|
Home > Archive > microsoft.public.cert.exams.mcse > September 2003 > Supernett?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
|
|
| snoopy51 2003-09-27, 4:23 pm |
|
You are 100% correct!, this means 24 bits are used for
network id and 8 bits are used for host id. How ever
supernetting is different:
Supernetting is used when multiple, consecutive network
addresses of the same Class are combined into blocks. For
example, if you need to assign 199.1.1.x and 199.1.2.x to
the same physical network and want to change subnet mask.
You would use the subnet mask of 255.255.254.0 and NOT
use DHCP superscopes to assign addresses from DHCP server
This is realy confusing to me too!
>-----Original Message-----
>192.168.14.0/24
>Can anyone please tell me what 24 is for? I have
>forgotten it.
>I think it stands for 24 bits for the network????
>I am prepairing for 70-215 and believe I have to
>understand this.
>:0)
>Thanks!!
>.
>
| |
| Herb Martin 2003-09-27, 5:23 pm |
| > You are 100% correct!, this means 24 bits are used for
> network id and 8 bits are used for host id. How ever
> supernetting is different:
Snoopy has the idea right but let's clean up the details a bit... (supernet
joke there.)
> Supernetting is used when multiple, consecutive network
> addresses of the same Class are combined into blocks. For
> example, if you need to assign 199.1.1.x and 199.1.2.x to
> the same physical network and want to change subnet mask.
Yes, except the "blocks" must not only be consecutive BUT ALSO
start on a "power of 2" and "occupy that power" - -then you subtract
that "power" from the mask to combine them:
So with "1-bit" as Snoopy showed, you must start on a "multiple of
2" (2^1) so you can combine 199.1.0.0 with 199.1.1.0 OR you can
combine 199.1.4.0 with 199.1.5.0 but not staring on 199.1.1.0,
or anything else that is odd in the third octet.
> You would use the subnet mask of 255.255.254.0 and NOT
> use DHCP superscopes to assign addresses from DHCP server
If you subtract 2-bits, /22 or 255.255.252.0 then you have to start
on a "Multiple of 4" and combine EXACTLY 4 nets, e.g., 199.1.12.0-
199.1.15.0 etc.
--
Herb Martin
| |
| Dolphus Brown 2003-09-28, 5:23 pm |
| It is much easier to understand if you convert the IP
address and subnet mask into binary first (which is how
the computer sees it). You obtain the Network ID portion
of an IP address by performing a logical "AND" of the IP
address and subnet mask. For example, the subnet mask
255.255.255.0 in binary is written
11111111.11111111.11111111.00000000. So, any bit in the
IP address that falls where there is a "1" in the subnet
mask(first 24 bits in Class C mask) is part of the network
ID. If you changed the subnet mask to 255.255.254.0 or
11111111.11111111.11111110.00000000, then
you will see that the network ID is now the first 23 bits,
and the host ID is now 9 bits.
so with this subnet mask, 199.1.1.x and 199.1.2.x are
actually two different subnets.
Dolphus Brown
A+ Network+ MCSA 22
>-----Original Message-----
>> You are 100% correct!, this means 24 bits are used for
>> network id and 8 bits are used for host id. How ever
>> supernetting is different:
>
>Snoopy has the idea right but let's clean up the details
a bit... (supernet
>joke there.)
>
>> Supernetting is used when multiple, consecutive network
>> addresses of the same Class are combined into blocks.
For
>> example, if you need to assign 199.1.1.x and 199.1.2.x
to
>> the same physical network and want to change subnet
mask.
>
>Yes, except the "blocks" must not only be consecutive BUT
ALSO
>start on a "power of 2" and "occupy that power" - -then
you subtract
>that "power" from the mask to combine them:
>
>So with "1-bit" as Snoopy showed, you must start on
a "multiple of
>2" (2^1) so you can combine 199.1.0.0 with 199.1.1.0 OR
you can
>combine 199.1.4.0 with 199.1.5.0 but not staring on
199.1.1.0,
>or anything else that is odd in the third octet.
>
>
>> You would use the subnet mask of 255.255.254.0 and NOT
>> use DHCP superscopes to assign addresses from DHCP
server
>
>If you subtract 2-bits, /22 or 255.255.252.0 then you
have to start
>on a "Multiple of 4" and combine EXACTLY 4 nets, e.g.,
199.1.12.0-
>199.1.15.0 etc.
>
>--
>Herb Martin
>
>
>.
>
| |
| Herb Martin 2003-09-28, 7:23 pm |
| That is the mathematically correct way but it is actually far more
tedious to do, and far more difficult for most people to understand.
--
Herb Martin
"Dolphus Brown" <DBrown48@satx.rr.com.nospam> wrote in message
news:1629201c38603$9230a030$a6
01280a@phx.gbl...
> It is much easier to understand if you convert the IP
> address and subnet mask into binary first (which is how
> the computer sees it). You obtain the Network ID portion
> of an IP address by performing a logical "AND" of the IP
> address and subnet mask. For example, the subnet mask
> 255.255.255.0 in binary is written
> 11111111.11111111.11111111.00000000. So, any bit in the
> IP address that falls where there is a "1" in the subnet
> mask(first 24 bits in Class C mask) is part of the network
> ID. If you changed the subnet mask to 255.255.254.0 or
> 11111111.11111111.11111110.00000000, then
> you will see that the network ID is now the first 23 bits,
> and the host ID is now 9 bits.
> so with this subnet mask, 199.1.1.x and 199.1.2.x are
> actually two different subnets.
> Dolphus Brown
> A+ Network+ MCSA 22
>
> >-----Original Message-----
> >> You are 100% correct!, this means 24 bits are used for
> >> network id and 8 bits are used for host id. How ever
> >> supernetting is different:
> >
> >Snoopy has the idea right but let's clean up the details
> a bit... (supernet
> >joke there.)
> >
> >> Supernetting is used when multiple, consecutive network
> >> addresses of the same Class are combined into blocks.
> For
> >> example, if you need to assign 199.1.1.x and 199.1.2.x
> to
> >> the same physical network and want to change subnet
> mask.
> >
> >Yes, except the "blocks" must not only be consecutive BUT
> ALSO
> >start on a "power of 2" and "occupy that power" - -then
> you subtract
> >that "power" from the mask to combine them:
> >
> >So with "1-bit" as Snoopy showed, you must start on
> a "multiple of
> >2" (2^1) so you can combine 199.1.0.0 with 199.1.1.0 OR
> you can
> >combine 199.1.4.0 with 199.1.5.0 but not staring on
> 199.1.1.0,
> >or anything else that is odd in the third octet.
> >
> >
> >> You would use the subnet mask of 255.255.254.0 and NOT
> >> use DHCP superscopes to assign addresses from DHCP
> server
> >
> >If you subtract 2-bits, /22 or 255.255.252.0 then you
> have to start
> >on a "Multiple of 4" and combine EXACTLY 4 nets, e.g.,
> 199.1.12.0-
> >199.1.15.0 etc.
> >
> >--
> >Herb Martin
> >
> >
> >.
> >
| |
| Dolphus Brown 2003-09-28, 9:23 pm |
| Aye, it is more tedious to do it like that, but an easier
way to learn it, methinks. Understanding how something
works is generally better than learning rules to follow.
If someone asked me how a subnet mask works, I wouldn't
even know how to explain it without going into binary.
Dolphus Brown
(and I would like to think that an MCSE candidate wouldn't
qualify as "most people" (8oP))
>-----Original Message-----
>That is the mathematically correct way but it is actually
far more
>tedious to do, and far more difficult for most people to
understand.
>
>--
>Herb Martin
>"Dolphus Brown" <DBrown48@satx.rr.com.nospam> wrote in
message
> news:1629201c38603$9230a030$a6
01280a@phx.gbl...
>> It is much easier to understand if you convert the IP
>> address and subnet mask into binary first (which is how
>> the computer sees it). You obtain the Network ID
portion
>> of an IP address by performing a logical "AND" of the IP
>> address and subnet mask. For example, the subnet mask
>> 255.255.255.0 in binary is written
>> 11111111.11111111.11111111.00000000. So, any bit in the
>> IP address that falls where there is a "1" in the subnet
>> mask(first 24 bits in Class C mask) is part of the
network
>> ID. If you changed the subnet mask to 255.255.254.0 or
>> 11111111.11111111.11111110.00000000, then
>> you will see that the network ID is now the first 23
bits,
>> and the host ID is now 9 bits.
>> so with this subnet mask, 199.1.1.x and 199.1.2.x are
>> actually two different subnets.
>> Dolphus Brown
>> A+ Network+ MCSA 22
>>
>> >-----Original Message-----
>> >> You are 100% correct!, this means 24 bits are used
for[c
olor=darkred]
>> >> network id and 8 bits are used for host id. How ever
>> >> supernetting is different:
>> >
>> >Snoopy has the idea right but let's clean up the[/color]
details
>> a bit... (supernet
>> >joke there.)
>> >
>> >> Supernetting is used when multiple, consecutive
network
>> >> addresses of the same Class are combined into blocks.
>> For
>> >> example, if you need to assign 199.1.1.x and
199.1.2.x
>> to
>> >> the same physical network and want to change subnet
>> mask.
>> >
>> >Yes, except the "blocks" must not only be consecutive
BUT
>> ALSO
>> >start on a "power of 2" and "occupy that power" - -then
>> you subtract
>> >that "power" from the mask to combine them:
>> >
>> >So with "1-bit" as Snoopy showed, you must start on
>> a "multiple of
>> >2" (2^1) so you can combine 199.1.0.0 with 199.1.1.0 OR
>> you can
>> >combine 199.1.4.0 with 199.1.5.0 but not staring on
>> 199.1.1.0,
>> >or anything else that is odd in the third octet.
>> >
>> >
>> >> You would use the subnet mask of 255.255.254.0 and
NOT[c
olor=darkred]
>> >> use DHCP superscopes to assign addresses from DHCP
>> server
>> >
>> >If you subtract 2-bits, /22 or 255.255.252.0 then you
>> have to start
>> >on a "Multiple of 4" and combine EXACTLY 4 nets, e.g.,
>> 199.1.12.0-
>> >199.1.15.0 etc.
>> >
>> >--
>> >Herb Martin
>> >
>> >
>> >.
>> >
>
>
>.
>[/color]
| |
| Herb Martin 2003-09-29, 2:23 pm |
| > Aye, it is more tedious to do it like that, but an easier
> way to learn it, methinks. Understanding how something
> works is generally better than learning rules to follow.
> If someone asked me how a subnet mask works, I wouldn't
> even know how to explain it without going into binary.
> Dolphus Brown
> (and I would like to think that an MCSE candidate wouldn't
> qualify as "most people" (8oP))
I meant most MCSE candidates or network engineers actually.
Usually it takes me about an hour to explain (better) what normally
takes a day to a day and a half in binary. I teach the students to
SEE how the binary works but directly from decimal (since they
are written that way.)
It requires noticicing the following "high-bit sums":
128 (1bit), 192(2-bits), 224 (3-bits), 240(4bits), 248, 252, 254, 255
--
Herb Martin
| |
| Dolphus Brown 2003-09-30, 4:23 am |
| No doubt that they should be very familiar with those
numbers, but in order to understand what a "high-bit sum"
is, you would have to know binary.
128......192......224......240......248......252........
10000000,11000000,11100000,111
10000,11111000,11111100...
eh, nevermind, scratch that, people seeking a cert don't
really need to know HOW something works, they just need to
know WHAT something does. I, however, would have gone
nuts trying to understand how IP addresses and routing
worked if the binary wasn't explained. I'm 22, I've never
(for the most part) worked in the field, and you,
apparently, teach computer courses. I would have to
concede that teaching binary to a group of students could
be a pain in the XXX (very large). On a side note..did you
know that you can ping an IP address with two, one or no
seperating dots.
For example, www.yahoo.com is 66.218.71.94 or,
66.218.18270, or 66.14305118, OR 1121601374; they are all
the same IP address, but you would need to understand it's
binary form to convert it (you must take the dots off from
right to left, of course). Now, I understand that this is
completely impractical, but I found it very interesting
when I discovered it and just thought I'd share.
Dolphus Brown
SATX MCSA
>It requires noticicing the following "high-bit sums":
>128 (1bit), 192(2-bits), 224 (3-bits), 240(4bits), 248,
252, 254, 255
>
>--
>Herb Martin
>
>
>.
>
| |
| Herb Martin 2003-09-30, 3:23 pm |
| > 128......192......224......240......248......252........
> 10000000,11000000,11100000,111
10000,11111000,11111100...
> eh, nevermind, scratch that, people seeking a cert don't
> really need to know HOW something works, they just need to
> know WHAT something does. I, however, would have gone
> nuts trying to understand how IP addresses and routing
> worked if the binary wasn't explained. I'm 22, I've never
> (for the most part) worked in the field, and you,
> apparently, teach computer courses. I would have to
Not if I explained it to you the "right way" -- you would actually know
how IP works, how the numbers relate, and how to troubleshoot IP
by sight using the decimal.
What I did was model what experts do AFTER they learn the binary
and use it for years.
Watch an IP expert solve problems and you will notice that they do
so almost instantly; ask for an explaination and you will almost invariably
get a long, tedious binary explanation.
It quickly becomes obvious they are not "doing it" that way.
Neat thing is that the method I created works for ALL problems, not just
one type and after you learn subnetting my way, you can learn supernetting
in about 5-10 minutes.
Twenty stations on a (sub)network? Look at them and immediately see
which is "out of place" or "misconfigured."
Then there are the (stupid) MCSE questions where someone gives you
a dozen or so semi-random addresses and says, "pick a subnet mask" --
stupid question because NO ONE does it that way; we pick the mask
first, then the addresses.
But that type of problem is nearly impossible the "traditional" way (at
least
under any sort of time pressure.) Actually those that do solve it using
only
the traditional way will really be using a crude form of my method but
largely guessing instead of having a procedure.
--
Herb Martin
|
|
|
|
|