I have a question about NATing I do not fully understand.
Normally when I think of NATing an IP Address, I think in terms of the outside interface and public address subnet. If I have a server on the inside network and an available public address in my subnet, I can NAT that server to the available address. The users accessing the NATed address are routed to the outside interface and then hit the NATed address which is translated to an inside or DMZ address.
So it is as if a device not on the outside interface of the firewall is routed to hit the interface, which has a NATed address from the public subnet on it
Suppose I have an inside interface:
a DMZ interface
with a server
Now, If I have a host that I want to allow access to the DMZ server, and I NAT the inside host to the DMZ:
This is a quirk of Cisco firewalls and NAT. On most other firewalls you only need to NAT when you are actually changing the IP address whereas with Cisco (unless you turn off NAT) even if you want to present the same inside address to the outside you still need a NAT statement for it. I haven't seen this behaviour on any other firewalls but i don't claim to have worked on all different types of firewalls.
The DMZ server does not see 10.10.10.10 on the same interface. The DMZ server merely works out that 10.10.10.10 is on a different subnet than it is on and therefore sends it to it's default-gateway which should be the DMZ interface on the Pix. The pix then knows where to send it. But the server has no knowledge of this.
Table of ContentsIntroductionVersion HistoryPossible Future
UpdatesDocuments PurposeNAT Operation in ASA 8.3+ SectionsRule Types
Network Object NATTwice NAT / Manual NATRule Types used per SectionNAT
Types used with Twice NAT / Manual NAT and Network Obje...
Table of Contents Introduction:This document describes details on how
NAT-T works. Background: ESP encrypts all critical information,
encapsulating the entire inner TCP/UDP datagram within an ESP header.
ESP is an IP protocol in the same sense that TCP an...