Most EIGRP adjacency issues are easily resolved - mistyped IP address, bad wildcard mask in the neighbor statement, mistyped EIGRP AS number, and so forth. You know the AS number has to match, but always double-check that. Next, check the wildcard mask to make sure the correct interfaces are EIGRP-enabled. In the following example, we've configured three loopbacks on R2 and want to advertise all three loopbacks via EIGRP, but R4 only sees one.
R2(config)#int loopback1
R2(config-if)#ip address 10.1.1.1 255.255.255.0
R2(config-if)#int loopback2
R2(config-if)#ip address 10.1.2.1 255.255.255.0
R2(config-if)#int loopback3
R2(config-if)#ip address 10.1.3.1 255.255.255.0
R2(config-if)#router eigrp 100
R2(config-router)#no auto
R2(config-router)#network 10.1.1.0 0.0.0.255
R2(config-router)#network 172.12.23.0 0.0.0.255
R4#show ip route eigrp
10.0.0.0/24 is subnetted, 1 subnets
D 10.1.1.0 [90/409600] via 172.12.23.2, 00:00:25, Ethernet0
We know it's not an adjacency issue, because R4 has to have an adjacency to get the route. Perhaps the config on R2 is the problem.
R2(config-if)#router eigrp 100
R2(config-router)#no auto
R2(config-router)#network 10.1.1.0 0.0.0.255
R2(config-router)#network 172.12.23.0 0.0.0.255
The only one of the three networks that matches that network statement is 10.1.1.0 /24. We could use 10.1.0.0 0.0.255.255 to successfully advertise the loopbacks, but by breaking the three networks down into binary, we can see that the three networks have the first 22 bits in common. Therefore, a better wildcard mask to use is 0.0.3.255.
R2(config-router)#no network 10.1.1.0 0.0.0.255
R2(config-router)#network 10.1.0.0 0.0.3.255
R4#show ip route eigrp
10.0.0.0/24 is subnetted, 3 subnets
D 10.1.3.0 [90/409600] via 172.12.23.2, 00:00:36, Ethernet0
D 10.1.2.0 [90/409600] via 172.12.23.2, 00:00:36, Ethernet0
D 10.1.1.0 [90/409600] via 172.12.23.2, 00:00:36, Ethernet0
Another factor that must match between potential EIGRP neighbors is the K-metrics, or metric weights. I've brought R2's Ethernet0 interface back up, and I'll now change the metric weights on R2 to see what happens.
R2(config)#router eigrp 100
R2(config-router)#metric ?
holddown Enable IGRP holddown
maximum-hops Advertise IGRP routes greater than <hops> as unreachable
weights Modify IGRP metric coefficients
R2(config-router)#metric weights ?
<0-8> Type Of Service (Only TOS 0 supported)
R2(config-router)#metric weights 0 ?
<0-4294967295> K1
R2(config-router)#metric weights 0 1 ?
<0-4294967295> K2
R2(config-router)#metric weights 0 1 2 ?
<0-4294967295> K3
R2(config-router)#metric weights 0 1 2 3 ?
<0-4294967295> K4
R2(config-router)#metric weights 0 1 2 3 4 ?
<0-4294967295> K5
R2(config-router)#metric weights 0 1 2 3 4 5 ?
<cr>
R2(config-router)#metric weights 0 1 2 3 4 5
R2(config-router)#
03:33:18: %DUAL-5-NBRCHANGE: IP-EIGRP 100: Neighbor 172.12.23.4
(Ethernet0) is d
own: metric changed
03:33:18: %DUAL-5-NBRCHANGE: IP-EIGRP 100: Neighbor 172.12.23.3 (Ethernet0)
is d
own: metric changed
R2(config-router)#
03:33:20: %DUAL-5-NBRCHANGE: IP-EIGRP 100: Neighbor 172.12.23.4
(Ethernet0) is d
own: K-value mismatch
03:33:20: %DUAL-5-NBRCHANGE: IP-EIGRP 100: Neighbor 172.12.23.3 (Ethernet0)
is d
own: K-value mismatch
R2(config-router)#
03:33:24: %DUAL-5-NBRCHANGE: IP-EIGRP 100: Neighbor 172.12.23.4 (Ethernet0)
is d
own: K-value mismatch
03:33:25: %DUAL-5-NBRCHANGE: IP-EIGRP 100: Neighbor 172.12.23.3 (Ethernet0)
is d
own: K-value mismatch
The adjacencies go down, and the router's kind enough to tell you that it's the result of the metrics changing. The router's also going to keep telling you that the neighbors are down because of the K-value mismatch.
We'll also see that message on R3 and R4, but just in case you don't see it, debug eigrp packet will indicate the K-value mismatch as well.
R4#debug eigrp packet
EIGRP Packets debugging is on
(UPDATE, REQUEST, QUERY, REPLY, HELLO, IPXSAP, PROBE, ACK)
3w2d: EIGRP: Received HELLO on Ethernet0 nbr 172.12.23.2
3w2d: AS 100, Flags 0x0, Seq 0/0 idbQ 0/0
3w2d: K-value mismatch
3w2d: %DUAL-5-NBRCHANGE: IP-EIGRP 100: Neighbor 172.12.23.2 (Ethernet0)
is down:
K-value mismatch
The changed K-values will also appear in the running configuration.
Whether it's on your CCNP certification exams or in a production network, you really shouldn't change the EIGRP K-values unless you have a very powerful reason to do so - and an understanding of the consequences!
To your success,
Chris Bryant
CCIE #12933