Bit of a real world scenario that I'm currently working on, but need to
pick brains about why I've had to do what I've done.

Scenario is that I have a router in a DMZ performing IPSEC VPN to a 3rd
party. This is to act as a backup service in the event of my primary
frame-relay service dying a death. My new router sits in a separate DMZ to
my internet so is essentially a router on a stick. A key requirement is to
perform some monitoring over the VPN service and to alert in the event the
remote end server that we need to access goes down. Other issue is that I
need to PAT all traffic behind an IP, as provided by the 3rd party. So far
so good.

My configuration is as follows -

1. I've configured my one interface with NVI capability (ip nat enable)
2. I've configured my IPSEC VPN tunnel
3. I've configured my PAT
4. I've configured my IP SLA to monitor the end point and send a SNMP trap

Now - points 1 - 3 all work very well. My machines/devices in other subnets
can route to the LAN on the other of VPN tunnel and they get nat'd. My
issues comes with the IP SLA. My IP SLA uses a loopback as the source of
the ICMP echos to my remote server, but the traffic is NEVER natd. I
understand that with the traditional NAT method of defining nat
inside/outside NATing of source/destination determines if a router routes
first then nat's or the reverse. I thought that with a NVI this concept
becomes defunct??

To overcome my issue I've introduced a second loopback and configure PBR
and applied the PBR policy to the local policy of my router i.e. NOT
interface policy. So end IP SLA traffic sourced by loopback 1 get's a next
hop of loopback 2, due to PBR, and then it is NATd correctly and sent down
my VPN tunnel.

Can someone explain to me why I need to introduce the extra hop??
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Are you a CCNP or CCIE and looking for a job? Check out 
www.PlatinumPlacement.com

http://onlinestudylist.com/mailman/listinfo/ccie_rs

Reply via email to