(Thanks for uncloaking your identity, by the way. This discussion
just got infinately more interesting. :))

"Schouten, Diederik (Diederik)" wrote:
> 
> > And this brings me to another point: fingerprinting leaks.
> > In some cases, knowing the MAC address tells you a lot about
> > a box. (Thinking of non-PC boxes with built-in NICs.)
> 
> Sure, can't argue with that.
> But all this is based on the assumption that the attacker or trojan is
> already within your "secured" network.

Hm, no. Just on the other side of a bridging firewall. In the case
 internet --- router --- empty stub network -- firewall -- internal
this is a non-issue, but as soon as you add DMZs, or have other hosts
right outside your firewall, or have a router than can be SNMP-
polled by anyone (view the ARP cache), it becomes an issue.

I'll give you that "open for an additional layer of information 
gathering" doesn't necessarily mean "open for an additional
avenue of attack", but.. Yeah, I'm paranoid. :)

> [on using ARP to see if several IPs resolve to the same machine]
> Indeed, and there are many more ways to find out of several 
> machines resolve to the same machine.

Hmm.. Which ones would that be?
(Sans allowing telnet in to the machine and seeing that you 
get the same hostname banner.)

This is about as easy as it gets, IMHO, since a bridging firewall
can't make L3/4 decisions on whether an external box gets to
learn about internal network topology (hey, there's another 
argument! I can see if internal IPs get routed somewhere 
else! :)) -- a bridging firewall needs to pass ARPs to the inside 
network even when the external box isn't allowed to initiate a 
single connection to the inside hosts.

(Yes, I agree that for 90% of all networks, this probably doesn't
mean squat, but for the rest -- the _large_ ones -- this 


> > And, here, have another low blow while I'm at it: proxy ARP
> > does indeed answer using the firewall's MAC address for all
> > published boxes. Even if they're down, or temporarily out and
> > traveling, or has a physical L1 switch moving it back and forth
> > between separate physical networks once every few minutes.
> > (Yes, these things do exist :))
> 
> Ehm... you're attacking your own standpoint now? ;)

Hm, no, I don't think so. "Always answering" versus "seeing of
the host is not there" is a good thing, I think. Unless you
assume that the firewall itself can (easily) be attacked. 
I don't. :)


-- 
Mikael Olsson, Clavister AB
Storgatan 12, Box 393, SE-891 28 �RNSK�LDSVIK, Sweden
Phone: +46 (0)660 29 92 00   Mobile: +46 (0)70 26 222 05
Fax: +46 (0)660 122 50       WWW: http://www.clavister.com

"Senex semper diu dormit"
_______________________________________________
Firewalls mailing list
[EMAIL PROTECTED]
http://lists.gnac.net/mailman/listinfo/firewalls

Reply via email to