CCNP ROUTE: 5. EIGRP Metric tuning: Offset Lists

-Offset Lists allow adding a value (an offset) to the calculated metric (FD and AD) for a destination network.
-The syntax for creating an offset list is: offset-list {access-list-number | access-list-name} {in | out} offset [interface-type interface-number]
-Offset Lists match prefixes/prefix lengths using an IP ACL, so that the offset is applied only to routes matched by the ACL with a permit clause.

Consider the same destination network as in previous examples, with its R6's two possible routes to it:
R6#show ip eigrp topology 
...
P 172.16.0.12/30, 1 successors, FD is 2195456
        via 172.16.0.6 (2195456/2169856), FastEthernet0/1
        via 172.16.0.2 (2681856/2169856), Serial0/0

Now, let's say we want traffic from R6 to 172.16.0.12/30 to be routed through 172.16.0.2 (R4), instead of 172.16.0.6 (R5), without manually changing any metrics on R6.
For this, we should configure on R6 an offset list which adds a value to the metric of the 172.16.0.12/30 route received from R5, so that R5's route metric will be larger than R6's route metric for the same destination.
This way, R4 will be preferred as the next-hop for 172.16.0.12/30 network.
The offset list will be configured inbound on the port facing R5.
First, we should configure an ACL which matches the destination network with a permit clause.
To match the exact prefix and prefix length, use an extended ACL. When doing so, use the destination address field to match the prefix length.

R6(config)#access-list 100 permit ip host 172.16.0.12 host 0.0.0.3 
R6#show access-lists 
Extended IP access list 100
    10 permit ip host 172.16.0.12 host 0.0.0.3

Then, configure the offset list on FastEthernet0/1 and add a value so that R5's route metric will be larger than R6's route metric for the same destination:

2681856 - 2195456 = 486400 (so let's add a rounded 500000 to the metric through R5)

R6(config)#router eigrp 1
R6(config-router)#offset-list 100 in 500000 fastEthernet 0/1
*Mar  1 00:39:57.163: %DUAL-5-NBRCHANGE: IP-EIGRP(0) 1: Neighbor 172.16.0.6 (FastEthernet0/1) is resync: route configuration changed

R6#show ip eigrp topology
...
P 172.16.0.12/30, 1 successors, FD is 2195456
        via 172.16.0.2 (2681856/2169856), Serial0/0
R6#show ip route
...
D       172.16.0.12/30 [90/2681856] via 172.16.0.2, 00:04:05, Serial0/0

Now we see that R4 is indeed preferred to reach 172.16.0.12/30.
Also, we notice that no feasible successor exists anymore for this destination.
The reason is that when the offset list added the value of 500000 to the FD, it also added it to the AD, making the AD larger than the successor's FD, thus not satisfying the feasibility condition.
The former successor is still visible, confirming that the offset list added the configured value to both FD and AD:

R6#show ip eigrp topology all-links 
...
P 172.16.0.12/30, 1 successors, FD is 2195456, serno 11
        via 172.16.0.2 (2681856/2169856), Serial0/0
        via 172.16.0.6 (2695456/2669856), FastEthernet0/1

Extracting 500000 from the current metric values (2695456/2669856) of the second route results in the initial values of this route (2195456/2169856).

Now, R6 will choose the path to R4 on Serial0/0 to reach 172.16.0.12/30.