subnetting question.

[Marquee]

I had this question on my CCNA exam... I wasn't quite sure what the correct answer was, none of the mult choice answers seemed right. Could someone maybe shed some light?

(Not exact phrasing, but gets the general idea across)
You have a Class C address with 26 bits of subnet. How many hosts and subnets do you have?

a) 4 subnets with 64 hosts each
b) 64 subnets with 4 hosts each
c) 2 subnets with 62 hosts each
d) 62 subnets with 2 hosts each

none of the answers seemed right to me, could someone help?
i passed the exam anyways, but this question has just been bugging me... i thought i knew subnetting pretty well!

 

[rburke]

The answer I would have marked would be answer C.

The reasoning:


Class C network bits = 24

26(subnet mask) - 24(Class C network bits) = 2 subnet bits

2^2 = 4 total subnets
but, Cisco still says that a subnet of all 0's and all 1's cannot be used so you have to subtract out 2 subents from the 4, leaving you with 2 usable subnets.

4 - 2(reserved) = 2 usable subnets


Class C host bits = 8

8(Class C host bits) - 2(subnet bits) = 6 host bits

2^6 = 64 total hosts per subnet
but, like subnets the all 0's and all 1's are reserved for subnet number and broadcast address, respectively. So you have to subtract out those 2 cases:

64-2 = 62 usable host addresses per subnet

Which comes out to answer C, 2 subnets with 62 hosts each


That is how I would have answered the question. By the way, I just passed my CCNA a couple days ago. Congrats.

Burke

 

[Marquee]

oh... i guess i overlooked the part where a subnet of all 1's or all 0's isnt valid. i picked C too, only because i was so absolutely sure there was only 62 hosts each.. but if they put a choice e) 4 subnets and 62 hosts each, i would've gotten the question wrong

thanks for the help!

 

[rburke]

Anytime...

Actually on the new IOS versions it allows you to use the subnet of all zero's, and *I think* the subnet of all 1's, but don't quote me on the all 1's one. You use the command "ip subnet-zero" to allow for the subnet of all 0's, not sure about the all 1's. Just thought I'd let you know.

Burke

 

[CCIEWANNABE]

IP Address : 192.168.1.1
Address Class : C
Network Address : 192.168.1.0

Subnet Address : 192.168.1.0
Subnet Mask : 255.255.255.192
Subnet bit mask : 110nnnnn.nnnnnnnn.nnnnnnnn.nnhhhhhh
Subnet Bits : 26
Host Bits : 6
Possible Number of Subnets : 4
Hosts per Subnet : 62

(2 *[/color]hhhhhh)- 2 = Host Answer is 62
92 * nn) = Possible Number of Subnets answer Four
Route once; switch many

 

[rburke]

Well, like I said earlier, yes there are 4 subnets, but froming straight fom the new Cisco Press CCNA 640-607 Certification guide, they still want you to disregard the subnets of all o's and all 1's, although they are now usable.

Burke

 

[wybnormal]

You disregard the all ones and all zeros unless EVERY device at layer three understands that subnet zero is valid. And that is rare indeed in a production network.

Even in the CCDA and CCDP course, you ignore subnet zero unless it's specificly said to use it.

MikeS
Find me at

http://www.packetattack.com

"Take advantage of the enemy's unreadiness, make your way by unexpected routes, and attack unguarded spots."
Sun Tzu

 

[Guest_imported]

I think your answer should be "A". I will explain. The new IOS supports CIDR(Classless Inter Domain Routing) with supports the use of 0's and 1's in the addressing scheme. Yes everyone is correct this is not the 'normal' thing to do. however we are not talking about the norm but what is actually valid give the equipment.

 

[nakona]

No, it's C.

You can specify subnet zero in your router but for exam purposes it's still 2^x-2.

Don't forget, this is the same company that puts an implicit denial in every access list by default and then tells you to override it with an implicit permit.

schizophrenic

 

[wybnormal]

No- thats being cautious.. an ACL is a security device by design and so they designed like one. Unlike MS which makes everything user friendly to the point of being completely porous to security unless you remember to put it all in place. Hopefully that will change over the next few years.

MikeS
Find me at

http://www.packetattack.com

"Take advantage of the enemy's unreadiness, make your way by unexpected routes, and attack unguarded spots."
Sun Tzu

 

[nakona]

No Mike, what I mean is...

I understand why they put in the implicit deny and I agree that it is a good idea.

It's just that they THEN tell you to end each ACL with an implicit permit.

Schizophrenic.

 

[wybnormal]

Ahh.. the old permit IP any any ? or permit TCP any/any

I can still why they do it this way. You set up your list.. you put in what you think are all the exceptions.. and then you do the any/any by hand. I dont know abou anyone else but when I have to write the permit any/any, it causes me to pause and really think about the list I just made. Which is a good thing since sometimes I find something I've overlooked.

It's just a different way of doing things. I have a feeling we will see more of this type of default in the near future.

MikeS
Find me at

http://www.packetattack.com

"Take advantage of the enemy's unreadiness, make your way by unexpected routes, and attack unguarded spots."
Sun Tzu