To Quote RFC1878
...... using calculations which exclude all-zeros and all-ones subnets.
Many vendors support subnetting based upon this premise. This practise
is obsolete! Modern software will be able to utilize all definable
networks.
i.e. They will be able to use the first and last subnets.
So, as long as you are using ComOS 3.5 or above the PM3 will support
this. Most implementations of IPV4 should support this if they were made
after December 1995 (when this RFC was written) This (amazingly)
includes almost all Microsoft implementations, and quite a few unixes,
but it (strangely enough) excludes the TCP/IP included in the Novell
Client32 for dos/win before version 2.1 and tcpip.nlm that shipped with
netware prior to version 4.11.
Cheers
Phil
btw I think you mean x.x.x.1-62,65-126,129-190 and 193-254 these are the
only USEABLE address ranges, the one above and below being for broadcast
and subnet address.)
> ----------
> From: Chad Scott[SMTP:chad@txdirect.net]
> Sent: Tuesday, April 15, 1997 8:54 PM
> To: Robert Hayne
> Cc: portmaster-users@livingston.com
> Subject: Re: IP filters/subnetting.
>
> On Mon, 14 Apr 1997, Robert Hayne wrote:
>
> > 1.2.3.1 to 1.2.3.63 1st subnet
> >
> > 1.2.3.64 to 1.2.3.127 2nd subnet
> >
> > 1.2.3.128 to 1.2.3.191 3rd subnet
> >
> > 1.2.3.192 to 1.2.3.255 4th subnet
>
> Depending on who you ask, this arrangement is invalid. The first and
> last
> subnets are unusable. I'm not sure whether the PM2E supports this
> arrangement or not (MegaZone?).
>
> > What I am not sure of is the isolation of each subnet from one
> another. This
> > is being done on a PM2E.
>
> The subnets should be completely isolated (as much as is possible,
> anyway)
> as long as the PM2E and the router routing to the PM2E are configured
> correctly.
>
>