[olsr-dev] Improvements in the algorithm

[Bernd Petrovitsch]

On Wed, 2005-11-16 at 17:19 +1100, Dan wrote:
[...]
> There is a problem though.  If the node with the IP address of 10.10.100.1
> moves to another area, it may no longer be within cabled range of the
> 10.10.100.0/28 network.  So it must change it's IP address.  This is also
> true of wireless networks that are subnetted the same way - with each
> geographical location covering a different range of IP addresses.  If a node
> moves across a geographic subnet border it must change it's IP address.  Any
> current data connections must be closed and re-established after the new
> address is configured.  Also, because it has a new IP address, other nodes
> may now not know where to find it.  Traditional methods of keeping track of
> dynamic IP addresses, such as dynamic DNS or mobile IP are basically too
> slow to keep track of a fast-roaming node, and they require centralised,
> human-managed servers to maintain the identity-to-location lookup tables.

And - especially - the security is the major problem. Think of ssh and
moaning about chnaged IP adresses for a given host key.
So - if you want security - you need also a working PKI as a
prerequisite.

And there is also another thing: At least in Vienna we have lots of
static nodes which basically do not move (since the antenna is mounted
on the top of the roof). For these it makes absolutely sense to have a
network/mask (as opposed to really mobile nodes).

	Bernd

PS: Sorry, I didn't read the rest of the mail up tonow.
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services