OSPF Adjacency Not Forming Between Neighbors

Step 1: Check Neighbor State

# Cisco IOS
show ip ospf neighbor detail

# Linux (FRR/Quagga)
sudo vtysh -c "show ip ospf neighbor detail"

# Key states and their meaning:
# DOWN      = No Hello received
# INIT      = Hello received but Router ID not in Hello's neighbor list (one-way)
# 2-WAY     = Hello received bidirectionally (non-DR/BDR stop here on broadcast)
# EXSTART   = Negotiating master/slave for DD exchange
# EXCHANGE  = Sending Database Descriptions
# LOADING   = Requesting missing LSAs
# FULL      = Adjacency complete, database synchronised

Step 2: Verify Hello Parameters Match

# Cisco IOS
show ip ospf interface GigabitEthernet0/0
# Look for: Hello 10, Dead 40, Wait 40, Retransmit 5
# Both sides MUST match: Hello interval, Dead interval, Area ID, Authentication type, MTU

# Linux FRR
sudo vtysh -c "show ip ospf interface eth0"

Step 3: Confirm Area ID and Network Type Match

# Area 0 (backbone) and area 1 are different — peers in different areas cannot form adjacency
# Network type: both must be same (point-to-point, broadcast, non-broadcast)

# Cisco
show running-config | section ospf
# Check: network area and ip ospf network type

# FRR
sudo vtysh -c "show running-config" | grep -A5 "ospf"

Step 4: Examine OSPF Debug Output

# Cisco IOS (limit output with ACL before enabling)
debug ip ospf adj   # Shows adjacency state machine transitions
debug ip ospf hello # Shows Hello packet details

# FRR
sudo vtysh
conf t
debug ospf event
debug ospf packet all
end
# Watch: sudo journalctl -u frr -f

# Disable debug after diagnosis:
no debug ip ospf adj

Step 5: Check MTU Mismatch

# OSPF will not form adjacency if MTU differs between peers (DD packets are rejected)
# On interface connected to OSPF peer:
ip link show eth0
# MTU in ip link must match both sides

# Cisco
show interface GigabitEthernet0/0 | include MTU
# Or bypass MTU check (not recommended for production):
ip ospf mtu-ignore

Fix: Match Hello and Dead Intervals

# Cisco IOS — set on the interface
interface GigabitEthernet0/0
  ip ospf hello-interval 10
  ip ospf dead-interval 40

# FRR
sudo vtysh
conf t
interface eth0
  ip ospf hello-interval 10
  ip ospf dead-interval 40
end
write memory

Fix: Correct Area Assignment

# Cisco IOS
router ospf 1
  no network 10.1.1.0 0.0.0.255 area 1   # Remove wrong area
  network 10.1.1.0 0.0.0.255 area 0       # Add correct area

# FRR
router ospf
  no network 10.1.1.0/24 area 1
  network 10.1.1.0/24 area 0

Fix: Resolve MTU Mismatch

# Option 1: Fix the interface MTU to match
sudo ip link set eth0 mtu 1500

# Option 2: Enable OSPF MTU ignore (use only if MTU mismatch is by design)
# Cisco:
interface GigabitEthernet0/0
  ip ospf mtu-ignore

# FRR:
interface eth0
  ip ospf mtu-ignore

Fix: Configure Matching Authentication

# Cisco IOS — MD5 authentication
interface GigabitEthernet0/0
  ip ospf authentication message-digest
  ip ospf message-digest-key 1 md5 MySecretKey

# FRR
interface eth0
  ip ospf authentication message-digest
  ip ospf message-digest-key 1 md5 MySecretKey

• Standardise OSPF interface parameters (Hello/Dead intervals, authentication, network type) in a configuration template and apply via automation
• Implement OSPF neighborship monitoring: alert when any expected adjacency drops below FULL state for more than 60 seconds
• Use show ip ospf neighbor as a post-change verification step in all network change runbooks
• Enable OSPF authentication on all interfaces to prevent accidental or malicious neighborship formation
• Document MTU values for all OSPF-enabled interfaces and verify end-to-end path MTU with ping using the df-bit option during commissioning