Talk:Subnetting Explained For CCNA Exam

From Proprofs

I am studying for the Network+ test and have observed that some texts (I've read several books preparing you for

the test) state that taking n bits from the name space for subnetting gives you 2^n subnets while other texts explicitly state that you only get 2^n - 2, for you can't have a subnet with all zeros nor one with all ones within these subnetted bits. This latter claim agrees with this study guide's view in its subnetting section.  Pat Reagan on p. 205 of  "Networking with Windows 2000 and 2003 2ed."  states that this latter requirement specified in RFC 950 (which first defined subnetting) was necessary so as to not confuse the classful routers solely in use at the time.  He goes on to state, "...Today, a router can be both classful and classless at the same time.  But a classless router requires that each routing table update include the route/<prefix-length> pair to differentiate between a route to the all 0's subnet and a route to the entire network.  The same can be said for the all-1's subnet. Since it indicates the number of bits for the network and the host, you can use all 0's and all 1's for the subnet mask."

    If I get a question on the exam with the two choices 2^n and 2^n - 2 I'll still probably go with 2^n - 2 but it seems that this answer is probably outdated.