There is probably a fundamental gap in my understanding of how NAT works,
otherwise I wouldn't need to ask this question...

I have a client with a NW5.0 SB server, probably running BM3.0, if that even
makes a difference. A LinkSys DSL router allows them to get out to the
internet. I want to configure a filter exception that accepts SMTP mail
ONLY from their ISP, since they filter it for SPAM and viruses. Some junk
mail has been addressed directly to their mail server's IP address instead
of their domain name, bypassing that filtering.

So, my question is, when the server receives mail through the LinkSys NAT
device, what IP address does it appear to come from? Is it coming from the
address of the sending server, from the public IP address of the LinkSys
router, or the private address of the LinkSys router?

Surely, since so many people are using these "consumer" level routers to act
as PPPoE clients for Netware networks, someone elxse out there must have run
into this. As I mentioned in my opening statement, is this merely a gap in
my understanding of the fundamentals of NAT? I am unable to simulate the
environment here at home (lack of equipment), but figure I should be able to
learn what I need to know using Windows and some IP tools, but have so far
been unable to figure out how. I have the two Craig Johnson books, and
figure what I want is in there (everything else is), but I am unable to nail
it down...

Can someone suggest how I might restrict incoming SMTP traffic to only one
source IP address, without buying more upscale edge-gear?


Rob Pelletier