5.2 OSPFv2 and OSPFv3 Multi-Area Architecture: LSAs, Area Types, and Summarization

Key Takeaways

  • OSPF multi-area design constrains the link-state database (LSDB) footprint, isolates SPF recalculations to local areas, and requires all non-backbone areas to connect to Backbone Area 0.

  • OSPF neighbor adjacency establishes through eight discrete states: Down, Attempt, Init, 2-Way, ExStart, Exchange, Loading, and Full, with DR/BDR election occurring at 2-Way.

  • LSA Types 1 through 5 and Type 7 construct the OSPF topology, distinguishing intra-area router links, multiaccess network segments, inter-area summaries, and external redistributed routes.

  • Special area types (Stub, Totally Stubby, NSSA, and Totally NSSA) prune external and summary LSAs, substituting default routes to reduce router memory and CPU overhead.

  • OSPFv3 supports IPv6 and IPv4 natively, operates per-link rather than per-subnet, and introduces Link-LSAs (Type 8) and Intra-Area-Prefix-LSAs (Type 9) to decouple IP prefix announcements from SPF topology graphs.

Last updated: October 2026

OSPFv2 and OSPFv3 Multi-Area Architecture: LSAs, Area Types, and Summarization

Open Shortest Path First (OSPF) is an open-standard, link-state routing protocol engineered for high scalability, rapid convergence, and vendor interoperability. In large enterprise infrastructures, operating a single flat OSPF area causes the Link-State Database (LSDB) to expand dramatically. Every link flap triggers full Shortest Path First (SPF) calculations across every router. Multi-area OSPF partitions the network into a structured two-tier hierarchy that localizes link-state updates, conserves memory and CPU resources, and enables administrative route summarization.


OSPF Multi-Area Hierarchical Design

OSPF enforces a strict two-tier hierarchical architecture centered on Backbone Area 0 (also referenced as 0.0.0.0). All non-backbone areas (e.g., Area 1, Area 20) must physically connect to Area 0, or link to it through a virtual link across a transit area.

                      +---------------------------------+
                      |         BACKBONE AREA 0         |
                      |    (High-Speed Transport Core)  |
                      +---------------------------------+
                                /             \
                               /               \
                  [ABR Router 1]               [ABR Router 2]
                              /                 \
                             /                   \
        +-----------------------+             +-----------------------+
        |    STANDARD AREA 1    |             |    STUB / NSSA AREA 2 |
        |  (Campus Access Pod)  |             |     (Branch Office)   |
        |  [Internal Routers]   |             |     [ASBR Router]     |
        +-----------------------+             +-----------------------+

OSPF Router Classifications

Routers in a multi-area topology fulfill distinct roles based on their interface locations:

  • Internal Router: A router with all active interfaces assigned to a single OSPF area.
  • Backbone Router (BR): Any router maintaining at least one active interface inside Area 0.
  • Area Border Router (ABR): A router with interfaces attached to two or more different areas, with at least one interface residing in Area 0. The ABR maintains a discrete LSDB for each attached area, runs independent SPF calculations per area, and summarizes topological information between areas.
  • Autonomous System Boundary Router (ASBR): A router that redistributes routes learned from external routing domains (such as BGP, EIGRP, or static routes) into the OSPF domain.

OSPF Neighbor Adjacency Lifecycle

Before exchanging routing information, OSPF routers must discover adjacent neighbors and transition through eight defined states:

Down ---> Attempt ---> Init ---> 2-Way ---> ExStart ---> Exchange ---> Loading ---> Full
                        ^          ^                                                  ^
                        |          |                                                  |
                  Hello seen   DR/BDR Elected                                   LSDB Synchronized
  1. Down: Initial state; no Hello packets have been received from the neighbor.
  2. Attempt: Applicable only to Non-Broadcast Multi-Access (NBMA) networks; the router sends unicast Hellos to statically configured neighbors.
  3. Init: A Hello packet was received from a neighbor, but the local router's own Router ID (RID) is not yet listed in the neighbor's active neighbor list (unidirectional communication).
  4. 2-Way: Bi-directional communication is established; the local router observes its own RID inside the neighbor's Hello packet. DR/BDR elections occur at this state on multiaccess broadcast links. Routers that are neither DR nor BDR remain in the 2-Way state with each other (DROTHER).
  5. ExStart: The initial step of LSDB synchronization. Routers negotiate a Master/Slave relationship and determine an initial Database Description (DBD) sequence number based on the highest Router ID.
  6. Exchange: Routers exchange DBD packets describing the summary contents of their respective LSDBs.
  7. Loading: Routers send Link-State Requests (LSRs) for any missing or newer LSAs identified during DBD exchange, receiving the full LSA records via Link-State Updates (LSUs).
  8. Full: The LSDB is completely synchronized between adjacent routers. Full routing adjacency is established.

Designated Router (DR) and Backup Designated Router (BDR) Election

On multiaccess segments (such as Ethernet), if every router formed full adjacencies with every other router, N×(N−1)/2N \times (N - 1) / 2 adjacencies would result. To curb excessive LSA flooding, OSPF elects a DR and a BDR. All other routers (DROTHER) form full adjacencies exclusively with the DR and BDR, reducing total adjacencies to 2N−32N - 3.

  • Election Logic: The router with the highest interface OSPF priority (range 0–255, default 1) is elected DR. A priority of 0 permanently disqualifies a router from becoming DR or BDR. In case of a priority tie, the router with the highest 32-bit Router ID (RID) wins.
  • Non-Preemptive Behavior: DR and BDR elections are non-preemptive. If a router with a higher priority or RID boots up after the election has concluded, it will not overthrow the existing DR or BDR until the current DR/BDR process restarts.
  • Multicast Addresses:
    • 224.0.0.5 (AllSPFRouters): Listened to by all OSPF routers.
    • 224.0.0.6 (AllDRouters): Listened to exclusively by the DR and BDR to receive link-state updates from DROTHER devices.

Master OSPF LSA Types Reference

Link-State Advertisements (LSAs) represent the operational currency of OSPF. Each LSA type conveys a specific category of routing or topological data:

LSA TypeNameOriginating RouterFlooding BoundaryPurpose & Contents
Type 1Router LSAEvery OSPF routerLocal area onlyDescribes the router's active interfaces, IP prefixes, interface types, and link costs. Never crosses an ABR.
Type 2Network LSADesignated Router (DR)Local area onlyGenerated on multiaccess networks. Lists the subnet mask and all adjacent routers attached to the transit link.
Type 3Summary Net LSAArea Border Router (ABR)Inter-area (flooded across adjacent areas)Advertises intra-area subnets learned from one area into other areas. Represents reachability rather than full topology.
Type 4Summary ASBR LSAArea Border Router (ABR)Inter-area (flooded to all areas except origination area)Advertises a host route (/32) pointing to an ASBR located in a different area, allowing remote routers to calculate cost to the ASBR.
Type 5AS External LSAASBRAutonomous System-wide (all standard areas)Advertises external routes redistributed into OSPF (e.g., BGP, static). Can be external type 1 (E1, metric accumulates) or type 2 (E2, fixed metric).
Type 7NSSA External LSAASBR inside an NSSANSSA area onlyAdvertises external routes inside a Not-So-Stubby Area. Converted into Type 5 LSAs by the ABR before injection into Area 0.

Special OSPF Area Types Matrix

To shield low-end branch routers or edge sites from massive LSDB overhead, OSPF defines specialized area types that selectively block external (Type 5) and inter-area summary (Type 3) LSAs:

Area TypeAllowed LSAsBlocked LSAsDefault Route InjectionUse Case / Design Role
Standard AreaTypes 1, 2, 3, 4, 5NoneNone by default; optional manual configurationDefault OSPF area behavior; full LSDB synchronization.
Stub AreaTypes 1, 2, 3Types 4 and 5ABR automatically injects a Type 3 default route (0.0.0.0/0)Branch offices with a single exit point to Area 0; eliminates external routes.
Totally Stubby Area (Cisco Proprietary)Types 1, 2Types 3, 4, and 5ABR automatically injects a Type 3 default route (0.0.0.0/0)Extreme LSDB optimization; removes all inter-area summaries and external routes.
Not-So-Stubby Area (NSSA)Types 1, 2, 3, 7Types 4 and 5Not automatic; configured via area nssa default-information-originateBranch sites requiring an ASBR to redistribute local routes while blocking core Type 5 LSAs.
Totally NSSA (Cisco Proprietary)Types 1, 2, 7Types 3, 4, and 5ABR automatically injects a Type 3 default route (0.0.0.0/0)Optimal NSSA variant; blocks inter-area summaries while supporting local ASBR redistribution.

NSSA P-Bit Translation: When an ASBR inside an NSSA generates a Type 7 LSA, it sets the Propagate bit (P-bit) in the LSA header. When the Type 7 LSA reaches the ABR, the ABR translates it into a standard Type 5 LSA and floods it across Backbone Area 0. If multiple ABRs attach to the NSSA, the ABR with the highest Router ID performs the translation.


Inter-Area vs. External Route Summarization

Route summarization shrinks routing tables, suppresses downstream route calculation during link flaps, and maintains clean address boundaries.

1. Inter-Area Summarization on ABRs (area range)

Summarizes Type 1 and Type 2 LSAs from an originating area into a single consolidated Type 3 summary LSA advertised into adjacent areas:

router ospf 1
 router-id 1.1.1.1
 area 1 range 10.1.0.0 255.255.0.0

When this command is configured, the ABR automatically generates a discard route to Null0 (10.1.0.0/16 -> Null0) to prevent routing loops if packets arrive for unallocated subnets within the summarized range.

2. External Route Summarization on ASBRs (summary-address)

Summarizes redistributed external routes into a single consolidated Type 5 (or Type 7) external LSA before flooding into the OSPF domain:

router ospf 1
 summary-address 172.16.0.0 255.255.0.0

OSPFv3 Architectural Enhancements

OSPFv3 updates OSPF to route IPv6 and IPv4 natively under a unified framework:

  • Runs Per-Link, Not Per-Subnet: OSPFv3 interfaces communicate across a common physical link regardless of configured subnets. Neighbors establish adjacencies over IPv6 link-local addresses (fe80::/10).
  • Decoupling Topology from IP Addressing: In OSPFv2, Type 1 and Type 2 LSAs carry both physical topology links and IPv4 subnet prefixes. Any IP subnet change causes an SPF calculation. OSPFv3 solves this by carrying pure topology in Type 1 and 2 LSAs, moving IP prefix advertisements into two new LSA types:
    • Link-LSA (Type 8): Flooded strictly link-local; advertises the router's link-local address and all IPv6 global unicast prefixes assigned to that interface.
    • Intra-Area-Prefix-LSA (Type 9): Flooded area-wide; associates IPv6 prefixes with a router or transit network without altering the SPF tree structure.
  • Support for Multiple Address Families: Under the router ospfv3 process, engineers configure dedicated address family submodes (address-family ipv4 unicast and address-family ipv6 unicast). The address families are kept apart by the 8-bit Instance ID field in the OSPFv3 header; RFC 5838 assigns ranges such as 0–31 for IPv6 unicast and 64–95 for IPv4 unicast.

Configuring and Verifying Multi-Area OSPF

Topic 3.2.b asks you to configure simple OSPFv2 and OSPFv3 environments with multiple normal areas, summarization, and filtering, including neighbor adjacency, point-to-point and broadcast network types, and passive interfaces. The router below is an ABR between area 0 and area 1.

OSPFv2 on an ABR

interface Loopback0
 ip address 10.255.0.1 255.255.255.255
!
interface GigabitEthernet0/0
 description Link to core (area 0)
 ip address 10.0.12.1 255.255.255.252
 ip ospf network point-to-point
!
interface GigabitEthernet0/1
 description Access LAN (area 1)
 ip address 10.1.10.1 255.255.255.0
!
router ospf 1
 router-id 10.255.0.1
 network 10.0.12.0 0.0.0.3 area 0
 network 10.255.0.1 0.0.0.0 area 0
 network 10.1.10.0 0.0.0.255 area 1
 passive-interface GigabitEthernet0/1
 area 1 range 10.1.0.0 255.255.0.0

Interface-level configuration is an alternative to network statements: ip ospf 1 area 1 under an interface does the same job and is easier to audit.

OSPFv3 for IPv6

ipv6 unicast-routing
!
interface GigabitEthernet0/0
 ipv6 address 2001:db8:0:12::1/64
 ospfv3 1 ipv6 area 0
!
interface GigabitEthernet0/1
 ipv6 address 2001:db8:1:10::1/64
 ospfv3 1 ipv6 area 1
!
router ospfv3 1
 router-id 10.255.0.1
 address-family ipv6 unicast
  passive-interface GigabitEthernet0/1
  area 1 range 2001:db8:1::/48

OSPFv3 still uses a 32-bit router ID. If the router has no IPv4 address to borrow, you must set router-id manually or the process will not start. The older syntax, ipv6 router ospf 1 with ipv6 ospf 1 area 0 on interfaces, is still common.

Neighbor adjacency requirements

Must match on the linkSymptom if it does not
Area IDNo neighbor; the router logs a mismatched-area message
Hello and dead intervalsNo neighbor (never reaches 2-Way)
Primary IP subnet (and the mask on broadcast networks)No neighbor
Authentication type and keyNo neighbor
Stub or NSSA area flagNo neighbor
Interface MTUNeighbors stick in ExStart/Exchange because database description packets are rejected
Network type (point-to-point vs. broadcast)Hellos can match and the adjacency may form, but routes are missing because the two sides describe the link differently

Router IDs must also be unique. A duplicate router ID causes log errors and unstable routing.

Broadcast vs. point-to-point network types

Network typeDefault onHello / deadDR/BDR election
BroadcastEthernet10 s / 40 sYes
Point-to-pointSerial (HDLC/PPP) and GRE tunnels10 s / 40 sNo

On a two-router Ethernet link, ip ospf network point-to-point removes the DR/BDR election and the Type 2 LSA, so the adjacency forms faster and the LSDB is simpler. Both ends must use the same network type.

Passive interfaces

passive-interface keeps the interface's subnet in OSPF but stops sending Hellos on it, so no neighbor can form there. Use it on user LANs and loopbacks. passive-interface default makes every interface passive; then enable only the uplinks with no passive-interface GigabitEthernet0/0.

Filtering OSPF routes

ToolWhere it worksEffect
area 1 range 10.1.0.0 255.255.0.0 not-advertiseABRSuppresses the Type 3 summary for that range so other areas do not learn it
area 1 filter-list prefix NO-LAB inABRFilters Type 3 LSAs entering area 1 using a prefix list
distribute-list prefix NO-LAB inAny routerKeeps matching OSPF routes out of the local routing table; LSAs are still flooded, so other routers are unaffected
summary-address 172.16.0.0 255.255.0.0 not-advertiseASBRSuppresses the matching external routes

Verification commands

CommandWhat to check
show ip ospf neighborNeighbor router ID, state (FULL, or 2WAY between DROTHERs), DR/BDR role, and dead timer
show ip ospf interface briefArea, cost, state (DR, BDR, DROTHER, P2P), and neighbor count per interface
show ip ospf interface GigabitEthernet0/0Network type, hello/dead timers, priority, and passive status
show ip route ospfO (intra-area), O IA (inter-area), O E1/E2, and O N1/N2 routes
show ip ospf databaseLSAs by type and area
show ospfv3 neighborOSPFv3 adjacencies
Test Your Knowledge

An enterprise network architect configures an OSPF area as a 'Totally Stubby Area'. What LSAs will the Area Border Router (ABR) block from entering this area, and how do internal routers reach external destinations?

A

The ABR blocks Type 5 external LSAs only, and floods Type 3 summary LSAs for all inter-area prefixes.

B

The ABR blocks Type 1 and Type 2 LSAs, replacing them with a single Type 4 ASBR summary.

C

The ABR blocks Type 3, Type 4, and Type 5 LSAs, automatically injecting a single Type 3 default route (0.0.0.0/0).

D

The ABR converts Type 5 LSAs into Type 7 LSAs, requiring manual configuration of a default static route.

Test Your Knowledge

On an OSPF broadcast Ethernet segment with five active routers, what criteria determines which device is elected Designated Router (DR)?

A

The highest interface OSPF priority wins; if priorities tie, the highest router ID (RID) wins.

B

The router that has been running the longest, with the lowest interface MAC address used to break any ties.

C

The router configured with the lowest loopback IP address, regardless of the interface priority values set on the segment.

D

The router with the fastest physical interface bandwidth and lowest delay metric.

Test Your Knowledge

Which statement accurately describes how OSPFv3 decouples IP addressing from the physical topology graph compared to OSPFv2?

A

OSPFv3 eliminates Area Border Routers and uses BGP path attributes to advertise IPv6 subnets across areas.

B

OSPFv3 removes Type 1 and Type 2 LSAs completely, relying entirely on Type 5 external LSAs for IPv6 reachability.

C

OSPFv3 embeds IPv6 prefixes directly into modified Type 4 LSAs, while link metrics are carried in Type 7 LSAs.

D

Type 1 and 2 LSAs carry only topology; prefixes move to Intra-Area-Prefix (Type 9) and Link (Type 8) LSAs.

Sections you finish are checked off in the contents.