8.3 OSPF Designated Routers and Multi-Area Deployments

OSPF Designated Routers and Multi-Area Deployments

Quick Summary: In campus networks with multi-access broadcast media (such as shared VLANs interconnecting multiple Layer 3 switches), establishing full mesh adjacencies between every pair of routers creates excessive control plane overhead. OSPF solves this by electing a Designated Router (DR) and Backup Designated Router (BDR) to serve as central focal points for LSA synchronization. To scale beyond local broadcast domains, OSPF implements a two-tier hierarchical multi-area architecture centered around Area 0 (the Backbone Area). Multi-area designs limit Shortest Path First (SPF) calculation domains, reduce Link-State Database (LSDB) memory consumption, and restrict the flooding of Type 1 (Router), Type 2 (Network), and Type 3 (Summary) LSAs.

---

OSPF Network Types: Broadcast vs. Point-to-Point

OSPF dynamically alters its operational behavior based on the underlying Layer 2 media type. The two predominant network types deployed in enterprise campus environments are:

1. Broadcast Networks

• Default On: Ethernet interfaces and Switched Virtual Interfaces (SVIs) configured for routing.
• Topology: Multi-access segment where three or more routers can connect simultaneously across a shared Layer 2 broadcast domain.
• DR/BDR Requirement: Elects a DR and BDR to manage flooding.
• Timers: Default Hello interval = 10 seconds; Dead interval = 40 seconds.
• Addressing: Exchanging traffic uses multicast addresses 224.0.0.5 and 224.0.0.6.

2. Point-to-Point (P2P) Networks

• Default On: Serial links or explicitly configured Ethernet links (ip ospf network point-to-point).
• Topology: Connects exactly two routers across a dedicated routed link (such as a /30 or /31 inter-switch link).
• DR/BDR Requirement: No DR or BDR is elected. The two routers form a direct, full adjacency without election delays.
• Timers: Default Hello interval = 10 seconds; Dead interval = 40 seconds.
• Addressing: Traffic is transmitted exclusively to 224.0.0.5.

AOS-CX Campus Design Tip: For dedicated switch-to-switch routed uplinks carrying only two routers, network engineers should explicitly configure ip ospf network point-to-point. This bypasses DR/BDR election wait times, streamlines LSDB representation, and accelerates network convergence.

---

DR and BDR Election Mechanics

On a broadcast network with n routers, if every router established a full adjacency with every other router, the total number of full adjacencies would grow exponentially according to the formula:

Total Adjacencies = [n * (n - 1)] / 2

In a subnet with 10 routers, this requires [10 * (10 - 1)] / 2 = 45 full adjacencies, generating 45 separate streams of LSDB exchanges and redundant LSA flooding. To prevent this overhead, OSPF elects two specific roles on multi-access segments:

• Designated Router (DR): The primary router responsible for generating Type 2 Network LSAs and synchronizing LSDB updates for the segment.
• Backup Designated Router (BDR): A standby router that listens to all OSPF traffic and seamlessly takes over the DR role if the primary DR fails.
• DROther Routers: All other routers on the segment. DROther routers form Full adjacencies only with the DR and BDR. Between two DROther routers, the adjacency stops and remains permanently in the 2-Way state.

With a DR and BDR in place, the number of required adjacencies scales linearly to just 2n - 3, reducing control plane overhead dramatically.

+-----------------------------------------------------------------------------------+
|                         DR / BDR ELECTION HIERARCHY                              |
|                                                                                   |
|  1. OSPF INTERFACE PRIORITY (0 to 255)                                            |
|     Highest priority wins DR; second-highest wins BDR.                            |
|     Priority = 0: COMPLETELY INELIGIBLE (Never becomes DR or BDR).               |
|                                      |                                            |
|                                      v (Tie in priority)                          |
|  2. ROUTER ID (RID) TIEBREAKER                                                    |
|     Highest 32-bit dotted-decimal Router ID wins the election.                    |
|                                                                                   |
|  3. NON-PREEMPTIVE ELECTION RULE                                                  |
|     Once a DR and BDR are active, a newly added router with a higher priority      |
|     or higher RID will NOT take over the DR role. Elections are non-preemptive!   |
+-----------------------------------------------------------------------------------+

The Election Algorithm and Non-Preemption

1. Interface Priority: The administrator assigns priority per interface (ip ospf priority <0-255>). The default priority is 1.
   • A router configured with priority 0 immediately disqualifies itself from becoming DR or BDR, remaining a permanent DROther.
   • The router with the highest priority is elected DR; the second highest becomes BDR.
2. Router ID Tiebreaker: If priorities are identical (e.g., all switches remain at default priority 1), the switch with the numerically highest Router ID is elected DR, followed by the second highest as BDR.
3. Non-Preemptive Operation: OSPF DR/BDR elections are strictly non-preemptive. If a switch with priority 255 boots up on an active segment where a switch with priority 1 has already been elected DR, the existing DR retains its role. Preemption is prevented because forcing a new election would tear down active adjacencies and interrupt packet forwarding across the campus.

Multicast Addressing on Broadcast Segments

• 224.0.0.5 (AllSPFRouters): Listened to by all OSPF-enabled routers. Hellos are sent here, and the DR floods LSUs to this address to update all DROthers.
• 224.0.0.6 (AllDRouters): Listened to only by the DR and BDR. When a DROther detects a network topology change, it transmits an LSU specifically to 224.0.0.6. The DR receives it, processes the update, and floods it out to 224.0.0.5 for the entire broadcast domain.

---

Multi-Area OSPF Architecture

While single-area OSPF works well for small networks, running all campus routers in a single flat area creates scaling bottlenecks in large environments:

• Every router must maintain the entire enterprise LSDB, consuming extensive RAM.
• Any link flap anywhere in the network causes every router to rerun the CPU-intensive Dijkstra SPF calculation.
• Route summarization cannot be performed within an area; it is only permitted at area boundaries.

To solve this, OSPF employs a two-tier hierarchical design:

+-----------------------------------------------------------------------------------+
|                         MULTI-AREA OSPF HIERARCHY                                 |
|                                                                                   |
|        +---------------------------------------------------------+                |
|        |                 AREA 0: BACKBONE AREA                   |                |
|        |         Core Switches (High-speed transit core)         |                |
|        +---------------------------------------------------------+                |
|                         /                       \                                 |
|                        /                         \                                |
|             +---------------------+   +---------------------+                     |
|             |     ABR Switch 1    |   |     ABR Switch 2    |                     |
|             +---------------------+   +---------------------+                     |
|                        |                         |                                |
|        +-------------------------+     +-------------------------+                |
|        |     AREA 1: CAMPUS A    |     |     AREA 2: CAMPUS B    |                |
|        | Access/Agg Layer 3 SVIs |     | Access/Agg Layer 3 SVIs |                |
|        +-------------------------+     +-------------------------+                |
+-----------------------------------------------------------------------------------+

Area Terminology and Rules

• Area 0 (Backbone Area): The central core of the OSPF network. All non-backbone areas must physically or logically connect directly to Area 0. All inter-area transit traffic must traverse Area 0.
• Non-Backbone Areas (e.g., Area 1, Area 2): Connect access and distribution layers. Routers within an area only run SPF for links inside their own area.

Router Classifications

In a multi-area topology, routers are categorized by their physical link placements:

1. Internal Router: A switch whose OSPF-enabled interfaces all belong to the same single area (e.g., an access switch residing entirely in Area 1).
2. Backbone Router: Any switch that maintains at least one active interface connected to Area 0.
3. Area Border Router (ABR): A router that attaches to two or more areas, with at least one interface in Area 0. ABRs maintain separate, independent LSDBs for each attached area, run SPF independently per area, and summarize/translate routing information between areas.
4. Autonomous System Boundary Router (ASBR): A router that connects the OSPF domain to an external routing source (such as static routes, BGP, or another IGP) and redistributes those external routes into OSPF.

---

Essential OSPF LSA Types

Topological information is partitioned into discrete Link-State Advertisement (LSA) types. An associate-level network engineer must master LSA Types 1, 2, and 3:

---

AOS-CX Multi-Area and DR Tuning Configuration

The following configuration demonstrates tuning DR election priorities on an access switch, configuring an Area Border Router (ABR) connecting Area 1 to Backbone Area 0, and optimizing point-to-point links:

! --- SWITCH A: Aggregation Switch (ABR Role) ---
switch-agg(config)# router ospf 1
switch-agg(config-ospf-1)# router-id 10.0.0.10
switch-agg(config-ospf-1)# area 0
switch-agg(config-ospf-1)# area 1
switch-agg(config-ospf-1)# exit

! Interface connecting to Core (Area 0)
switch-agg(config)# interface 1/1/48
switch-agg(config-if)# routing
switch-agg(config-if)# ip address 10.0.0.2/30
switch-agg(config-if)# ip ospf 1 area 0
switch-agg(config-if)# ip ospf network point-to-point
switch-agg(config-if)# exit

! Interface connecting to Campus Access Switch (Area 1, Force DR Role)
switch-agg(config)# interface 1/1/1
switch-agg(config-if)# routing
switch-agg(config-if)# ip address 10.1.0.1/24
switch-agg(config-if)# ip ospf 1 area 1
switch-agg(config-if)# ip ospf priority 255
switch-agg(config-if)# exit


! --- SWITCH B: Campus Access Switch (DROther Role in Area 1) ---
switch-acc(config)# router ospf 1
switch-acc(config-ospf-1)# router-id 10.1.0.20
switch-acc(config-ospf-1)# area 1
switch-acc(config-ospf-1)# exit

switch-acc(config)# interface 1/1/48
switch-acc(config-if)# routing
switch-acc(config-if)# ip address 10.1.0.2/24
switch-acc(config-if)# ip ospf 1 area 1
switch-acc(config-if)# ip ospf priority 0
switch-acc(config-if)# exit

Essential Verification Commands

---

Common Exam Traps

• Preemption Assumption: Candidates frequently assume that changing a switch's OSPF priority to 255 immediately forces it to become the DR on an active network. In reality, DR elections are non-preemptive; the current DR remains active until its interface goes down or its OSPF process is restarted.
• Priority 0 Behavior: Setting ip ospf priority 0 does not give the switch the lowest chance of winning; it makes the switch completely ineligible to ever become DR or BDR.
• LSA Type Scopes: Type 1 and Type 2 LSAs never cross an Area Border Router. Only Type 3 Summary LSAs travel between OSPF areas.
• Direct Area 0 Connectivity: In standard OSPF design, non-backbone areas cannot connect to each other directly; they must connect to Area 0. Traffic between Area 1 and Area 2 must flow through Area 0.