192.168.0.* vs 192.168.1.*
David
as the 3rd part of an IP address as opposed to 1-255?
It's just supposed to use the netmask to determine what part of the IP
address is actually the host-part wildcard right?
Florian E.J. Fruth
wrote in <394d1bce...@nntp.ix.netcom.com>
at the moment no programm should have problems with the 0 (also microsoft
did sth. against this !). but if u don't need the 65536 ips in your net u
can use the 1...
fejf
Brian
David wrote in message <394d1bce...@nntp.ix.netcom.com>...
>Are there any complications or things that can't be done by using a 0
>as the 3rd part of an IP address as opposed to 1-255?
>It's just supposed to use the netmask to determine what part of the IP
>address is actually the host-part wildcard right?
Only the in last 8 bit address space should you avoid using 0 or 255 - in
other words everything between;
192.168.0.1 to 192.168.255.254
is fine.
192.168.x.0 and 192.168.x.255
are reserved.
Best regards,
Brian
DanH
(David) wrote:
> Are there any complications or things that can't be done by using a 0 as
> the 3rd part of an IP address as opposed to 1-255?
>
> It's just supposed to use the netmask to determine what part of the IP
> address is actually the host-part wildcard right?
The last number is the 'network number' I've used x.x.0.x x.x.1.x and just
about every other series there.
Doesn't really make a difference, just make sure they're all on the same
sub-net.
DanH
--
UNIX - Not just for vestal virgins anymore
Linux - Choice of a GNU generation
Francois Labreque
Brian wrote:
>
> Hi David:
>
> David wrote in message <394d1bce...@nntp.ix.netcom.com>...
> >Are there any complications or things that can't be done by using a 0
> >as the 3rd part of an IP address as opposed to 1-255?
>
> >It's just supposed to use the netmask to determine what part of the IP
> >address is actually the host-part wildcard right?
>
> Only the in last 8 bit address space should you avoid using 0 or 255 - in
> other words everything between;
That's not necessarily true. Back in the forlorn days before everyone
used Cisco routers with IOS version 12*, the first and last subnets of a
network were reserved. In other words, if you used 255.255.255.0 as you
mask, 192.168.0.x and 192.168.255.x were not useable.
Depending on the routers used at your location and the routing protocol
they use, your mileage may vary. As others have pointed out, since
192.168.x.x is a class B address** it gives you 65535 possible hosts.
Unless you really have that many devices, there's really no harm in
using 192.168.1.x, just to be safe.
Note:
* For those interested in such a debate. Cisco still does not allow
using the first subnet of a network unless you configure it with "ip
subnet-zero" which is the new default on IOS 12.x.
** In this day and age where *almost* everyone uses "Classless
Inter-Domain Routing" - CIDR for short, Class A, B or C addresses don't
matter much, but there are still places using RIP v1 out there.
--
Francois Labreque | In the future, performance will be measured
flabreq | by the size of your pipe.
@ | - Dogbert, on networking
attglobal.net
Paul Colquhoun
|
|
|Brian wrote:
|>
|> Hi David:
|>
|> David wrote in message <394d1bce...@nntp.ix.netcom.com>...
|> >Are there any complications or things that can't be done by using a 0
|> >as the 3rd part of an IP address as opposed to 1-255?
|>
|> >It's just supposed to use the netmask to determine what part of the IP
|> >address is actually the host-part wildcard right?
|>
|> Only the in last 8 bit address space should you avoid using 0 or 255 - in
|> other words everything between;
|
|That's not necessarily true. Back in the forlorn days before everyone
|used Cisco routers with IOS version 12*, the first and last subnets of a
|network were reserved. In other words, if you used 255.255.255.0 as you
|mask, 192.168.0.x and 192.168.255.x were not useable.
|
|Depending on the routers used at your location and the routing protocol
|they use, your mileage may vary. As others have pointed out, since
|192.168.x.x is a class B address** it gives you 65535 possible hosts.
Actually, the 192.168.x.y reserved address range is 256 contiguous "Class C"
addresses, and 192.168.0.x and 192.168.255.x are quite safe to use.
192 decimal is 11000000 binary. Class A addresses started 0xxxxxxx
Class B addresses started 10xxxxxx
Class C addresses started 11xxxxxx
If you want to be really picky Class C addresses started 110xxxxx
Class D addresses started 1110xxxx
Class E addresses started 1111xxxx
Class D & E as address classes were never in widespread use, but
were used in experimental areas.
|Unless you really have that many devices, there's really no harm in
|using 192.168.1.x, just to be safe.
|
|Note:
|* For those interested in such a debate. Cisco still does not allow
|using the first subnet of a network unless you configure it with "ip
|subnet-zero" which is the new default on IOS 12.x.
|** In this day and age where *almost* everyone uses "Classless
|Inter-Domain Routing" - CIDR for short, Class A, B or C addresses don't
|matter much, but there are still places using RIP v1 out there.
Agreed. The "Don't use 1st or last subnet" only really applied when
you were dividing up a Class-stype address with a local mask.
No there is no such thing as a "subnet", just variable-sized
netmasks defining different sized networks. Very semantic, but
make things much more generalized and easier to understand, IMHO.
--
Reverend Paul Colquhoun, postm...@andor.dropbear.id.au
Universal Life Church http://andor.dropbear.id.au/~paulcol
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
xenaphobia: The fear of being beaten to a pulp by
a leather-clad, New Zealand woman.
Francois Labreque
Paul Colquhoun wrote:
>
> On Sun, 18 Jun 2000 22:52:37 -0400, Francois Labreque <fla...@attglobal.net> wrote:
> As others have pointed out, since
> |192.168.x.x is a class B address** it gives you 65535 possible hosts.
>
> Actually, the 192.168.x.y reserved address range is 256 contiguous "Class C"
> addresses, and 192.168.0.x and 192.168.255.x are quite safe to use.
I know, I know.
[Note to self: Don't post this late at night!]
>
> Class D & E as address classes were never in widespread use, but
> were used in experimental areas.
"Class D" is used for multicasting and IS very widespread. It is used
for such things as building OSPF adjacencies, and routing table
updates, amongst other things. Class E is still labeled as
"experimental, however.
Sean
you want after the 192.168.y.z. If your private ip reaches the
internet the router will know to dump it after reading the
192.168.x.z. Have fun
Sean
On Sun, 18 Jun 2000 19:00:06 GMT, Davi...@ix.netcom.com (David)
wrote:
Joe Chiasson
192.168.0.1, which if placed on an existing IP subnet of 192.168.0 could cause
some grief until you reconfigure it
moonie
>Are there any complications or things that can't be done by using a 0
>as the 3rd part of an IP address as opposed to 1-255?
>
>It's just supposed to use the netmask to determine what part of the IP
>address is actually the host-part wildcard right?
Not that I know of, I use 192.168.0.2 for my linux box (Win98SE ICS connected
to I-net, don't ask) and I have SAMBA, APACHE, FTP, and a Half-Life Dedicated
Server, as well as all of my normal net services running just fine (netmask:
255.255.255.0)
moonie ;)
Sean
It is fine to use 192.168.0.z or 192.168.1.z. If you use the
192.168.0.z make sure all your computers have 192.168.0.z and that
your subnetmasks are the same. Look belowe for an example...
192.168.0.1 = router/gateway
192.168.0.2 = first machine
192.168.0.3 = second machine and so forth.
just make sure your gateway is set to 192.168.0.1 on your machines. As
to the wildcard, are you talking about wild card masking for
firewalls? All the subnetmask does it determin what is on what network
and what is not, and which host is on what network, thats it. 1s and
0s..pain the ass! LoL. If you dont supply a netmask then it should
take the default class c mask. If you have any questions email me.
Sean
On Fri, 30 Jun 2000 14:15:16 -0400, moonie;) <who...@mybox.mydomain>
wrote:
John Mazza
address. The rule with using "0" or "255" in TCP/IP is that you can't have
either a host address or network number resolve to all "0" or all "1" in
binary. I'll explain in a bit more detail below:
An IP address is a 32-bit unsigned long integer. Since human beings do not
like to think in binary, we express this 32-bit number as four digital (base
10) numbers separated by dots. (ie: 192.168.0.1). This is strictly for
convenience and ease of use. BTW - 192.168.0.1 is C0A80001 in Hexadecimal,
or 11000000 10101000 00000000 00000001 in binary (spaces between bytes of
address for clarity).
The IP address is only part of the picture. It must be combined with a
subnet mask before it has any relevance. The subnet mask is a number that
is boolean AND-ed with the IP address to tell what portion of the IP address
is the network number and what portion is the node address. The network
number identifies the unique network, and the node address indentifies which
machine in that network we are attempting to contact. Just like an IP
address, a subnet mask is usually given in dotted decimal notation. For
example, it is standard practice to give a Class C address a subnet mask of
255.255.255.0 (FF.FF.FF.00 in hex, or 11111111 11111111 11111111 0000000 in
binary). When we logically AND 192.168.0.1 with a mask of 255.255.255.0 we
get:
11000000 10101000 00000000 00000001 (192.168.0.1)
11111111 11111111 11111111 00000000 (255.255.255.0)
---------------------------------------------------------------
11000000 10101000 00000000 00000000 (192.168.0.0)
Logically AND-ing 192.168.0.1 with the inverse of the subnet mask gives us
the node number:
11000000 10101000 00000000 00000001 (192.168.0.1)
00000000 00000000 00000000 11111111 (0.0.0.255)
-------------------------------------------------------------
00000000 00000000 00000000 00000001 (0.0.0.1)
Now, we can see from the above that 192.168.0.1 is an okay address, as
neither the network number (192.168.0.0) nor the node number (0.0.0.1) is
all binary 0 or binary 1.
Where we can get into trouble is when we do non-standard subnet masks, such
as 255.255.255.128 and forget to watch out for the unusable node numbers.
For example, let's use an address of 192.168.1.127 and a mask of
255.255.255.128. Converting these to binary shows that our network number
is:
|----------- Net No -------------| | Node |
11000000 10101000 00000001 0 1111111 (192.168.1.127)
11111111 11111111 11111111 1 0000000 (255.255.255.128)
------------------------------------------------------------------
11000000 10101000 00000001 0 0000000 (192.168.1.0)
And our node number is:
10000000 10101000 00000001 0 1111111 (192.168.1.127)
00000000 00000000 00000000 0 1111111 (0.0.0.127)
---------------------------------------------------------------
00000000 00000000 00000000 0 1111111 (0.0.0.127)
UH OH! We have a node number that is all "1's". We also can't use
192.168.1.128, as it is all "0" in the subnet of 192.168.1.128.
Oh - above I mentioned Class C addresses. Address classes are an old
Internet convention and are determined by the first octet of the address:
1-126 Class A Mask 255.0.0.0 (Originally intended for huge networks)
127 Loopback
128-191 Class B Mask 255.255.0.0 (For medium sized networks)
192-223 Class C Mask 255.255.255.0 (Smaller networks).
There are only 125 Class A networks possible, with millions of nodes each.
There are 65000 or so Class B networks, with roughly 65,000 nodes each, and
millions of Class C networks, with million of nodes each.
As IP network numbers become a more and more scarce resource, subnetting
into the third octet and using technologies such as NAT to allow an entire
network to access the global Internet through a single IP address will
continue to gain popularity. Once IPv6 becomes widespread, this will cease
to be an issue, as its 128-bit addresses will provide plenty of address
space for the forseeable future.
I hope this primer helps in understanding how to plan an IP addressing
scheme. Please let me know if I can be of any further assistance in this
subject.
moonie;) <who...@mybox.mydomain> wrote in message
news:962389089...@nntp.northstate.net...
C.J.
was not permitted. As such, some older devices (some routers, etc) may not
understand it correctly. Having said that, almost all newer devices and hosts
have absolutely no problem with the zero subnet. Just make sure you don't
have any really old stuff on your network making routing decisions.
Even newer devices may need to be configured to support it properly (though
almost all do by default.) For example, cisco's IOS has an ip "subnet-zero"
command to explicitly enable/disable this support.
In article <Q0S75.27597$A%3.31...@news1.rdc2.pa.home.com>, "John Mazza"