From: Edwards, Andrew M (andrew.m.edwards@boeing.com)
Date: Wed Mar 03 2004 - 13:43:03 GMT-3
I experienced a little tweak the other day while working simple static
NAT from solie PS I....
I tried lab32 with eigrp before setting up NAT translations. Worked in
simple configuration had neighbors, etc.
Then, I set the interface that went to the EIGRP domain as ip nat
outside.
Then I set the static NAT mappings from old routable addresses to new
addresses.
To my suprise, it drops the eigrp neighbors... took me a while to notice
it cause I moved on to DLSW, but after not having the outside NAT routes
anymore, I started poking around and noticed one half of the serial link
said I had a neighbor (the nat outside interface) and the other side of
the serial line said there was no neighbor.
Debug ip packet shows i'm not getting any multicast updates from the ip
nat outside interface. Debug ip packet on ip nat outside interface shows
I'm getting multicast updates from other end.... this explains why I'm
half an adjacency. So I did a little more poking around to see the
traffic and do a debug ip nat on the nat outside interface router
(really on a whim).
DING!!! nat encapsulation failed! Hrmmmm... seems that I'm not able to
source the eigrp multicast updates from the IP NAT OUTSIDE source ip
address.
So, I set a static NAT translation from outside source IP to outside
source IP and recover my adjacency.
e.g. ip nat inside source static x.x.x.x x.x.x.x
Is this a normal occurrance? Only asking the group cause I am familiar
with firewall NAT/PAT which doesn't really entail much with routing
protocols. Plus you get nifty tools like tcpdump/windump that help
resolve issues quickly across interfaces.
I just want to know if I attacked that problem correctly or if there is
a better way to do it. Anyone?
andy
_____
Find things fast with the new MSN Toolbar - includes FREE pop-up
blocking! <http://g.msn.com/8HMBENUS/2752??PS=>
This archive was generated by hypermail 2.1.4 : Thu Apr 01 2004 - 08:15:13 GMT-3