OSPF Areas and Router Types

OSPF Areas and Router Types

Quick Answer: OSPF (Open Shortest Path First) is a link-state IGP. Routers flood LSAs, build an identical LSDB per area, and run SPF to place best routes in the RIB. Design uses areas with Area 0 as the backbone. Know internal, backbone, ABR, and ASBR roles—multi-area OSPF fails on the exam if you forget that non-backbone areas must connect through Area 0.

IP routing chapters already covered static routes, preference, and longest-match. Real campus and enterprise networks need dynamic discovery when links fail and when dozens of prefixes exist. OSPFv2 (RFC 2328 family; IPv4) is the primary dynamic IGP tested on HCIA-Datacom H12-811. This section builds the topology model: what OSPF is, why areas exist, and which router type does which job. Neighbor formation, LSA types, DR/BDR, and VRP CLI follow in the next sections.

Link-state versus distance-vector (exam contrast)

OSPF routers do not simply pass “I can reach 10.1.0.0/16 via cost 20” as the only story. They advertise links and states. Each router in an area assembles the same database, then independently computes shortest paths. That independence is why consistent LSDB contents matter: if two routers see different topology, they can install inconsistent paths.

Process, Router ID, and network type (mental model)

Before areas, fix three identity ideas:

1. OSPF process ID on Huawei is locally significant. ospf 1 on Router A does not need to match ospf 1 on Router B. Adjacency cares about Area ID, Hello parameters, authentication, network type, and addresses—not equal process numbers.
2. Router ID is a 32-bit dotted-decimal value that uniquely identifies the router inside the OSPF domain. Best practice: set it explicitly (router-id x.x.x.x). If unset, devices typically choose a stable interface address (often highest loopback, then highest active interface—exact selection rules are vendor-detailed, but HCIA expects you to configure Router ID deliberately).
3. Network type (broadcast, P2P, NBMA, etc.) influences Hello behavior and whether a DR/BDR is elected. Ethernet LAN segments usually run as broadcast multi-access; serial/P2P style links often skip DR election. Details appear in the DR/config section.

Why hierarchical areas exist

A flat OSPF domain with hundreds of routers means:

• Every topology change floods widely.
• Every router runs SPF over a huge graph.
• Memory and CPU for the LSDB grow with the whole domain.

Areas partition the domain so most detailed topology stays inside an area. Between areas, OSPF uses summary information (Type 3 LSAs and related behavior) instead of full intra-area link detail. Benefits HCIA expects you to recite:

• Contain flooding of Type 1/2 LSAs within an area.
• Reduce SPF scope when a leaf link flaps (other areas do not always recompute full internal detail).
• Enable structured addressing so ABRs can summarize prefixes at area borders (design skill; exact summary commands deepen later).

Area 0: the backbone rule

Area 0 (0.0.0.0) is the backbone area. Rules that generate exam traps:

• In a multi-area design, non-backbone areas must connect to Area 0 (classically via an ABR that has interfaces in both Area 0 and the non-zero area).
• Traffic between two non-backbone areas is expected to transit the backbone (ABR → Area 0 → ABR).
• A design with only Area 1 and Area 2 and no Area 0 is not a valid standard multi-area OSPF hierarchy for inter-area reachability as taught at associate level.

Single-area OSPF (everything in Area 0, or everything in one non-zero area only for a tiny lab) is fine for small networks. The moment you introduce multiple areas, backbone discipline becomes mandatory in exam answers.

HCIA focuses on standard areas and the backbone rule more than deep stub optimization, but you should recognize that special area types exist to control Type 5 external flooding.

OSPF router types (roles)

Roles are defined by where interfaces live and what routes the router injects, not by chassis model.

Important distinctions:

• A router can be both ABR and ASBR (connects areas and injects externals).
• A router with interfaces only in Area 1 is internal, even if it has a default static route in the RIB that is not redistributed—without redistribution it is not an ASBR.
• “Backbone router” is not a separate election; it is a classification. An ABR is always a backbone router because it must touch Area 0 in the classic model.

Worked campus example

Imagine:

• Core routers C1/C2: interfaces only in Area 0 → backbone internal routers.
• Building distribution D1: GE to core in Area 0, VLANs toward access in Area 10 → ABR.
• Access router A1: only Area 10 interfaces → internal in Area 10.
• Edge router E1 in Area 0 redistributes a branch static default or BGP prefixes → ASBR (and backbone router).

Inter-building traffic between Area 10 and Area 20 goes Area 10 → ABR → Area 0 → ABR → Area 20. If someone “simplifies” by deleting Area 0 and linking Area 10 directly to Area 20 without a backbone, associate-level OSPF rules say the design is wrong.

Metrics and path preference (area context)

OSPF cost on an interface is commonly derived from bandwidth (reference bandwidth / interface bandwidth). Lower cost is better. Within an area, SPF yields the tree. Across areas, routers prefer routes in a well-known order that advanced exams stress; at HCIA, remember at least:

• Intra-area paths are preferred over inter-area paths to the same prefix when both exist in the decision process OSPF defines.
• External routes (from ASBRs) are a different class from pure internal OSPF routes.

You do not need full RFC preference tables here, but you must not claim that a Type 5 external is “just another Type 3 summary.”

Huawei VRP placement (preview)

Configuration belongs in detail later, but area membership is declared under the process:

[Router] ospf 1 router-id 1.1.1.1
[Router-ospf-1] area 0.0.0.0
[Router-ospf-1-area-0.0.0.0] network 10.0.0.0 0.0.0.255

or interface-centric enable patterns. The area context is what makes a router an ABR when the same process has both area 0 and area 10 with networks. Verification concepts: display ospf peer, display ospf lsdb, display ospf routing.

Design checklist for HCIA

1. Prefer explicit Router IDs on every OSPF speaker.
2. Keep Area 0 contiguous in multi-area designs; do not create discontiguous backbones without understanding advanced repair mechanisms.
3. Place ABRs where hierarchy naturally sits (distribution/core border).
4. Limit ASBRs and redistribution points so external LSAs do not surprise the domain.
5. Align IP addressing with areas so future summarization is possible.
6. Remember process ID is local; Area ID and parameters must match for neighbors.

Exam checklist

• Define OSPF as link-state: LSA → LSDB → SPF → RIB/FIB.
• State that Area 0 is the backbone and multi-area non-backbone areas attach via ABR to Area 0.
• Classify internal vs backbone vs ABR vs ASBR from interface and redistribution facts.
• Contrast OSPF with hop-count distance-vector ideas at a one-sentence level.
• Avoid the trap: “ABR means any two areas” without Area 0—classic ABR definition includes the backbone connection in standard hierarchical OSPF teaching.

Next: how two routers actually become neighbors and progress through states to Full adjacency (or stop at 2-Way on multi-access segments).