via 192.168.1.1
S    12.0.0.0/6 [1/0] via 192.168.1.1
share|improve this question

marked as duplicate by Michael Hampton, EEAA, Jacob, MDMarra, Wesley Apr 5 '13 at 3:05

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

    
Your book is a history lesson. –  MDMarra Apr 4 '13 at 22:26
    
2012 is already history. I starting studying less than a month ago and now they've already updated the exam e.g. my book has RIP v2 and now it's OSPF. and 640-822 is no longer a choice. Now only 100-101 is offered. cisco.com/web/learning/exams/list/icnd1b.html#~Topics –  server-builder Apr 5 '13 at 7:55

3 Answers 3

You mention classless addressing, then immediately mention a Class C network? Your book is outdated, class-based addressing is no more.

While there are definitely certain conventions, you can technically assign any subnet mask to any IP address you want. There are many reasons why this may be a bad idea, however.

Regarding your questions 222.0.0.0 is the network address (network and subnet address are being used interchangeably here).

share|improve this answer
1  
CCNA books (and ICND) still mention class-based addressing. not sure why maybe for the same reason they still talk about frame relay. –  Epaphus Apr 4 '13 at 22:22
    
Could be that the info in the book reall is outdated. It's the Pearson 640-822 official cert guide 2012. pgs 403-404. According to their method S = P - N (subnet, prefix, host bits respectively). In this case S = 8 - 24 < 0. Is 222.0.0.0 really the network address?? I don't get the math on that. 220.0.0.0 is the subnet, and I don't see how the network can ever have a value (in prefix part of IP) greater than its subnet. –  server-builder Apr 4 '13 at 22:29
    
@Jim G. The book shows that method of finding subnets (which btw includes class in calculations) immediately after their definition of CIDR. I see that any subnet mask can be assigned to any host IP, but my question is more specifically, can any subnet mask be assigned to any subnet or network IP? –  server-builder Apr 4 '13 at 22:33
    
@Epaphus People are still using frame relay, believe it or not. –  Michael Hampton Apr 4 '13 at 23:14
    
Exactly, Some people still refer to a /24 as a Class C network. Which is most likely why it is still covered in the books. –  Epaphus Apr 5 '13 at 8:34

Your first sentence contains two conflicting pieces of information. You say both "assuming CIDR" and "a class C network." If you are considering CIDR, there is no "Class C" and 220.0.0.0 could have a mask of 255.0.0.0. If you are dealing with 220.0.0.0 as a class C network, then you could not use a 255.0.0.0 mask.

Cisco equipment can be configured either way. For some time you needed to add the "ip classless" statement if you wanted CIDR, but that has been the default setting for probably at least 10-15 years at this point. You can however turn it off ("no ip classless").

While it is almost universally true that networks are now classless, it is still good to understand the class distinction.

There are some "corner cases" where this knowledge is useful. For instance, in the past 3 years, I have come across two consumer devices and one SCADA device that while they learned their IP address and mask through DHCP properly, behaved in a classful manner on the network (and as such had issues). This knowledge allowed me to understand the way they were actually behaving and provide a resolution more quickly (and in one case communicate accurately what was happening to the vendor to get a fix).

So while classes may be obsolete from a network perspective, you can come across devices where the software was was written by developers mistakenly still referencing old code or documentation that is based on classes.

Edit: Now that you have clarified your question some, when you are talking strictly on the technical mechanics, you are correct in saying you cannot have a "subnet" of 220.0.0.0 using a mask of 255.0.0.0. A subnet in this strictly technical sense is referring to the number of additional bits used to subnet the original class. Your example is more accurately speaking a supernet (combining more than one smaller network into a larger one), although the standard for supernetting was meant more to bridge the gap between class B and class C.

In today's networks, generally you can both subnet and supernet and generally classless addressing is considered to include both.

share|improve this answer

Forget about classes, classful networks and their subnets. Since CIDR got introduced in 1993 we just have prefixes and more specific prefixes.

share|improve this answer

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