Ok I am fried. subnetting.

CoryS

Alright, I can do a few things with subnetting but this one completely dissolves my brain.. Sad I know but here it is..

If you have existing computers with the ips of

10.2.12.1
10.2.41.23
10.2.41.100
10.2.41.101

Which subnet mask would you give them

255.0.0.0
255.255.0.0
255.255.255.0

I know its a class A network so it starts with 255.0.0.0 but I am having a very hard time figuring out why the answer is 255.255.0.0

Going through the book and stumbling through learntosubnet has cramped my style.. Any sort of detailed explanation on this one would be very helpful for me and I am sure countless others..

I understand how to figure out a given subnet mask when requiring a certain amount of hosts and in reverse just fine but when it comes to actual addressing I dont get how people determine how to increment the ranges (ex 1-30, 31-60 etc..)

Again any help is GREATLY appreciated.

Slowhand

I could go through and explain the binary, show you all kinds of technical details of why the answer is 255.255.0.0, but the short answer is this (since your brain is fried, and I know that feeling all too well):

The first two octets of all the addresses are the same. They're all 10.2.x.x networks, so the host addresses are anything in the .x.x portion. You can "cover up" the 10.2 portion (network portion) with the subnet mask.

10.2.x.x
255.255.0.0

In this case, their class doesn't matter. It's just a matter of how many octets you want to "block out", and how many you want to use for the network addresses.

tomiwa

If I understand your question, I believe the reason why you have a class B subnet mask as the correct answer is may be for the following reasons

1. Probably you were given something like 10.2.x.x/16. (Class A = 8 bit, Class B = 16 bit, Class C = 24bit)

2. Another reason which is most obvious in this case is just so that you can limit the number of host to 65534 regardless of the ip being a Class A.

If you go back to the chapter on subnetting, you'll see that Subnet Mask basically helps you know what network you belon to and it helps limit you network to certain number of hosts that can be supported.

This is how Subnet works (Using Class C as example)

255.255.255.0

255 in binaries is 11111111 (8bit) so since there is "three 8bits" then you have 8x3 = 24bit.
0 in binaries is 00000000 (8bit). The number of 0s indicates how many ip addresses you can have in a subnet... then you will have 2^8bits = 256 ipaddress. the first and last ipaddresses cannot be used. First IP address is the SUBNET ADDRESS and Last IP address is the BROADCAST ADDRESS, which means if we minus these two IP addresses we are left with 254 ipaddress available for your hosts.

I wouldn't like to go further just in case this ain't what you looking for exactly but if it is then I won't mind explaining how subnet works, and how to use it.

Not so sure if I answered your question. If I didn't, if you won't mind posting the question as is then maybe I can help.

CoryS

sadly that was the question in its entirety, if I had been given a little more information I may have stumbled to the answer on my own but I digress. In your response you mention 10.2.x.x can we further on that example, how would one begin addressing something like that then? How would you know what to put after the 10.2.

I really appreciate the replies guys, I am sure others will get use out of them as well. Thanks again!

CoryS

For anyone that wants to see what I am refering to, if you have the MSPress book for the 291 its page 2-24 Exercise 1

tomiwa

just take "x.x" as any number. in the example you had one of them was 41.100.

so since you have 10.2.41.100 and your subnet is 255.255.0.0

The first thing you should take note is that your network will only support 65534 host (that is 2^16bits of 0s)

To know your ip range you need to do the ANDING process. you change your ip addresses to binary and then you have to AND it with your subnet mask.

geezer

I know it can be hard to see sometimes and mindblocks don't help.

Basically, to know the range of hosts you need to find your subnet block size which will dictate your 'range' of hosts.

Your block size is found by subtracting the subnet from 256 or seeing where the lowest order bit becomes a 1 if that makes sense (do the Binary and see). So, 256-255 = 1 block size and each subnet will increase by this value giving you 10.0, 10.1, 10.2, 10.3 and so on up to 255. The hosts would be an increase to the lowest order bit by one: 10.0.0.1, for example, up to 10.0.255.254 as the last valid host (10.0.255.255 is the broadcast address for the 10.0 subnet)

In the example you show you could have a subnet of 255.254.0.0 and it would still work because your block size is 256-254 (second octet) = 2. Your steps would be 0 (ip Subnet-zero),2, 4, 6,....250, 252, 254.

Like Todd Lammle says "...know your block sizes..."

P.S. In your example you could use the 255.255.255.0 mask if the third octect was 41 but since one is 12 this can't be used.

CoryS

Awesome replies, this will really help me out. I can do all the binary work just fine and have just recently been introduced to the AND concept. So if I had a 255.255.252.0, would the block size be 4, so 192.168.4.x 192.168.8.x etc..

geezer

Bingo! By jove he's got it!

CoryS

You gents have done more for me then many hours reviewing books and videos.. Many many thanks.

Slowhand

Not a problem. Sometimes all it takes is a "plain English" description, despite the chapters and chapters you'll find on a subject in the books. Never be afraid to ask questions, that's why we're all here: to learn and teach.

CoryS

"P.S. In your example you could use the 255.255.255.0 mask if the third octect was 41 but since one is 12 this can't be used."

Ok this one threw me for a loop and I am trying to figure out why that would have been possible, and presumably if it is possble, you would use this only if you needed to subdivide your network more thoroughly and for no other reason, correct?

Thanks for the tips again, I pretty much went to bed with this and woke up and started pounding away again, its been years since i had my CCNA courses and I remember why I hated this part of the class.

JdotQ

Check out this thread I started a while back - sort of covers the same idea;

Clients on same subnet?

Hopefully some of the posts/information will help you

And ask away if you still have questions

geezer

CoryS wrote:

"P.S. In your example you could use the 255.255.255.0 mask if the third octect was 41 but since one is 12 this can't be used."

Ok this one threw me for a loop and I am trying to figure out why that would have been possible, and presumably if it is possble, you would use this only if you needed to subdivide your network more thoroughly and for no other reason, correct?

Thanks for the tips again, I pretty much went to bed with this and woke up and started pounding away again, its been years since i had my CCNA courses and I remember why I hated this part of the class.

Glad to help and be helped of course!!

The mask is there to highlight which portion of the IP address is network and which part is host, i.e. where the mask is a zero then the corresponding bits in the IP address are host bits. So if the first three octets of the IP address are the same and only the last octet differs then the first three octets could be in the same network of: 10.2.41.x (x = host bits of 1 through to 254 [255=b/cast]).

Remember that 256-255=1 and this is the block size for the subnet of 10.2.41 and the next block would be 10.2.42.x and so forth until you reach the next subnet of 10.3.0.0 but this might be more than you need to know at present so ignore if unsure or it confuses you.

CoryS

Fantastic. This clears it up. If I could buy you guys a round I would Take care.