RE: Where NAT disenfranchises the end-user ...

From: Jared Mauch [mailto:jared@puck.Nether.net]
Sent: Sunday, September 09, 2001 2:49 PM

  Let me reprhase my inital statement, "In most cases i've seen
where someone is using NAT it's part of a security policy and not due
to lack of available address space".

Jared, those whom depend on an accident, for security, deserve what happens
when the accident undoes itself. I was just over on www.netcraft.com,
checking out their stats for the CodeRed worm. I was amazed at how fast IIS
admins responded by applying the patches. If NAT were suddenly "fixed", any
incidental security is toast. NAT was never designed for, and was never
intended as, a security method. Any current protection is strictly the
result of a side-effect. The side-effect that breaks the internet
connection. It's a result of the connection being broken. A properly built
firewall is much more effective and definitely more deterministic. Neither
is it vulnerable to a "fix patch".

Roeland Meyer <rmeyer@mhsc.com> writes:

>> From: Jared Mauch [mailto:jared@puck.Nether.net]
>> Sent: Sunday, September 09, 2001 2:49 PM

>> Let me reprhase my inital statement, "In most cases i've seen
>> where someone is using NAT it's part of a security policy and not due
>> to lack of available address space".

Jared, those whom depend on an accident, for security, deserve what happens
when the accident undoes itself. I was just over on www.netcraft.com,
checking out their stats for the CodeRed worm. I was amazed at how fast IIS
admins responded by applying the patches. If NAT were suddenly "fixed", any
incidental security is toast. NAT was never designed for, and was never
intended as, a security method. Any current protection is strictly the
result of a side-effect. The side-effect that breaks the internet
connection. It's a result of the connection being broken. A properly built
firewall is much more effective and definitely more deterministic. Neither
is it vulnerable to a "fix patch".

I don't understand what kind of "fix patch" you're talking about
here...NAT uses the same techniques that a stateful firewall uses; if
you can find some kind of "fix patch" to bypass NAT, chances are
excellent it will work on a stateful firewally, too.

I've actually seen the question of how NAT breaks the Internet more
than a good stateful firewall come up more than once, and haven't
really seen a satisfactory answer. Where does a stateful firewall
configured to only allow outgoing connections work that NAT doesn't?

I ask not to drag this discussion on, but because I use NAT for
address conservation and security on a couple networks that I operate,
and am curious if I'd be much better off with something different...

-----ScottG.

in the case of IPSec, the IP addresses need to be preserved end-to-end
as part of the whole security scheme.

richard

From: "Scott Gifford" <sgifford@tir.com>

I've actually seen the question of how NAT breaks the Internet more
than a good stateful firewall come up more than once, and haven't
really seen a satisfactory answer. Where does a stateful firewall
configured to only allow outgoing connections work that NAT doesn't?

Anywhere the IP address is a part of the protocol, and a proxy for that
protocol does not exist. Peer election protocols, replication protocols, etc.

True, but ONLY because the Internet Architecture lacks an alternative
namespace that could identify the box associated with a given network
interface. (The IP address is used in this context to identify the network
interface associated with the Security Association). So that's all true
today, but is driven by a shortcoming in the Internet Architecture.

Ran
rja@Inet.org

perhaps, but this doesn't invalidate either his question or my answer
to it. this whole discussion is really pointless due to the fact that
right this minute, reality sucks, and cannot be instantly fixed by a
stupid flame war.

richard

right this minute, reality sucks, and cannot be instantly fixed by a
stupid flame war.

Oh so true, and of far wider applicability than the subject at
hand. If this mailing list had a FAQ, this should be the answer
to most of the questions.

Alex Bligh
Personal Capacity

Q1: What are the common flame wars on NANOG?
A1: RFC1918, ORBS/MAPS, ARIN, PI /28s, DSL for business use.

Q2: What are some proper ways to configure/run a network? For instance,
should I use RFC1910 space to number my routers?
A2: See Q1.