Tony Crouch wrote:
Hi All,
Seeing as though this query is not 100% Fedora related it may be better
if someone is able to help me with this query if they reply to me
personally off-list.
On reading the post made recently by Thom, it got me thinking ...
At the school I work out we have a setup where we have 2 computer
laboratories (primary-lab & secondary-lab -- each lab contains about 40
machines) and a rather large number of computers scattered throughout
classrooms.
Everything runs from a DHCP router which supplies each machine with its
required network details.
The machines scattered throughout the school (non-lab machines) has the
following details:
IP-Address: 10.10.?.* -- ? starts at 2
subnet-mask: 255.255.0.0
The machines in the computer labs have the following details
IP-Address: 10.10.2.*
subnet-mask: 255.255.255.0
The discretion of whether a machine receives the B or C-class subnet
mask is based on MAC addresses.
What has confused in the fact these two networks can talk to one
another. Is it because the admin has freed up the firewall between the
two, but I can't really see the point of specifying two seperate masks
and then opening up the firewall to additional traffic.
Was wondering if someone might be able to shed some light into either
why this happens (or shouldn't happen :P ) in the world of networking.
It shouldn't happen and its kind of accidental that it works with some
equipment. The 10.10.2.x machines will bounce things outside their
netmask range through a router. That part is OK and the router may
apply some firewall restrictions. The problem comes when a machine in
the larger range tries to send or respond to the ones with the smaller
netmaks. They'll see it as part of their own subnet and arp for the
address instead of sending to the router - and they should send the arp
with a broadcast address of 10.10.255.255 while the 10.10.2.x boxes
should only respond to a broadcast address of 10.10.2.255.
--
Les Mikesell
lesmikesell@xxxxxxxxx
