I have recently encountered an ultra wide band radio bridge for
ethernet packets that presents problems for etharp. The radio bridge
pair modifies the source ethernet address of the packets that it
bridges. This seems to present problems for etharp_ip_input, which
calls update_arp_entry with incoming source IP address and source
ethernet address pairs. Even if a correct ethernet address has
previously been obtained from an ARP reply, this update_arp_entry call
will overwrite it with the ethernet address that has been modified by
the bridges.
The radio bridges seem to be a disservice by altering the source
ethernet address as they do. Nevertheless, they are able to work with
Windows workstations, which apparently do not modify their arp entries
based on the incoming IP traffic.
Who's right here? Is it legitimate to glean ethernet source addresses
from incoming IP packets, or should one rely exclusively on the
information presented in ARP replies?
I haven't been able to find any reference to the gleaning method in
RFCs or my elderly copy of Stevens. If gleaning is wrong, the solution
is would seem to be as simple as deleting one's calls to
etharp_ip_input.
--
Jim Gibbons
|
address@hidden
|
Gibbons and Associates, Inc.
|
TEL: (408) 984-1441
|
900 Lafayette, Suite 704, Santa Clara, CA
|
FAX: (408) 247-6395
|
|