Technical question of networks

- well for me anyway.

Given I understand basic 8 octlet network addressing and subnets; can
someone clarify the notation w.x.y.z/a for me. I've determined 'a' as a
subnet number (does /16 gives me the whole range w.x.y.??) but what is the
basis; is it always 0 OR a or is it in segments? I wonder if this notation
is older than describing subnets themselves or a modern equivalent. Taa.

"William4" <(E-Mail Removed)> wrote in message
news:n9lJg.129$(E-Mail Removed)...
>- well for me anyway.
>
> Given I understand basic 8 octlet network addressing and subnets; can
> someone clarify the notation w.x.y.z/a for me. I've determined 'a' as a
> subnet number (does /16 gives me the whole range w.x.y.??) but what is the
> basis; is it always 0 OR a or is it in segments?

Written in binary, a subnet mask is n ones followed by 32-n zeros.
Therefore, a subnet mask can be specified by simply stating n, the number of
ones. That is what the notation is: a concise way to describe a
(sub)network. For instance, 192.168.0.0/255.255.255.0 means the same as
192.168.0.0/24.

Alex

Alex Fraser wrote:
> "William4" <(E-Mail Removed)> wrote in message
> news:n9lJg.129$(E-Mail Removed)...
>> - well for me anyway.
>>
>> Given I understand basic 8 octlet network addressing and subnets; can
>> someone clarify the notation w.x.y.z/a for me. I've determined 'a'
>> as a subnet number (does /16 gives me the whole range w.x.y.??) but
>> what is the basis; is it always 0 OR a or is it in segments?
>
> Written in binary, a subnet mask is n ones followed by 32-n zeros.
> Therefore, a subnet mask can be specified by simply stating n, the
> number of ones. That is what the notation is: a concise way to
> describe a (sub)network. For instance, 192.168.0.0/255.255.255.0
> means the same as 192.168.0.0/24.
>
Yes.
The common subnet masks are
24 bit - class c - 255 available addresses
16 bit - class b - 65 thousand odd addresses
8 bit - class a - 16 million odd addresses per network

--
Alex

Piece by piece the penguins have taken my sanity
www.drzoidberg.co.uk www.ebayfaq.co.uk

"Alex Fraser" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> "William4" <(E-Mail Removed)> wrote in message
> news:n9lJg.129$(E-Mail Removed)...
>>- well for me anyway.
>>
>> Given I understand basic 8 octlet network addressing and subnets; can
>> someone clarify the notation w.x.y.z/a for me. I've determined 'a' as a
>> subnet number (does /16 gives me the whole range w.x.y.??) but what is
>> the
>> basis; is it always 0 OR a or is it in segments?
>
> Written in binary, a subnet mask is n ones followed by 32-n zeros.
> Therefore, a subnet mask can be specified by simply stating n, the number
> of ones. That is what the notation is: a concise way to describe a
> (sub)network. For instance, 192.168.0.0/255.255.255.0 means the same as
> 192.168.0.0/24.
>
> Alex
Thanks, I think I was working this backwards (from the lsb). So anything/0
is the whole caboodle and anything/31 is just two address then? - will they
always be .0 & .1 or can you have w.x.y.120/31 for .121 & .121 ?

"William4" <(E-Mail Removed)> wrote in message
news:IZmJg.919$(E-Mail Removed)...
> "Alex Fraser" <(E-Mail Removed)> wrote in message
> news:(E-Mail Removed)...
[snip]
>> That is what the notation is: a concise way to describe a (sub)network.
>> For instance, 192.168.0.0/255.255.255.0 means the same as 192.168.0.0/24.
>
> Thanks, I think I was working this backwards (from the lsb). So
> anything/0 is the whole caboodle and anything/31 is just two address then?
> - will they always be .0 & .1 or can you have w.x.y.120/31 for .121
> & .121 ?

A /30 is the smallest useful subnet you can have, but yes, you could have
w.x.y.120/30 for w.x.y.120-123. However, the address ANDed with the
complement of the subnet mask must be zero. So, for example, w.x.y.121/30 is
not valid (use a monospaced font):

mask = /30 (255.255.255.252) = 11111111 11111111 11111111 11111100
~mask = 00000000 00000000 00000000 00000011
address = w.x.y.121 = wwwwwwww xxxxxxxx yyyyyyyy 01111001
address & ~mask = 00000000 00000000 00000000 00000001

This is equivalent to saying that for /30, the last octet of the address
must be a multiple of four. For /29, it must be a multiple of eight, and so
on. For /24 down to /17, similar applies to the third octet and the last
octet must be zero, eg for /22, the third octet must be a multiple of four.
You can probably guess the rest .

Alex