Main Troubleshooting Flowchart
Neighbor Check
Flowchart Notes
Issue the show ip eigrp interface command to verify.
Issue the show interface serial command to verify.
Flowchart Notes
Issue the show ip interface command to verify.
Redistribution Check
Flowchart Notes
Issue the show ip eigrp topology net mask command to verify.
Route Check
Flowchart Notes
Issue the show ip route eigrp command to verify.
Issue the show ip eigrp topology command to verify. If routes are not seen in the topology table, issue the clear ip eigrp topology command.
Flowchart Notes
Issue the show ip eigrp topology net mask command, to find the Router ID (RID). You can find the local RID with the same command on the locally generated external router. In Cisco IOS Software Release 12.1 and later, the show ip eigrp topology command shows the RID.
Reasons for Neighbor Flapping
The stability of the neighbor relationship is of primary concern. A failure in the neighbor relationship is accompanied by increased CPU and bandwidth utilization. EIGRP neighbors can flap for these reasons:
- Underlying link flaps. When an interface goes down, EIGRP takes down the neighbors that are reachable through that interface and flushes all routes learned through that neighbor.
- Misconfigured hello and hold intervals. The EIGRP hold interval can be set independently of the hello interval if you issue the ip hold-time eigrp command. If you set a hold interval smaller than the hello interval, it results in the neighbors flapping continuously. Cisco recommends that the hold time be at least three times the hello interval. If the value is set less than 3 times the hello interval, there is the chance for link flapping or neighborship flapping.
- Loss of hello packets: Hello packets can be lost on overly congested links or error-prone links (CRC errors, Frame errors, or excessive collisions).
- Existence of unidirectional links. A router on a unidirectional link can be able to receive hello packets, but the hello packets sent out are not received at the other end. The existence of this state is usually indicated by the retry limit exceeded messages on one end. If the routers generating retry limit exceeded messages has to form neighborship, then make the link bidirectional for both unicast and multicast. In case tunnel interfaces are used in the topology make sure that the interfaces are advertised properly.
- Route goes stuck-in-active. When a router enters the stuck-in-active state, the neighbors from which the reply was expected are reinitialized, and the router goes active on all routes learned from those neighbors.
- Provision of insufficient bandwidth for the EIGRP process. When sufficient bandwidth is not available, packets can be lost, which causes neighbors to go down.
- Bad serial lines.
- Improperly set bandwidth statements.
- One-way multicast traffic.
- Stuck in active routes.
- Query storms.
EIGRP Neighbors are not Recognized
The EIGRP neighbor relationship is not established over the multipoint GRE tunnel if there is an incorrect NHRP association in the spoke. Next Hop Resolution Protocol (NHRP) is used to discover the addresses of other routers and networks behind the routers that are connected to a nonbroadcast multiaccess (NBMA) network. When a network statement under Eigrp covers both the physical interface and tunnel interface (tunnel interface ip address and physical interface ip address belong to the same major class) and if the phyiscal interface is the source of the tunnel, then the both interfaces have to be separately advertised in the Eigrp to avoid issues with DMVPN. The best practice is to advertise the interfaces using specific subnet advertisements.
This issue can be resolved when you clear the NHRP associations with this command:
Source : ciscoRouter#clear ip nhrp