Re: subnetting confusion

> What I want to do is get a better understanding of when and why changing a
> subnet mask is or is not necessary.

You are correct in your assumption that you can carve out a subnet from you
existing network (alienate) by simply changing the mask. You can also do it
by changing the IP to be outside of the same mask. But, you do have to
understand how this works.

A mask of 255.255.255.0 means that to be on the same network (subnet), the
first three octets have to be the same. The fourth octet can be anything, up
to 254. If you change a machine's address to be
same.same.different.different, he will be off your network (on it's own
subnet). It will not be able to communicate with the original network unless
you place a router between them and manually forward the networks back and
forth.

You could also change the mask. For this, I need an IP calculator (crutch)
because I am not that good at it. However, I will use my own scenario as an
example because know it is right.

I have purchased a set of 8 static IPs. My ISP gave me the IPs, told me
which one to assign to my router, which one was used to define the network,
which one to assign to my firewall (using 3 total at this point). That
leaves me 5 I can use internally for the machines where I want to have
public IPs (not often, but it happens). So... Here is what I use:
IPs 61.50.80.45 to 61.50.80.52. You'll notice that is 8 total IPs (I made
them up for this post...hehe...). Anyway, my mask is 255.255.255.248
(really mine). The ending 248 limits my (sub)network to only those 8
addresses. If I configured an address out of that range he would be off the
original network. In this case, you are subneting using a mask.

The actual IP and the mask have both got to be taken into consideration to
determine if you are on the same subnet or not. You can't tell with just
one.

> I am looking possibly at a better way to organize my network

Use AD and OUs in Windows. Don't bother with subnets unless you can absorb
the costs of a router between each subnet.

> Maybe my confusion lays in the fact that there are always several ways to
> accomplish the same thing.

Yes, probably.

> just trying to get a grasp on the big picture really.

Suggest a good read from the bookstore or online about TCP/IP. It's all
based on binary math. Kinda tedious but helps in the long run.

-Frank


.

Relevant Pages

  • Re: subnetting confusion
    ... >> subnet mask is or is not necessary. ... >existing network by simply changing the mask. ... >I have purchased a set of 8 static IPs. ...
    (microsoft.public.windows.server.networking)
  • Re: LAN IP addresses
    ... some operating systems that the lowest and highest _subnet_ were ... "outside", you probably have no idea what mask may be used on the LAN, ... and I believe it should not matter to any network equipement either. ...
    (comp.os.linux.networking)
  • Re: Cant seem to get packets to route
    ... 5 usable IPs and broadcast IP. ... > especially when that network overlaps your assigned IP range. ... > the internet as the IP of your internet interface. ... the same /24 subnet. ...
    (comp.os.linux.networking)
  • RE: CIDR Explanation
    ... The network is 10.2.3.0 and this host is 55 on that network ... In the above example the mask is 255.255.255.0 each number represents on ... If you sub net the above example into two networks (allowing subnet ... which give a /25 CIDR network. ...
    (Security-Basics)
  • Re: Need more IP addresses
    ... My network understanding is a bit sketchy. ... If it really is one subnet, I think you merely need to use a different subnet mask to define a larger range. ... You'll need a DHCP server for each distinct subnet - most routers can provide this facility. ...
    (microsoft.public.windows.server.networking)