Learning paths › Scenarios › OSPF, EIGRP and BGP adjacency faults
OSPF, EIGRP and BGP adjacency faults
Adjacency faults across three protocols, one cause at a time: an interface addressed in the wrong subnet, an ACL dropping IP protocol 88, an OSPF area or network-type mismatch, a letter O typed as a zero in an authentication key, a key chain pointing at nothing, a passive interface left on a transit link, authentication applied to the wrong area, and an MTU mismatch that stalls database exchange — closing on BGP sessions that need TCP 179 up and a route to the peer's loopback before any prefix appears.
Lessons
- Fix an interface addressed in the wrong subnet
- Find an ACL that drops IP protocol 88
- Catch an OSPF area mismatch on a transit link
- Replace nonbroadcast with a broadcast model
- Catch a letter O typed as a zero in a key
- See where redistribution and area borders sit
- Keep the network type and work around it
- Unpick an AS mismatch, a VLAN mismatch and a filter
- Match the EIGRP selector to the real interface
- Choose the wildcard that matches the whole subnet
- Start unicast hellos from the hub
- Make both peers agree on authentication type
- Change Frame Relay mappings without adding new ones
- See how network type shapes the computed routes
- Remove a passive interface from a transit link
- Catch a key-chain name that points at nothing
- Find an MTU mismatch that stalls database exchange
- Apply authentication to the area that actually peers
- See an interface setting override the area policy
- Get TCP 179 up before expecting any routes
- Keep the BGP AS and the EIGRP AS apart
- State the loopback-to-loopback goal and its return route
- Route to the peer loopback before peering with it
- Keep the route-map constraint in the picture