# OSPF network type

> Source: Courseiva IT Certification Glossary — https://courseiva.com/glossary/ospf-network-type

## Quick definition

OSPF network type is a setting that tells the OSPF routing protocol how to behave on a network connection. It decides whether routers automatically find each other, whether they elect a special router to manage updates, and how they send routing information. Think of it as choosing between a small group discussion where everyone talks freely versus a large meeting where one person speaks for the group. The right choice depends on whether your network connects two routers directly, multiple routers on one wire, or virtual connections over the internet.

## Simple meaning

Imagine you are organizing a neighborhood watch group with several houses on different streets. If two houses are directly next to each other with no one else around, they can have a simple one-on-one conversation about suspicious activity. That is like a point-to-point OSPF network type, where two routers talk directly without needing a manager. Now imagine a block party where all 20 houses on the street need to share information. If everyone shouted at once, it would be chaos. So the group elects one person to be the block captain, who collects information from everyone and then announces it to all. That is like a broadcast OSPF network type, where routers elect a Designated Router (DR) and a Backup Designated Router (BDR) to manage updates efficiently. In some neighborhoods, the houses are spread out and cannot hear each other directly, so each house must individually knock on every other door. That is like a non-broadcast multi-access (NBMA) network, common in older Frame Relay setups, where every router must be manually told about its neighbors. There are also hybrid types like point-to-multipoint, where one router can talk to multiple others but without needing a central manager, similar to a teacher in a classroom who can address all students one at a time but without a student leader. The OSPF network type matters because it affects how quickly routers learn about changes, how much bandwidth they use, and how stable the network is. Choosing the wrong type can cause routers to fail to form adjacencies, waste bandwidth with unnecessary updates, or even prevent the network from converging after a failure. Network engineers set network types based on the underlying physical or virtual topology to ensure OSPF runs efficiently. For example, on an Ethernet link, the default is broadcast, but on a serial point-to-point link, it is point-to-point. The key is that OSPF adapts its behavior to the network type you configure, so understanding each type helps you design robust and scalable networks.

## Technical definition

OSPF network type is a per-interface attribute that dictates the operational behavior of the Open Shortest Path First (OSPF) routing protocol version 2 (IPv4) and version 3 (IPv6). It is defined in RFC 2328 for OSPFv2 and affects how Hello packets are sent, how neighbors are discovered, whether a Designated Router (DR) and Backup Designated Router (BDR) are elected, and how Link State Advertisements (LSAs) are flooded. There are five standard OSPF network types: broadcast, non-broadcast multi-access (NBMA), point-to-point, point-to-multipoint, and point-to-multipoint non-broadcast. Cisco IOS supports a virtual-link network type which is used for OSPF virtual links. The network type is configured at the interface level under OSPF, often using the command 'ip ospf network <type>' on Cisco routers, or equivalent configuration in other vendors like Juniper or Arista. In broadcast network type, the interface uses multicast to send Hellos to 224.0.0.5 (AllSPFRouters) and LSAs to 224.0.0.6 (AllDRouters). DR and BDR election occurs based on the highest OSPF priority (default 1) and then highest Router ID. This reduces LSA flooding overhead on multi-access segments such as Ethernet. Neighbors are discovered dynamically via multicast Hellos. Adjacencies form only between routers and the DR/BDR, not between all pairs, which is a key distinction. The NBMA network type is used on technologies like Frame Relay, ATM, or X.25 where multicast is not natively supported. Because these networks do not support broadcast, OSPF must use unicast to send Hellos and LSAs. Neighbors must be manually configured using the 'neighbor' command under the OSPF router process. DR/BDR election still occurs in NBMA networks, but since multicast is unavailable, all communication is unicast. This can lead to full mesh requirement for optimal performance. Point-to-point network type is designed for direct links between exactly two routers, such as serial HDLC or PPP links. No DR/BDR election takes place because there is only one other router. Hellos are multicast to 224.0.0.5, and adjacency forms quickly and simply. This type is efficient and fast to converge. Point-to-multipoint network type is used when multiple routers are connected over a single interface, like a hub-and-spoke Frame Relay or DMVPN configuration. OSPF treats each remote router as a point-to-point link, which means no DR/BDR is elected. Hellos are multicast or unicast depending on the sub-type. This type simplifies configuration in non-broadcast environments that are not fully meshed. Point-to-multipoint non-broadcast is similar but requires manual neighbor configuration because no multicast is available. The choice of OSPF network type directly impacts LSDB synchronization, convergence time, and scalability. For example, using broadcast on a serial point-to-point link does not cause errors but adds unnecessary DR election overhead and two additional seconds of wait time. Using point-to-point on an Ethernet segment with more than two routers would break neighbor relationships because OSPF would expect only one neighbor. Network engineers must match the network type to the actual Layer 2 topology. Modern implementations in data centers often use broadcast on VLANs, while WAN links use point-to-point or point-to-multipoint. Cloud environments such as AWS VPCs often use point-to-point for VPN tunnels or transit gateways. Understanding network types is crucial for passing certification exams like CCNA and Network+ and for real-world network troubleshooting. Misconfiguration of network type is a common cause of OSPF neighbor states stuck in INIT, EXSTART, or DOWN. OSPF network type influences the contents of Router LSAs (Type 1) and Network LSAs (Type 2). Broadcast and NBMA networks generate Type 2 LSAs from the DR, while point-to-point and point-to-multipoint networks use Type 1 LSAs with multiple stub links. This difference affects the OSPF database size and router memory consumption. OSPF network type is a fundamental concept linking Layer 2 topology to Layer 3 protocol behavior, and mastering it is essential for OSPF design and troubleshooting.

## Real-life example

Think of OSPF network types like different communication styles in a workplace or community. A broadcast network type is like an open-plan office with a team manager. Everyone can speak loudly and everyone hears, but to avoid chaos, the team elects a manager (DR) and an assistant manager (BDR). Team members talk to the manager, and the manager then announces updates to everyone. This is efficient when there are many people in one room, like many routers on an Ethernet segment. A point-to-point network type is like a private one-on-one meeting between exactly two people. There is no need for a manager because only two are involved. They can talk directly, quickly, and without extra overhead. This is like a direct serial link between two routers. An NBMA network type is like a group of offices in different buildings where each office has a phone but no conference call feature. To communicate, each person must individually call every other person. There is no way to broadcast a message to all at once. A manager is still elected to reduce the number of calls, but the manager must manually dial each person. This is like old Frame Relay networks where PVCs are used, and the router must know each neighbor's address. A point-to-multipoint network type is like a teacher in a classroom with students sitting in groups. The teacher can directly address each group individually, but groups do not talk to each other directly. No group leader is needed because the teacher handles everything. This is like a hub-and-spoke VPN where the hub router directly communicates with each spoke router, and the spokes do not directly talk to each other. The network type choice affects how much work the routers do. If you use broadcast on a serial link, it is like having a manager for just two people, which is wasteful. If you use point-to-point on an Ethernet with 50 routers, it is like trying to have private one-on-one meetings with everyone simultaneously, which is impossible because the medium expects a manager. Understanding this analogy helps you see why network engineers must choose the correct OSPF network type for each interface to ensure efficient and stable routing.

## Why it matters

OSPF network type matters because it directly controls the efficiency, stability, and scalability of OSPF routing in a network. Using the correct network type reduces unnecessary protocol overhead, speeds up convergence after a failure, and conserves router CPU and memory resources. For example, in a broadcast network with many routers, electing a DR and BDR cuts down the number of adjacencies from n*(n-1)/2 to just n-1, which is a massive reduction in a large Ethernet segment. This prevents flooding storms and keeps the network stable. In a point-to-point network, eliminating DR/BDR election speeds up adjacency formation because two routers can form a full adjacency in seconds without any waiting period. On NBMA networks, using the wrong network type can cause neighbors to never form because OSPF tries to multicast Hellos on a medium that does not support multicast, leading to routing black holes. Network type affects how OSPF handles redistribution and summarization, impacting route propagation. For IT professionals, especially those preparing for CCNA, Network+, or cloud certifications like AWS SAA and AZ-104, understanding network types is crucial for configuring OSPF in hybrid or cloud environments. For instance, in AWS, OSPF is often used with Dynamic Multipoint VPN (DMVPN) or transit gateway, where point-to-multipoint or point-to-point types are used. Misconfiguration can cause costly downtime or routing loops. Therefore, mastering OSPF network types is not just an exam topic but a practical skill for designing resilient networks.

## Why it matters in exams

OSPF network type is a high-weight topic in several major certification exams, especially CCNA (Cisco Certified Network Associate), Network+, and to a lesser extent, AWS Solutions Architect Associate (SAA) and Azure AZ-104. In CCNA (200-301), OSPF network types appear in the 'Configure, verify, and troubleshoot OSPF' objective. Questions often require you to identify which network type is appropriate for a given topology, such as a serial link (point-to-point) or an Ethernet LAN (broadcast). You may be asked to interpret 'show ip ospf interface' output to determine the current network type and the state of DR/BDR election. Troubleshooting scenarios where OSPF neighbors are stuck in INIT or DOWN often point to a mismatch in network type. For example, one router configured as broadcast while the other is point-to-point will prevent adjacency. In Network+ (N10-008), OSPF network types are part of the routing protocol knowledge under objective 1.3 (Explain routing technologies). While less detailed than CCNA, you need to know the basic types and their characteristics, especially broadcast and point-to-point. In AWS SAA, OSPF appears in the context of Site-to-Site VPN and Direct Connect, where BGP is more common, but OSPF can be used with Dynamic Multipoint VPN (DMVPN) or in advanced networking questions. Knowing that OSPF network type must match the VPN tunnel topology (like point-to-multipoint for hub-and-spoke) may be tested. In Azure AZ-104, OSPF appears in the context of Azure Virtual Network Gateway and ExpressRoute, but BGP is primary. However, understanding OSPF network types can help when troubleshooting hybrid connectivity. In Security+, OSPF network type is not a direct objective but supporting knowledge for network segmentation and routing security. For the Google ACE, OSPF appears in the networking section with less depth. Overall, CCNA and Network+ are the exams where you must be solid on this topic. Expect multiple-choice questions, scenario-based drag-and-drop, and simulation questions that require changing the network type to fix adjacency issues. Memorize the five types, their multicast/unicast behavior, and whether DR/BDR election occurs.

## How it appears in exam questions

Exam questions about OSPF network type typically fall into three patterns: scenario selection, configuration verification, and troubleshooting. In scenario selection questions, you are given a network diagram or description and asked which OSPF network type should be configured on a specific interface. For example, 'A router has two serial interfaces connected to two different remote sites over a Frame Relay network with partial mesh. Which OSPF network type is most appropriate?' The answer would be point-to-multipoint non-broadcast or point-to-multipoint, depending on the use of multicast. Similarly, 'An Ethernet segment has 10 routers. Which OSPF network type is the default and why?' expects broadcast. Configuration verification questions present output from 'show ip ospf interface' or 'show running-config' and ask you to identify the network type currently configured, or why a router is not becoming the DR. For instance, looking at 'show ip ospf interface FastEthernet0/0' you might see 'Network Type BROADCAST, Cost: 10' and then a question like 'What is the role of this router if its priority is 0?' The answer is that it will never become DR or BDR. Troubleshooting questions are common. You might be given a scenario where two directly connected routers on a serial link fail to form an OSPF adjacency, and the output shows they are both configured with broadcast network type. The problem is that broadcast type expects a DR/BDR election and does not form adjacency until after the election timer expires, but on a serial link the election never completes because there are only two routers. The fix is to change both to point-to-point. Another common trap is where one router is configured as point-to-point and the other as broadcast on the same link, causing adjacency to never form because the Hello packets are sent differently (multicast vs unicast). Questions about the order of DR/BDR election in broadcast networks and how priority affects it also appear. In advanced CCNA questions, you may be asked to calculate the number of OSPF adjacencies required in a broadcast network with n routers, which is 2n-3 (since every router has adjacency with DR and BDR, plus DR and BDR have an adjacency between them). For point-to-point networks, it is simply the number of links. These questions test your understanding of how network type scales.

## Example scenario

Consider a small company with two branch offices and a main office. The main office has an Ethernet switch with three routers connected to it, each serving a different department. At the main office, the OSPF network type on the Ethernet interfaces should be broadcast because it is a multi-access segment with multiple routers. This allows automatic DR/BDR election so that routing updates are efficient. Now consider the WAN links connecting each branch office to the main office. Each branch has a single router connected to the main office router via a point-to-point serial link. These links should be configured with OSPF network type point-to-point because there are only two routers on each link. Using broadcast on these serial links would cause unnecessary DR election delays and could prevent adjacencies from forming properly because serial links often don't have multicast support in the same way. Now imagine one of the branch offices is connected via a Frame Relay circuit that also connects to a third site in a hub-and-spoke topology. In this case, the main office is the hub, and the two branches are spokes. The hub router has one physical interface but multiple virtual circuits. The appropriate network type is point-to-multipoint, which treats each PVC as a point-to-point link without requiring a DR. This simplifies configuration and ensures all spoke routers can reach each other through the hub. The scenario demonstrates how different network types are chosen based on the physical topology and the number of routers per segment.

## OSPF Network Type Overview: What They Are and Why They Matter

OSPF (Open Shortest Path First) is a link-state routing protocol that relies on adjacencies and LSA (Link-State Advertisement) flooding to build a complete topology database. A critical but often misunderstood concept is the OSPF network type assigned to an interface. The network type determines how OSPF behaves on that interface, including how it discovers neighbors, whether it elects a Designated Router (DR) and Backup Designated Router (BDR), and whether it sends hello packets to multicast or unicast addresses. Understanding OSPF network types is essential for network engineers and is heavily tested in CCNA, Network+, and other vendor-neutral certifications.

There are five primary OSPF network types: Broadcast, Non-Broadcast, Point-to-Point, Point-to-Multipoint, and Point-to-Multipoint Non-Broadcast. Each type is designed for a specific Layer 2 technology. Broadcast is the default on Ethernet and other multi-access networks that support multicast. It requires DR/BDR elections and uses multicast addresses 224.0.0.5 (AllSPFRouters) and 224.0.0.6 (AllDRouters). Non-Broadcast is used on Frame Relay or ATM networks that do not support multicast; neighbors must be statically configured, and DR/BDR elections still apply. Point-to-Point is the default on serial interfaces using HDLC or PPP; it skips DR/BDR elections and uses multicast, making it faster and simpler. Point-to-Multipoint is used when a hub-and-spoke topology exists over a non-broadcast medium; it treats each link as a point-to-point link, eliminating DR/BDR elections. Point-to-Multipoint Non-Broadcast is similar but requires static neighbor configuration.

Configuration mismatches are a common source of OSPF adjacency failures. Two routers on the same segment must have compatible network types; otherwise, they may not form an adjacency. For example, if one end is configured as Broadcast and the other as Point-to-Point, the routers will disagree on neighbor discovery methods and DR election participation. Exam questions often test your ability to identify and fix these mismatches. The network type affects the OSPF cost calculation-on point-to-point links, cost is based solely on bandwidth, while on broadcast links, the cost can also be influenced by the DR/BDR role. This section establishes the foundational knowledge needed to understand all subsequent details about OSPF network types.

## OSPF Network Type DR/BDR Election: How It Works and Exam Traps

One of the most significant differences among OSPF network types is whether they elect a Designated Router (DR) and Backup Designated Router (BDR). On multi-access networks like Ethernet, electing a DR reduces the number of adjacencies from O(n^2) to O(n) and prevents LSA flooding storms. Only network types that are multi-access-Broadcast and Non-Broadcast-elect a DR and BDR. Point-to-Point and Point-to-Multipoint network types do not require DR/BDR elections because they are not multi-access; they have only two routers on the link, so a DR is unnecessary.

The DR election process is deterministic and based on priority and Router ID. The router with the highest OSPF interface priority (0-255, default 1) becomes the DR. If priorities are tied, the router with the highest Router ID wins. A priority of 0 means the router cannot become the DR or BDR. The election is not preemptive-once the DR and BDR are established, they remain until they fail, even if a higher-priority router joins later. This is a common exam trap: adding a new router with a higher priority does not cause a new election unless the current DR or BDR goes down.

Network type also determines when the election occurs. On Broadcast networks, the election happens during the Two-Way state when both routers see each other in their hello packets. On Non-Broadcast networks with statically configured neighbors, the election also occurs but may take longer due to manual neighbor specification. Exam questions often ask which network types require a DR, how priority affects election, and what happens when a new router with a higher priority appears. Understanding these details is critical for CCNA and Network+ exams. DR/BDR placement can have performance implications-the DR should be a robust router because it processes all LSAs from other routers in the segment.

## OSPF Network Type Adjacency Formation Rules and Timers

Each OSPF network type has specific rules for how neighbors are discovered and how adjacencies form. These rules directly impact the timers-hello interval and dead interval-and the neighbor discovery method (multicast, broadcast, or unicast). The default hello interval for Broadcast and Point-to-Point networks is 10 seconds, with a dead interval of 40 seconds. For Non-Broadcast and Point-to-Multipoint Non-Broadcast, the default hello interval is 30 seconds, with a dead interval of 120 seconds. Mismatched timers will prevent adjacency from forming, as routers expect consistent intervals within a subnet.

On Broadcast and Point-to-Point networks, OSPF uses multicast address 224.0.0.5 for all OSPF routers and 224.0.0.6 for DR/BDR communication. This allows automatic neighbor discovery-routers dynamically detect each other through hello packets. On Non-Broadcast networks, multicast is unavailable, so neighbors must be manually configured using the neighbor command. This manual configuration makes Non-Broadcast network types less dynamic and more error-prone. Exam questions often include scenarios where an adjacency fails because one router is configured with a Non-Broadcast network type but no neighbor statements are added.

Point-to-Multipoint network types are unique because they treat each spoke as a separate point-to-point link. Adjacencies are formed independently with each neighbor, and no DR/BDR is needed. On Point-to-Multipoint Non-Broadcast, static neighbor configuration is required, but the adjacency still behaves as point-to-point once established. The network type also affects LSA flooding-on broadcast networks, only the DR floods LSAs to other routers, reducing overhead. On point-to-point networks, each router directly floods LSAs to its single neighbor. Understanding these adjacency rules helps troubleshoot issues like one-way or two-way adjacencies that never progress to Full state, which is a common exam focus.

## OSPF Network Type Cost Impact: How Network Type Affects Path Selection

The OSPF cost metric is inversely proportional to bandwidth, but the network type can influence how cost is calculated and applied. By default, OSPF uses the formula cost = reference bandwidth / interface bandwidth. The reference bandwidth defaults to 100 Mbps, which can lead to equal-cost paths on high-speed interfaces if not adjusted. However, the network type interacts with cost through the DR/BDR election and the handling of virtual links.

On broadcast and non-broadcast multi-access networks, the DR becomes a central point for LSA exchange. The cost to reach the DR is not inherently different from other routers, but the DR's role adds processing overhead. When a virtual link is configured to traverse a non-backbone area, the cost of that virtual link is derived from the cost of the underlying interface, which depends on its network type. Point-to-point links are generally preferred for virtual links because they have lower overhead and deterministic costs.

Network type mismatches can also lead to suboptimal routing. For example, if a link is configured as Point-to-Multipoint but the physical topology is a full mesh, OSPF may generate multiple host routes (each /32) instead of a single network route, increasing the routing table size and potentially affecting convergence. Conversely, using Broadcast on a point-to-point link wastes resources on DR/BDR election. Exam questions frequently test whether you can identify the cost implications of network types in multi-vendor or multi-topology environments. Adjusting the reference bandwidth or manually setting the cost with the ip ospf cost command is a common solution. Understanding this relationship helps network engineers design efficient OSPF networks and avoid metric-induced routing loops or suboptimal path selection.

## Common mistakes

- **Mistake:** Assuming all network types support multicast
  - Why it is wrong: NBMA and point-to-multipoint non-broadcast do not support multicast by default, so Hello and LSA packets must be sent via unicast or manually configured neighbors
  - Fix: Check the underlying Layer 2 technology; use broadcast or point-to-point only when multicast is available, otherwise use NBMA or point-to-multipoint non-broadcast
- **Mistake:** Configuring broadcast network type on a point-to-point serial link
  - Why it is wrong: Broadcast type invokes DR/BDR election on a link where only two routers exist, causing unnecessary overhead and delaying adjacency formation
  - Fix: Use point-to-point network type on serial links (HDLC, PPP) for faster convergence and simplicity
- **Mistake:** Forgetting to configure static neighbors on NBMA networks
  - Why it is wrong: Without the 'neighbor' command under OSPF, the router will not discover neighbors because multicast Hellos are not sent, so no adjacency forms
  - Fix: Add 'neighbor <ip-address>' for each directly connected neighbor on the NBMA interface
- **Mistake:** Mismatching network type between OSPF neighbors on the same link
  - Why it is wrong: If one router is configured as broadcast and the other as point-to-point, they will not form an adjacency because the Hello packets have different format and multicast expectations
  - Fix: Ensure both ends of a link are configured with the same OSPF network type
- **Mistake:** Using point-to-multipoint on a fully meshed network
  - Why it is wrong: Point-to-multipoint treats each neighbor as a point-to-point link, which is inefficient on a full mesh where broadcast type would be more scalable with DR election
  - Fix: Use broadcast or NBMA (with DR election) for full-mesh topologies to reduce adjacencies
- **Mistake:** Setting OSPF priority to 0 on all routers, expecting a DR election to happen
  - Why it is wrong: Priority 0 means the router cannot be elected DR or BDR; if all routers have priority 0, no DR/BDR is elected and no adjacency forms in broadcast or NBMA networks
  - Fix: Ensure at least one router has a priority greater than 0 (default is 1) to allow DR/BDR election

## Exam trap

{"trap":"In a broadcast OSPF network, if all routers have a priority of 0, what happens to neighbor relationships?","why_learners_choose_it":"Learners often think that OSPF will use some other mechanism to form adjacencies, or that DR still gets elected based on Router ID alone","how_to_avoid_it":"Remember that a priority of 0 means the router cannot become DR or BDR. In broadcast networks, every router must have adjacency with the DR and BDR. If no DR/BDR is elected, no full adjacency can form and the network will not converge. The correct answer is that OSPF routers will remain in 2-WAY state and will not exchange LSAs."}

## Commonly confused with

- **OSPF network type vs OSPF area type:** OSPF area type (such as standard, stub, NSSA, totally stubby) focuses on how LSAs are filtered within an area, while network type focuses on how OSPF operates on a specific interface. Area type affects route summarization and LSA propagation, while network type affects neighbor discovery and DR election. (Example: A broadcast network type can be configured inside a stub area; they are independent concepts.)
- **OSPF network type vs OSPF interface cost:** OSPF interface cost is a metric used to calculate the OSPF path cost based on bandwidth, while network type defines the protocol behavior on that interface. Cost influences route selection, but network type influences adjacency formation. (Example: Two routers on a broadcast network type may have different interface costs, but they still use multicast to discover each other.)
- **OSPF network type vs OSPF router ID:** Router ID is a 32-bit identifier that uniquely identifies a router in the OSPF domain, while network type determines how that router interacts on a specific link. Router ID is used in DR election if priority is tied, but network type decides whether DR election happens at all. (Example: A router with Router ID 1.1.1.1 on a point-to-point link does not need to be elected as DR.)
- **OSPF network type vs EIGRP network type:** EIGRP also has different interface types (broadcast, NBMA, point-to-point) but the behavior is different. EIGRP uses multicast for updates but does not use DR/BDR election; it uses a different neighbor discovery mechanism. OSPF network types directly affect LSA flooding and DR role. (Example: In an EIGRP network over Frame Relay, you use 'no ip split-horizon' on the hub, while in OSPF you would configure point-to-multipoint.)

## Step-by-step breakdown

1. **Determine the network medium** — Before choosing an OSPF network type, identify whether the underlying Layer 2 technology supports broadcast/multicast (like Ethernet), is point-to-point (like serial HDLC), or is non-broadcast (like Frame Relay). This physical reality dictates which OSPF network types are valid.
2. **Select the appropriate OSPF network type** — Based on the medium, choose from broadcast (Ethernet with multiple routers), point-to-point (serial links with two routers only), NBMA (non-broadcast multi-access like Frame Relay mesh), point-to-multipoint (hub-and-spoke with multicast), or point-to-multipoint non-broadcast (hub-and-spoke without multicast).
3. **Configure the network type on the interface** — Use the interface configuration command (e.g., 'ip ospf network point-to-point' on Cisco). Ensure both ends of a link are configured with the same network type, otherwise adjacency will not form. Verify with 'show ip ospf interface'.
4. **Optional: Configure neighbor statements** — If using NBMA or point-to-multipoint non-broadcast, you must manually specify neighbors using the 'neighbor' command under the OSPF router process. This tells the router the IP addresses of its neighbors because multicast is not available.
5. **Let Hello packets be exchanged** — Once configured, OSPF begins sending Hello packets based on the network type. In broadcast and point-to-point, these are multicast to 224.0.0.5. In NBMA, they are unicast to configured neighbors. The Hello Interval is 10 seconds (default) for broadcast/point-to-point, 30 seconds for NBMA.
6. **DR/BDR election (if applicable)** — In broadcast and NBMA network types, an election determines which router becomes Designated Router (DR) and Backup Designated Router (BDR). Only the DR and BDR form full adjacencies with all other routers. Other routers remain in 2-WAY state between themselves. Point-to-point and point-to-multipoint skip this step.
7. **Form adjacency and synchronize LSDB** — Routers exchange Link State Advertisements (LSAs) to build identical link-state databases. The mechanism of flooding differs: in broadcast, LSAs are sent to 224.0.0.6 (AllDRouters); in point-to-point, they are sent to 224.0.0.5. The DR manages the flooding in broadcast networks.
8. **Convergence and route calculation** — Each router runs the SPF algorithm to calculate the shortest path to each destination. The convergence time is influenced by network type: point-to-point converges fastest because there is no DR election; broadcast networks may have a slight delay due to election timers.
9. **Verification and troubleshooting** — Use commands like 'show ip ospf neighbor', 'show ip ospf interface', and 'debug ip ospf adj' to verify correct network type and neighbor states. Common issues include mismatched types, missing neighbor statements, or DR election problems.

## Practical mini-lesson

In practice, configuring OSPF network type is one of the first steps in enabling OSPF on a router interface. Professionals must always verify the Layer 2 technology before deciding. For example, on a Cisco router with a serial interface using PPP, the default OSPF network type is point-to-point, which is ideal. However, if the same serial interface uses Frame Relay, the default becomes NBMA, which often requires manual neighbor statements. One common mistake in real environments is that engineers forget to change the network type when migrating from Frame Relay to Ethernet backhaul, leading to adjacency failures. On modern DMVPN networks, the tunnel interface (multipoint GRE) typically uses broadcast or point-to-multipoint network type. For point-to-point GRE tunnels, the network type is usually set to point-to-point by default, but you might manually change it to broadcast for some design reasons, such as reducing routing protocol overhead. In cloud environments, such as AWS VPN connections, OSPF is sometimes used between a virtual private gateway and on-premises router. These connections are typically point-to-point, so the network type should reflect that. However, because AWS uses BGP as the preferred dynamic routing protocol, OSPF is less common. Still, for hybrid networks using OSPF, you must configure the correct network type on both ends. Another practical scenario is in large data centers with redundant spine-leaf architecture. OSPF is often used as an IGP, and the interfaces between spines and leaves are typically point-to-point links or VLANs. The network type on these interfaces is usually set to point-to-point for fast convergence, even if the physical medium is Ethernet. This is a common optimization because it eliminates DR/BDR election, which reduces the number of adjacencies and speeds up failover times. To configure point-to-point on an Ethernet interface, you would use 'ip ospf network point-to-point'. This is a non-standard but widely used practice in data center networks. The downside is that you lose the ability to have more than two routers on the segment, but in a spine-leaf design, each link has only two endpoints. Therefore, it's a perfect fit. Engineers must also be aware that changing network type resets OSPF adjacencies on that interface, so it should be done during maintenance windows. Practical knowledge of OSPF network types allows you to design scalable, fast-converging networks and to troubleshoot neighbor issues efficiently.

## Commands

```
ip ospf network point-to-point
```
Changes the OSPF network type on an interface to point-to-point. Use this on serial links or Ethernet links where only two routers are connected to suppress DR/BDR election and speed up adjacency.

*Exam note: Exams test that point-to-point network type disables DR/BDR election and uses multicast 224.0.0.5, reducing convergence time.*

```
ip ospf network broadcast
```
Sets the OSPF network type to broadcast on an interface. This is the default on Ethernet interfaces. Use it when you need DR/BDR election on a multi-access network.

*Exam note: Exams may ask why broadcast is default on Ethernet and how it affects neighbor discovery via multicast vs unicast.*

```
ip ospf network non-broadcast
```
Configures the OSPF network type to non-broadcast on an interface. Typically used on Frame Relay or ATM interfaces where multicast is not supported.

*Exam note: Key exam point: non-broadcast requires static neighbor configuration with the 'neighbor' command, or adjacency will not form.*

```
neighbor 10.0.0.2 priority 1
```
Manually specifies an OSPF neighbor on a non-broadcast network. Required when the interface is configured with 'ip ospf network non-broadcast' or 'point-to-multipoint non-broadcast'.

*Exam note: Exams test that this command is mandatory for non-broadcast networks and that the priority influences DR/BDR election.*

```
ip ospf network point-to-multipoint
```
Configures the interface as point-to-multipoint. Use this on hub-and-spoke topologies over a non-broadcast medium. It treats each spoke as a point-to-point link.

*Exam note: Exams emphasize that point-to-multipoint eliminates DR/BDR and automatically creates host routes for each neighbor.*

```
ip ospf network point-to-multipoint non-broadcast
```
Sets the network type to point-to-multipoint non-broadcast. Combines features of point-to-multipoint with static neighbor configuration.

*Exam note: Commonly tested in scenarios involving Frame Relay hub-and-spoke where multicast is not available.*

```
show ip ospf interface [interface]
```
Displays OSPF parameters on an interface, including the network type, timers, neighbor count, and DR/BDR information.

*Exam note: Essential troubleshooting command. Exams test that you can verify network type and timer consistency to debug adjacency issues.*

```
ip ospf hello-interval 5
```
Changes the hello interval on an OSPF interface. Must match between neighbors for adjacency to form. Not directly setting network type, but often used with point-to-point for faster convergence.

*Exam note: Exams test that mismatched hello/dead intervals prevent adjacency, regardless of matching network types.*

## Troubleshooting clues

- **Mismatched OSPF Network Types Between Routers** — symptom: OSPF adjacency stuck in INIT state or never forms; 'show ip ospf neighbor' shows no neighbors on one side.. When routers have different OSPF network types on the same link (e.g., one Broadcast, one Point-to-Point), they disagree on multicast vs unicast neighbor discovery and DR election behavior, preventing the two-way state. (Exam clue: Exams present a topology where one interface is set to 'point-to-point' and the other to 'broadcast'; you must identify that this mismatch causes adjacency failure and correct it by aligning the network types.)
- **Neighbor Not Forming on Non-Broadcast Network** — symptom: OSPF adjacency remains in DOWN state; 'debug ip ospf adj' shows no hello packets received from neighbors.. On non-broadcast network types, multicast is disabled, so OSPF cannot discover neighbors dynamically. The 'neighbor' command must be configured manually on both sides for adjacencies to form. (Exam clue: Exams test that forgetting the 'neighbor' command on a Frame Relay interface results in no adjacency. You must know to add a static neighbor statement.)
- **DR/BDR Election Not Happening as Expected** — symptom: A router with higher priority or higher Router ID does not become DR after being added to the network; the existing DR remains.. OSPF DR/BDR election is non-preemptive. Once the DR and BDR are elected, they stay until they fail. A new router with a higher priority will only become DR or BDR if the current DR or BDR goes down. (Exam clue: Exams often ask: 'If you add a router with priority 255 to an existing OSPF network, does it become the new DR?' Answer: No, unless the current DR fails.)
- **Point-to-Multipoint Generating /32 Host Routes Instead of Network Routes** — symptom: Routing table shows multiple /32 host routes for neighboring routers instead of a single /30 or /24 network route.. On point-to-multipoint network types, OSPF treats each neighbor as a separate point-to-point link, assigning a /32 host route to the neighbor's interface IP. This increases routing table size. (Exam clue: Exams test that point-to-multipoint creates host routes and that changing to broadcast may suppress this, but DR/BDR election introduces new overhead.)
- **OSPF Adjacency Forms but Does Not Reach FULL State** — symptom: Neighbor state is stuck at TWO-WAY or EXSTART/EXCHANGE and never reaches FULL on a multi-access network.. On broadcast networks, not all neighbors form full adjacencies. Routers that are not DR or BDR will remain in TWO-WAY state with each other to reduce LSDB synchronization. Only DR and BDR form full adjacencies with all routers. (Exam clue: Exams ask: 'Why are some OSPF neighbors stuck in TWO-WAY?' The answer is that only DR/BDR require full adjacency; DROTHER routers stay in two-way.)
- **Timers Mismatch Preventing Adjacency** — symptom: 'show ip ospf neighbor' shows no neighbors; OSPF hello packets are sent but not matched; 'show ip ospf interface' shows different hello/dead intervals.. OSPF requires that hello and dead intervals match between neighbors for adjacency to form. Even if network types match, different timer values cause routers to consider each other as ineligible. (Exam clue: Exams combine timer mismatches with network type questions; you must identify that both parameters need to align.)
- **High CPU on DR Router in Large Broadcast Network** — symptom: Router acting as DR shows high CPU utilization during LSA flooding or topology changes.. On broadcast network types, the DR is responsible for flooding LSAs to all other routers in the segment. In large segments with many routers, this can become CPU-intensive, especially during convergence events. (Exam clue: Exams may ask how to reduce DR load (e.g., changing to point-to-multipoint or adjusting reference bandwidth).)

## Memory tip

Remember 'P-P-B-N-M-P' for the five OSPF network types: Point-to-point, Point-to-multipoint, Broadcast, NBMA, Point-to-multipoint Non-broadcast. Or use the phrase 'Peter Picks Big Nasty Mangoes' to recall them in order of simplicity.

## FAQ

**What is the default OSPF network type on an Ethernet interface?**

The default OSPF network type on an Ethernet interface is broadcast because Ethernet supports multicast and can have multiple routers.

**Can I change the OSPF network type on a running interface without resetting OSPF?**

Changing the OSPF network type will reset all adjacencies on that interface, causing temporary routing disruption. It should be done during a maintenance window.

**What happens if two OSPF routers have different network types on the same link?**

They will not form an adjacency because Hello packets are sent differently (multicast vs unicast) and the neighbor discovery mechanism differs.

**Why does point-to-point network type converge faster than broadcast?**

Point-to-point does not require DR/BDR election, so adjacency forms immediately after Hello exchange, reducing convergence time by several seconds.

**Is it possible to use point-to-point network type on an Ethernet interface?**

Yes, you can manually configure 'ip ospf network point-to-point' on an Ethernet interface. This is common in data center spine-leaf designs where each link has only two routers.

**What is the difference between point-to-multipoint and point-to-multipoint non-broadcast?**

Point-to-multipoint uses multicast to send Hellos, so neighbors are discovered dynamically. Point-to-multipoint non-broadcast requires manual neighbor configuration because multicast is not available.

**How many OSPF adjacencies exist in a broadcast network with 5 routers?**

In a broadcast network, each router forms a full adjacency with the DR and BDR (2 each), and the DR and BDR have one adjacency between them, totaling (5*2) - 1 = 9 adjacencies. Alternatively, formula 2n-3 = 7? Actually, for n=5, 2*5-3 = 7 adjacencies? Let's recalc: Each router (including DR and BDR) has adjacency with DR and BDR. So 5 routers each have 2 adjacencies = 10, but each adjacency counted twice, so 10/2 = 5 adjacencies? No, DR has adjacency with all 4 others (including BDR) = 4, BDR has adjacency with all 4 others = 4, but DR-BDR counted twice, so total = 4+4-1 = 7. Yes, 7 adjacencies.

---

Practice questions and the full interactive page: https://courseiva.com/glossary/ospf-network-type