Subnetting question
gouki2005
Posts: 197Member
in CCNA & CCENT
How many subnets and hosts per subnet can you get from the network 172.17.0.0/23?
ok the host number is easy is /23 so 510 usable address but dont remenber how obtain the subnets number??
ok the host number is easy is /23 so 510 usable address but dont remenber how obtain the subnets number??
0
Comments
is
7
because we use 2^9 for the host and 2^7 for the subnets
So 2 ^ 7 = 128
Hmm to some extent, but I think it does some good for newbies. For example... if you tell a new guy "Hey, your classful address is 172.20.1.0 /16, but your going to use a classless address of 172.20.1.0 /24, so you <insert word> from the host portion of the classful mask for the network portion..."
Just remember for the hosts that it would be (2^9)-2, since you can not use the network address and the broadcast address. If you already knew this, then disregard.
And I believe that would confuse things more.
The reason we had classfull routing is because IGRP and RIP assumed the subnet mask based on the first octet. Now that we have RIPv2, EIGRP, IS-IS, OSFP and BGP that include the subnet mask the idea of class or classless addresses doesn't exist. I think since IOS 12.2 IP classless has been enabled by default.
When students are learning about IP address history and the reasons we now use NAT. Then classfull routing and address class can be considered but not if we're talking about subnetting and routing today.
Cheers,
Pete
Actually, we had classful routing because thats how IP worked in those days. Subnetting and VLSM didn't exist when RIPv1 and IGRP were developed.
And I would say that they assumed the subnet mask based on the classful boundary, not just the first octet. Though the first octet IS where those boundaries exist.
Bad wording on my part and I'm noticing some irony here.
Like you say, it's not the routing process that assumed the subnet mask. It's that the first octet was used to define the subnet mask.
Edit: I'm repeating myself here but I think it's really important when teaching to explain classfull addresses and routing aren't used any more and shouldn't be used to teach how classless routing works. The two concepts should be kept separate. Using terms like "borrow" cloud the process and imply that the class of an address is still important when it's not.
But ultimately my point was that if we're talking classful networking, I think "classful boundary" describes what's going on better than "based on the first octet." Remember, this is the forum where we nitpick little details and argue them for 5 pages.
I agree whole heartedly it should be left out.
Also I reckon leading bits is better :P
Heck, I wouldn't be shocked if the broke up CCNA into two parts, history and now.
RIP and IGRP, especially IGRP is barely touched on CCNA but you still need to know all aspects of it because you'll never know if it appears on your exam.
Subnetting for IPv4 will become a history lesson one day too when everyone uses IPv6. So will NAT'ing.
But it will take years, even decades I think before all networks are 100% purely on IPv6 infrastructure.
Luckily, Cisco has removed ISDN which I'm a little shocked but I'm not complaining.
Frame Relay will follow too one day, but just not yet. Frame Relay has so much to teach I don't think they want it removed....