Cisco IOS — OSPF Configuration

Overview

OSPF (Open Shortest Path First) is the most widely deployed interior gateway routing protocol in enterprise and service provider networks. It is link-state: every router builds a complete map of the network topology (the Link-State Database) and runs Dijkstra’s shortest-path algorithm to determine the best routes. Unlike distance-vector protocols such as RIP, OSPF routers do not send their routing tables to neighbours — they share raw topology information, giving each router the intelligence to calculate its own optimal paths.

OSPFv2 handles IPv4 (RFC 2328). This article covers the full IOS configuration workflow: process creation, router ID, the network command, interface-mode configuration, passive interfaces, neighbour verification, cost tuning, DR/BDR elections, network types, multi-area design, and the default route advertisement.

Starting OSPF

Process ID

router ospf 1

The process ID (1 in this example) is locally significant — it identifies this OSPF process on this router only. Two routers can use different process IDs and still become OSPF neighbours. Multiple OSPF processes can run on one router simultaneously (used for redistribution), though this is rare in simple designs.

Router ID (RID)

The Router ID uniquely identifies this router in the OSPF domain. Every router must have a unique RID — duplicate RIDs prevent adjacency formation.

IOS selects the RID automatically using this priority order:

1. Manual router-id command (highest priority)
2. Highest IP address on any loopback interface (up at time of OSPF startup)
3. Highest IP address on any physical interface (up at time of OSPF startup)

Manual assignment is strongly recommended — it makes the RID predictable and stable regardless of which interfaces are up when OSPF starts.

router ospf 1
 router-id 1.1.1.1

Use a loopback-style address (e.g., 1.1.1.1, 2.2.2.2) that clearly identifies the router. The RID does not need to be routable or even configured on an interface — it is an identifier, not a destination address.

After changing the router-id, reset OSPF to apply it:

clear ip ospf process

This resets all OSPF neighbour relationships and triggers a new RID election. Do this during a maintenance window — it causes a brief routing interruption.

Enabling OSPF on Interfaces

Classic Method — Network Statement

The network command activates OSPF on any interface whose IP address falls within the specified range and assigns it to an area.

router ospf 1
 router-id 1.1.1.1
 network 10.1.0.0 0.0.255.255 area 0
 network 192.168.1.0 0.0.0.255 area 0

The second argument is a wildcard mask — the bitwise inverse of a subnet mask. In the wildcard, 0 means “this bit must match” and 1 means “ignore this bit.”

To match a specific interface exactly (common best practice):

network 10.1.10.1 0.0.0.0 area 0

Interface Mode Method (Newer Style)

Configure OSPF directly on each interface instead of using a network statement. This is more explicit and easier to read.

interface GigabitEthernet0/0/0
 ip ospf 1 area 0

This activates OSPF process 1 on this interface in area 0. There is no need for a network statement when using this method. Both methods work; interface mode is becoming the preferred style on IOS XE platforms.

Passive Interfaces

By default, every interface enabled for OSPF sends and receives Hello packets. On LAN interfaces that connect to hosts (not other routers), sending Hellos wastes bandwidth and exposes topology information unnecessarily. Passive interfaces advertise the connected subnet into OSPF but suppress Hello packets entirely — no OSPF neighbours can form on that interface.

Passive Single Interface

router ospf 1
 passive-interface GigabitEthernet0/0/0

Passive Default (All Interfaces)

In networks where most interfaces face hosts and only a few face other routers, the passive-interface default approach is cleaner:

router ospf 1
 passive-interface default
 no passive-interface GigabitEthernet0/0/1
 no passive-interface GigabitEthernet0/0/2

All interfaces are passive by default. Use no passive-interface to re-enable Hello transmission only on the interfaces that actually connect to other OSPF routers.

Neighbour Requirements

Two routers form an OSPF adjacency only when all of these parameters match on the common interface:

MTU mismatch is a particularly nasty problem — the neighbours appear in show ip ospf neighbor but the Database Description exchange fails silently.

Verifying OSPF Neighbours

show ip ospf neighbor

Key output fields:

OSPF Neighbour States

Healthy states for Ethernet (broadcast) networks:

• FULL/DR — fully adjacent with the DR
• FULL/BDR — fully adjacent with the BDR
• FULL/- — fully adjacent (point-to-point link, no DR)
• 2WAY/DROTHER — normal state between two DROthers; not fully adjacent with each other

States that indicate a problem:

• INIT — one-way communication; own RID not in remote Hello
• EXSTART or EXCHANGE stuck — MTU mismatch or authentication issue
• LOADING stuck — corrupted LSA or MTU issue

show ip ospf neighbor detail

Shows extended information including Dead/Hello timers, MTU, authentication type, area, and adjacency uptime.

Verifying OSPF Routes

show ip route ospf

OSPF route codes:

• O — OSPF intra-area (within the same area)
• O IA — OSPF inter-area (route learned from another area via an ABR)
• O E1 / O E2 — OSPF external (redistributed from another protocol)
• [110/cost] — Administrative Distance 110 / OSPF metric (path cost)

show ip ospf interface brief

Shows each OSPF-enabled interface with: area, cost, DR/BDR state, hello/dead timers, and neighbour counts. The State column shows DR, BDR, or DROTH for broadcast networks; P2P for point-to-point.

show ip ospf

Summary of the OSPF process: RID, areas, SPF run count, reference bandwidth, and redistribution details.

show ip ospf database

The Link-State Database — all LSAs this router has collected. For single-area OSPF, you should see Type 1 (Router) LSAs from every router.

DR and BDR Election

On broadcast networks (Ethernet), OSPF elects a Designated Router (DR) and Backup Designated Router (BDR) to reduce the number of OSPF adjacencies. All other routers (DROthers) form full adjacency only with the DR and BDR — not with each other. This reduces LSA flooding overhead on multi-access segments.

Election Rules

1. Highest OSPF interface priority wins (range 0–255; default 1). Priority 0 means “never become DR or BDR.”
2. Tiebreaker: highest Router ID

There is no preemption. Once a DR is elected, it keeps that role until it fails, regardless of a higher-priority router joining later. The next election happens only when the current DR fails.

Configuring Priority

interface GigabitEthernet0/0/0
 ip ospf priority 99

Set priority to 0 on interfaces where you never want the router to become DR:

interface GigabitEthernet0/0/0
 ip ospf priority 0

This is useful on access switches that run OSPF but should not become DR for a distribution segment.

Multicast Addresses

OSPF Cost

OSPF cost is the metric used to determine the best path. The cost of a path equals the sum of outgoing interface costs along the entire path from source to destination.

Default Cost Formula

Cost = Reference_bandwidth (Kbps) / Interface_bandwidth (Kbps)

Default reference bandwidth is 100 Mbps (100,000 Kbps). This creates a problem: FastEthernet (100 Mbps), GigabitEthernet (1 Gbps), and 10 GigabitEthernet all calculate to cost 1 — they are indistinguishable. OSPF will treat them equally and may make poor routing decisions.

Setting the Reference Bandwidth

Set reference bandwidth on all routers to a value higher than the fastest link in your network:

router ospf 1
 auto-cost reference-bandwidth 10000

This sets the reference to 10,000 Mbps (10 Gbps). Now:

All routers in the OSPF domain must use the same reference bandwidth — inconsistent values produce incorrect SPF calculations.

Manual Cost Override

Set the cost directly on an interface, overriding the calculated value:

interface GigabitEthernet0/0/1
 ip ospf cost 50

Manual cost is useful when the calculated value does not reflect the real-world preference (for example, a high-latency satellite link that has a low numerical bandwidth cost but should be avoided for latency-sensitive traffic).

Bandwidth Statement

The bandwidth command in Kbps influences the cost calculation but does not change the physical speed:

interface Serial0/0/0
 bandwidth 512

This tells OSPF the link is 512 Kbps for cost calculation purposes. The physical rate is whatever the circuit provider delivers. Use this only when the IOS default bandwidth (1544 Kbps for serial) does not reflect the actual circuit.

OSPF Hello and Dead Timers

Hellos are sent every 10 seconds on Ethernet (default). Dead time is 4× Hello = 40 seconds. If a router does not hear from a neighbour within the dead time, it declares the neighbour down and triggers SPF recalculation.

Both timers must match on both sides of a link to form adjacency.

interface GigabitEthernet0/0/1
 ip ospf hello-interval 5
 ip ospf dead-interval 20

Reducing timers speeds up failure detection but increases CPU and bandwidth usage. Increasing them reduces overhead on large slow networks.

OSPF Network Types

The OSPF network type controls DR/BDR election and how neighbours are discovered:

Default: Ethernet = broadcast, Serial = point-to-point.

For Ethernet WAN links (where only two routers are connected), changing to point-to-point eliminates the DR/BDR election overhead and speeds up convergence:

interface GigabitEthernet0/0/1
 ip ospf network point-to-point

Both ends must use the same network type — a mismatch causes the neighbours to reach adjacency but fail to exchange routes correctly.

Multi-Area OSPF

As a network grows, a single OSPF area becomes unwieldy. Every router stores every LSA; the SPF calculation runs on the full topology. Multi-area OSPF partitions the network so each area only floods within itself.

Area Design Rules

• Area 0 (backbone) is mandatory. All other areas must connect to Area 0.
• ABR (Area Border Router) connects two or more areas. It has a full LSDB for each area it belongs to and generates Type 3 Summary LSAs to advertise routes between areas.
• Intra-area routes (O) — destination in the same area
• Inter-area routes (O IA) — destination in a different area, learned via ABR

LSA Types

Type 3 LSAs do not carry the full topology of the remote area — just the prefix, mask, and cost. Routers in one area cannot see inside another area’s topology. This reduces SPF complexity.

Multi-Area Configuration Example

R1 (Area 0 and Area 1 — ABR):

router ospf 1
 router-id 1.1.1.1
 network 10.0.0.0 0.0.0.3 area 0
 network 10.1.10.0 0.0.0.255 area 1

R2 (Area 0 only):

router ospf 1
 router-id 2.2.2.2
 network 10.0.0.0 0.0.0.3 area 0
 network 10.2.20.0 0.0.0.255 area 0

R3 (Area 1 only):

router ospf 1
 router-id 3.3.3.3
 network 10.1.10.0 0.0.0.255 area 1

R1 is the ABR — it sits in both Area 0 and Area 1 and generates Type 3 LSAs to advertise Area 1 prefixes into Area 0 and vice versa.

Route Summarisation at ABR

An ABR can summarise inter-area routes, reducing the number of Type 3 LSAs flooded into Area 0:

router ospf 1
 area 1 range 10.1.0.0 255.255.0.0

All routes from Area 1 that fall within 10.1.0.0/16 are summarised into a single Type 3 LSA advertised to Area 0. Summarisation reduces the routing table size on routers in other areas and limits the impact of topology changes within Area 1.

Advertising a Default Route

An OSPF router can inject a default route into the OSPF domain so all other routers point toward it as the internet exit:

ip route 0.0.0.0 0.0.0.0 203.0.113.1

router ospf 1
 default-information originate

default-information originate creates an OSPF Type 5 External LSA for 0.0.0.0/0 and floods it to all OSPF areas. Other routers install O*E2 0.0.0.0/0 in their routing tables.

The always keyword advertises the default route even if the router itself does not currently have a default route in its routing table:

router ospf 1
 default-information originate always

Use with care — if the internet link is down and always is configured, routers will continue pointing traffic at the OSPF exit router, which will blackhole it.

Quick Reference