[MLB-WIRELESS] IP Addressing

[Simon J Mudd]

jimmy at deefa.com ("James Healy") writes:

> >My understanding is that clients would be allocated an address from the
> >10.10.x.x range that has been allocated to the Access Point you connect to.
> >The routing nodes (ie backbone) would be using 172.16.x.x addresses.
> 
> So this would mean that on the wiki page, anyone with a node that will form
> the backbone (ie. most nodes in the db, apart from the few
> dog/car/wife/kitchen sink nodes :-P ) will choose 1 IP address for it from
> 172.16.x.x, as well as a bunch of addresses from 10.whatever that it is
> allowed to hand out to any node that requests it.


        wc1         wc3
          \         /
           \       /
            \     /wnx
             +------+ n1         n2 +-----+
wc2 ---------| node1|---------------|node2|
             +------+               +-----+
             /     \
            /       \
           /         \
          /           \
         wc4          wc5

wcx = wireless client
wnx = wireless node
n1 / n2 = ip addresses of internode links

The idea is to have ip addresses allocated to clients + wxn in one
range and ip addresses for internode links in another range.

In madrid we're using 10.64.0.0 for clients and 172.16.64.0 for
internode links. (still only a few of each).

The advantage of this is that for filtering traffic and protocols you
can for example make it much easier to avoid people sending you OSPF
messages over the client network.  The internode networks can be 2
point-to-point links with /32 or a /30 with 2 ip addresses at each
end.  Where more than one node can be heard over the "client network
interface" then a larger network could be used.

We've been experimenting in Madrid for a while with various Internet
tunnels as we don't have line of site between the nodes which are more
active, and also with OSPF and md5 authentication (which is causing
some problems), but at least my netstat shows part of a (still small)
network:

[root at phoenix root]# netstat -nr
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
172.16.64.0     0.0.0.0         255.255.255.252 U        40 0          0 tap0
172.16.64.20    0.0.0.0         255.255.255.252 U        40 0          0 tap1
10.64.0.160     172.16.64.22    255.255.255.224 UG       40 0          0 tap1
10.64.0.192     0.0.0.0         255.255.255.224 U        40 0          0 eth1
x.y.z.q         a.b.c.d         255.255.255.192 UG       40 0          0 eth0
a.b.c.0         0.0.0.0         255.255.255.192 U        40 0          0 eth0
127.0.0.0       0.0.0.0         255.0.0.0       U        40 0          0 lo
0.0.0.0         a.b.c.d         0.0.0.0         UG       40 0          0 eth0

One link is missing but we've had 4 nodes linked up with duplicate VPN
links and now that this is working we have to help other nodes to
setup their machines in a similar way.

Perhaps this example makes things clearer?

Simon
--
Simon J Mudd,   Tel: +34-91-408 4878,  Mobile: +34-605-085 219
Madrid, Spain.  email: sjmudd at pobox.com,  Postfix RPM Packager

To unsubscribe: send mail to majordomo at wireless.org.au
with "unsubscribe melbwireless" in the body of the message