Nmap Development mailing list archives

Re: ARP scan on Mac OS, only the gateway MAC is shown


From: David Fifield <david () bamsoftware com>
Date: Sat, 24 Sep 2011 09:15:04 -0700

On Wed, Aug 24, 2011 at 11:36:01AM +0100, Giuliano wrote:
Hi Guys,

  I'm on Mac OS (Lion 10.7.1), trying to get a list of live MAC
addresses on the connected network segment... nmap is being run as
root, across wireless.
When I probe the default gateway, everything looks fine:

# ./nmap-5.51/nmap -e en1 --send-eth -sP -PR 10.0.0.1
Host is up (0.0017s latency).
MAC Address: 00:64:DE:AD:BE:EF (Cisco Systems)

When I try to do the same on another host:

# ./nmap-5.51/nmap -e en1 --send-eth -sP -PR 10.0.0.234
Note: Host seems down. If it is really up, but blocking our ping probes, try -Pn

Turning on verbose/debug reveals that nmap isn't even attempting to do
ARP, as if host .234 was on another ethernet segment. But --iflist
shows:

************************INTERFACES************************
DEV (SHORT) IP/MASK TYPE UP MTU MAC
lo0 (lo0) 127.0.0.1/8 loopback up 16384
en1 (en1) 10.0.0.36/24 ethernet up 1500 xx:xx:xx:xx:xx:xx
**************************ROUTES**************************
[..a bunch of routes..]
127.0.0.0/8 lo0 127.0.0.1
0.0.0.0/0 en1 10.0.0.1

If I change the -PR in -PE, the host reports as UP but I'm still
seeing no ARP going on... A single ICMP packet is sent to the default
gateway, using the gateway's MAC. A response is received, with the
target host's MAC as source.
I tried different nmap versions, with/without the builtin libcap, as
root or as a normal user, etc. Needless to say, on Linux I've got no
issues whatsoever.
How nmap could possibly get confused about what networks are connected?

Perhaps it's one of the routes in [..a bunch of routes..] above? It's
possible for a route to override the address/netmask of an interface.

David Fifield
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/

Current thread: