What Is OSPF Network Types in Networking?
Also known as: OSPF network types, CCNP ENCOR OSPF, OSPF broadcast, OSPF NBMA, OSPF point to point
On This Page
Quick Definition
OSPF network types are settings that tell the OSPF routing protocol how to behave on a specific network link. They control how routers find each other, how often they say hello, and whether they need a special router to manage the network. Different network types are used for different kinds of connections, like a direct cable between two routers or a shared Ethernet network.
Must Know for Exams
OSPF network types are a heavily tested topic in the Cisco CCNP ENCOR (350-401) exam. The exam objectives explicitly cover OSPF operation, including the characteristics of different network types, how they affect neighbor adjacency formation, and the election of DR and BDR. Learners are expected to know the default network type for each interface type, the hello and dead timer values, and the neighbor discovery method. The exam often presents a scenario where a router is not forming an adjacency, and the candidate must identify that the network type mismatch is the cause. For example, one router is configured as broadcast and the other as point-to-point on the same link, causing the hello timers to be different and preventing adjacency.
In addition, the exam may ask about the effect of network type on LSA flooding and the size of the link-state database (LSDB). Candidates must understand that on a broadcast network, the DR generates a Type 2 Network LSA, while point-to-point networks use Type 1 Router LSAs. Questions may also cover the behavior of OSPF on loopback interfaces, which default to the loopback network type (treated as a stub host) unless changed. The exam also tests the configuration commands, such as ip ospf network point-to-point or neighbor x.x.x.x for NBMA. Beyond ENCOR, this knowledge is essential for the CCNP Enterprise concentration exams, such as ENSLD and ENARSI, where advanced OSPF features and complex network designs are tested. The Cisco certification blueprint considers network types as foundational to understanding OSPF scalability and convergence.
Simple Meaning
Think of OSPF network types as the rulebook for how routers talk to each other over a specific cable or connection. Imagine a neighborhood where houses are routers and the roads between them are network links. If the road is a narrow, private driveway connecting just two houses, the neighbors can talk directly to each other without any schedule.
This is like a point-to-point network type. But if the road is a busy street with many houses, like a cul-de-sac, they need a different approach. On a busy street, one house might be chosen as the mail carrier for the entire street, so that not every house has to talk to every other house individually.
This mail carrier is called a designated router. The network type tells the routers whether they should elect a designated router, how often they should check in with a hello message, and whether they need to send their updates to a special multicast address or to every router individually. On a point-to-point link, the two routers just send their updates directly to each other.
On a broadcast network like Ethernet, they elect a designated router to keep things efficient and send updates to a special group address that all OSPF routers listen to. On a non-broadcast network, like a frame relay cloud, the routers have to be manually told about their neighbors because the network itself doesnt allow automatic discovery. So, the network type is a crucial configuration that adapts OSPFs behavior to the physical and logical characteristics of the link.
Full Technical Definition
OSPF (Open Shortest Path First) is a link-state routing protocol that uses different network types to dictate how the protocol operates on a specific interface. The network type determines the method of neighbor discovery, the hello and dead timer intervals, whether a Designated Router (DR) and Backup Designated Router (BDR) are elected, and the method of sending Link-State Advertisements (LSAs). The five main OSPF network types are: Broadcast, Non-Broadcast Multi-Access (NBMA), Point-to-Point, Point-to-Multipoint, and Point-to-Multipoint Non-Broadcast.
Broadcast is the default for Ethernet interfaces. In this type, routers dynamically discover neighbors using OSPF hello packets sent to the multicast address 224.0.0.5. A DR and BDR are elected to reduce the number of adjacencies and LSA flooding. The DR becomes the central point for database synchronization, and all other routers form a full adjacency only with the DR and BDR. This type has hello timers set to 10 seconds and dead timers to 40 seconds by default.
NBMA (Non-Broadcast Multi-Access) is used on networks like Frame Relay or ATM that do not support multicast or broadcast. Neighbors must be manually configured using the neighbor command. DR and BDR are elected, but hello timers are 30 seconds and dead timers are 120 seconds. This network type requires careful configuration because the lack of broadcast capability prevents automatic neighbor discovery.
Point-to-Point is typically used on serial interfaces with PPP or HDLC encapsulation. It assumes only two routers on the link. No DR/BDR election occurs because there is no need for a central point on a two-router link. Hello timers are 10 seconds, dead timers are 40 seconds. This is the most efficient type for direct links because it reduces overhead and simplifies adjacency formation.
Point-to-Multipoint is used on NBMA networks where each router can communicate directly with any other router. It treats the network as a collection of point-to-point links. No DR/BDR election is required. Neighbors can be discovered dynamically if multicast is supported, or configured manually. Hello timers are 30 seconds and dead timers are 120 seconds. This type is suitable for partial-mesh or hub-and-spoke topologies.
Point-to-Multipoint Non-Broadcast is similar to Point-to-Multipoint but does not support multicast. Neighbors must be manually configured. No DR/BDR election occurs. This type is used when multicast is not available and the network is logically a point-to-multipoint topology. Understanding these types is critical for CCNP ENCOR, as misconfiguration can lead to neighbor adjacency failures, suboptimal routing, or routing loops.
Real-Life Example
Imagine a large office building with a central mailroom. The offices are routers, and the hallways are the network links. The building uses three different communication methods depending on the floor. On the first floor, there is a long, straight hallway with only two offices, one at each end. The two office managers can just walk directly to each others offices to exchange mail. This is like a point-to-point OSPF network type. No need for a central mailroom or a schedule; they just talk directly.
On the second floor, there is a large open area with many cubicles. Every morning, all the employees meet in the center of the floor and one person is elected the floor manager. That floor manager keeps a master list of all mail deliveries. If someone in a cubicle wants to send a message, they only need to tell the floor manager, and the manager passes it on to the correct cubicle. This is the broadcast network type with a designated router. The floor manager is the DR, and the employees are the other routers.
On the third floor, there are several closed rooms but the doors are fireproof and soundproof, so you cannot knock or call out. The office manager has to manually write down the names of all the people in the rooms and their locations. He then walks from room to room to deliver mail. This is like an NBMA network where neighbors must be manually configured because the network itself does not support automatic discovery or broadcast. The manager has to know exactly who is on the network and where to send updates. Each of these methods works well for the specific floor layout, and using the wrong method on the wrong floor would cause chaos.
Why This Term Matters
In real-world IT work, especially in enterprise networking, OSPF network types directly impact the stability, scalability, and performance of the routing infrastructure. A network engineer who configures an OSPF interface with the wrong network type can cause routers to never form adjacencies, leading to network outages or black holes where traffic cannot reach certain destinations. For example, if a point-to-point link is accidentally configured as broadcast, the routers will try to elect a DR and BDR, which is unnecessary and can cause a delay in convergence. Worse, if the network type is set to NBMA on a broadcast-capable Ethernet link, the routers will not automatically discover each other and the engineer must manually configure every neighbor, which is tedious and error-prone in a large environment.
Furthermore, OSPF network types affect how routing updates are flooded. On a broadcast network with a DR, the DR controls the database synchronization, which reduces the number of adjacencies from O(n^2) to O(n). This is critical in large data center or campus networks where hundreds of routers might be connected to the same VLAN. Without the broadcast network type, each router would have to form an adjacency with every other router, consuming memory and CPU. In service provider networks using Frame Relay or MPLS, the NBMA or point-to-multipoint types are common, and misconfiguration can prevent the entire network from converging. For cloud infrastructure, understanding how OSPF network types map to virtual network interfaces is essential for connecting on-premises networks to cloud VPCs using virtual private gateways. A network engineer troubleshooting a site-to-site VPN over a generic routing encapsulation tunnel must ensure the tunnel interface uses point-to-point or point-to-multipoint, as broadcast election on a tunnel can cause instability.
How It Appears in Exam Questions
In certification exams, OSPF network types appear primarily in scenario-based and troubleshooting questions. A typical question might describe a hub-and-spoke Frame Relay topology where the hub router is configured as broadcast but the spoke routers are configured as point-to-multipoint. The question asks why the spoke routers are not seeing routes from each other. The correct answer is that the network type mismatch prevents full adjacency, and the hub becomes the DR but the spokes do not form adjacencies with each other because the broadcast type expects a DR/BDR election, while the point-to-multipoint type does not elect a DR. Another common pattern is a simulation where a router is connected to an Ethernet switch, but the OSPF neighbor state is stuck in INIT or EXSTART. The candidate must examine the interface configuration and realize that the network type has been manually changed to NBMA, requiring the neighbor command, which is missing.
Configuration questions often ask the candidate to select the correct command to change the network type to point-to-point on a serial interface to avoid DR election. Multiple-choice questions may list the default timers for each network type, asking which combination (hello and dead timers) is correct for broadcast. Architecture questions might present a design with a large Layer 2 domain containing many routers and ask which OSPF network type minimizes the number of adjacencies and reduces convergence time. The answer is broadcast with DR/BDR election. Some questions ask about the effect of network type on the Type 2 LSA and whether a DR exists. For example, in a point-to-point network, there is no Type 2 LSA because no DR is elected. Every exam candidate must be comfortable mapping the network type to the underlying physical or logical medium and understanding how it changes OSPFs operational behavior.
Study encor
Test your understanding with exam-style practice questions.
Example Scenario
A company has three routers: Router A, Router B, and Router C. Router A is at the headquarters, while Router B and Router C are at branch offices. They are connected through a Frame Relay cloud, which is a non-broadcast multi-access network.
The network engineer configures OSPF on all three routers. On Router A, she sets the network type to broadcast, expecting automatic neighbor discovery. On Router B and Router C, she sets the network type to NBMA because she knows the cloud does not support broadcast.
Router B and Router C never form an adjacency with Router A. Why? Because Router A, using the broadcast network type, sends hello packets to the multicast address 224.0.0.5 and expects a DR election with 10-second hello timers.
Routers B and C, using NBMA, have 30-second hello timers and require the neighbor command. The mismatch in hello timers and the lack of interaction with the DR election process prevent the routers from ever reaching the FULL state. The engineer fixes the issue by configuring all three routers with the point-to-multipoint network type, which does not require a DR and works well over the non-broadcast cloud by treating each spoke link as a separate point-to-point connection.
Common Mistakes
Assuming all Ethernet interfaces automatically use broadcast network type.
While Ethernet defaults to broadcast, if the interface is configured as an OSPF passive interface or if the network type is manually changed, it will not behave as broadcast. Also, some virtual Ethernet interfaces may default to point-to-point.
Always verify the OSPF network type on the interface with the command show ip ospf interface, and do not assume the default. Read the documentation for non-standard interfaces.
Configuring OSPF on a serial point-to-point link with the broadcast network type to force DR election.
On a point-to-point link, DR election is unnecessary and wastes resources. The broadcast type also expects multicast capability, which may not be present on older serial links. It can cause adjacency formation to fail or be slow.
Use the default point-to-point network type on serial links with PPP or HDLC. Only use broadcast when there are more than two routers on the same Layer 2 segment.
Forgetting to configure the neighbor command on NBMA networks.
Without the neighbor command, routers on an NBMA network cannot discover each other, and no adjacencies will form. The routers will stay in the DOWN state.
After configuring the network type as NBMA, use the router ospf configuration mode and the neighbor command with the IP address of the neighbor for each remote router.
Assuming that point-to-multipoint and NBMA are the same.
Point-to-multipoint does not elect a DR or BDR, while NBMA does. Point-to-multipoint treats the network as a collection of point-to-point links, simplifying operations. They have different timer values and neighbor discovery methods.
Study the characteristics of each network type separately. Remember that point-to-multipoint is generally simpler and avoids the complexity of DR election on non-broadcast networks.
Changing the network type on a loopback interface to broadcast to advertise it as a network.
Loopback interfaces default to the loopback network type, which advertises the IP as a host route (/32). Changing it to broadcast can cause OSPF to try to form adjacencies over the loopback, which is not a physical link, leading to instability.
Leave loopback interfaces at their default network type. If you want to advertise the loopback network, use the network command under router ospf and let OSPF create a router LSA for it.
Exam Trap — Don't Get Fooled
In an exam question, they present a scenario where two routers are connected via a serial link with HDLC encapsulation, both using the default OSPF network type. They ask why the routers are not forming a full adjacency. The trap answer is that the network type must be changed to broadcast to force DR election.
Remember that serial links with HDLC or PPP default to the point-to-point network type, which does not require DR election. Adjacency formation on a point-to-point link is straightforward as long as the routers share the same subnet, are not in the same OSPF router ID, and have matching MTU. If adjacencies are not forming on a serial link, check for mismatched MTU, authentication, or IP addressing, not the network type.
Commonly Confused With
OSPF area types (such as standard, stub, totally stubby, NSSA) control the types of LSAs that are allowed within an area to reduce the routing table size. Network types control the behavior on individual interfaces, such as neighbor discovery and DR election. Area types affect the whole area, while network types affect a single link.
Changing the area type from standard to stub on a router does not change how that router discovers neighbors on an Ethernet link. The network type remains broadcast. The area type only filters the LSAs that the router accepts.
OSPF interface cost is a metric used to calculate the shortest path to a destination, and it is derived from the bandwidth of the interface. Network types do not affect cost calculation directly, though the network type can influence which interfaces are considered (e.g., loopback interfaces are always advertised as host routes regardless of cost).
Two routers on a Gigabit Ethernet link with a broadcast network type will have the same interface cost as if they were using point-to-point, because cost is based on bandwidth, not network type.
OSPF neighbor states (DOWN, INIT, 2-WAY, EXSTART, EXCHANGE, LOADING, FULL) describe the progress of adjacency formation. Network types influence which states are reached and how quickly. For example, on a broadcast network, routers reach the 2-WAY state after hello exchange, but only DR and BDR proceed to EXSTART. On a point-to-point link, routers quickly move to FULL without waiting for DR election.
On a broadcast network with three routers, Router A and Router C may remain in 2-WAY with each other because they are not the DR or BDR. On a point-to-point link, they would go to FULL directly.
Hello and dead intervals are timer values that determine how often a router sends hello packets and how long it waits before considering a neighbor dead. These timers are a direct consequence of the network type, but they are not the network type itself. Changing the timer values manually does not change the network type.
A broadcast network has default hello of 10 seconds and dead of 40 seconds. If an engineer changes the hello to 20 seconds, the network type remains broadcast, but the router may not form an adjacency with a neighbor that still uses 10 seconds.
Step-by-Step Breakdown
Interface Configuration and Encapsulation
The OSPF network type is associated with a specific interface. The underlying physical or logical medium (Ethernet, serial, tunnel) and the encapsulation (HDLC, PPP, Frame Relay) often determine the default network type. For example, an Ethernet interface defaults to broadcast, while a serial interface with HDLC defaults to point-to-point.
Default Network Type Assignment
When OSPF is enabled on an interface, it automatically assigns a default network type based on the interface hardware and encapsulation. Understanding the defaults is the first step. If the default is not suitable, the engineer must manually change it using the ip ospf network command.
Hello Packet Generation and Timer Configuration
The network type dictates the hello interval, dead interval, and the destination address of hello packets. On broadcast and point-to-point, hello is sent every 10 seconds to 224.0.0.5. On NBMA and point-to-multipoint, hello is sent every 30 seconds, and on multicast capable networks, to 224.0.0.5; on non-broadcast, it is unicast to configured neighbors.
Neighbor Discovery and Adjacency Formation
On broadcast and point-to-point, neighbors are discovered dynamically when hellos are received. On NBMA and point-to-multipoint non-broadcast, neighbors must be manually configured. The absence of automatic discovery on NBMA is a key difference that affects operations.
DR and BDR Election Process
Only broadcast and NBMA network types elect a DR and BDR. The election is based on the highest OSPF priority (default 1) and then the highest router ID. Point-to-point and point-to-multipoint types do not elect a DR, eliminating the overhead of the election process and the need for Type 2 LSAs.
LSA Flooding and Database Synchronization
On broadcast networks, the DR is responsible for flooding LSAs to all other routers. On point-to-point, each router floods its own LSAs directly to the single neighbor. Understanding this step is crucial for predicting network convergence behavior and bandwidth usage.
Verification and Troubleshooting
After configuration, verify the network type using show ip ospf interface. Check that timers match between neighbors. If adjacencies do not form, compare network types on both ends. Use debug ip ospf hello to see if hellos are being sent and received. A mismatch in network type is a common cause of adjacency failure.
Practical Mini-Lesson
OSPF network types are a fundamental configuration element that every networking professional must master. In practice, when you configure OSPF on a router, you must consider the interface type and the topology of the network. Begin by understanding the default network type for each interface. For standard Ethernet, the default is broadcast. For serial links using PPP or HDLC, the default is point-to-point. For Frame Relay, the default is typically NBMA, but this depends on the Cisco IOS version and how the Frame Relay map is configured. If you are building a point-to-point Frame Relay subinterface, the default may become point-to-point, which is simpler.
When configuring OSPF in an enterprise network, you will often need to change the network type on tunnel interfaces. For example, when building a DMVPN (Dynamic Multipoint VPN) tunnel, the tunnel interface should be configured as point-to-multipoint or point-to-point to avoid DR election, as the tunnel may have variable connectivity. Similarly, for a Generic Routing Encapsulation (GRE) tunnel between two routers, using point-to-point is standard and efficient. A common mistake is to leave the default on a tunnel interface, which is often NBMA, leading to complex DR election and neighbor configuration issues.
Professionals should also know that OSPF network types affect the LSA types generated. On a broadcast network, the DR generates a Type 2 Network LSA that lists all attached routers, reducing the amount of LSAs needed. On point-to-point, each router generates a Type 1 Router LSA with a point-to-point link description. This difference is important for large networks where reducing LSA flooding overhead is critical. Additionally, the network type influences how OSPF handles multicast versus unicast. On a non-broadcast network, you must use the neighbor command, and the router will send unicast hellos to the specified IP. This requires that the neighbor IP is reachable at Layer 2.
What can go wrong? The most common problem is a network type mismatch between two routers on the same link. If Router A is configured as broadcast (hello 10, dead 40) and Router B is configured as point-to-point (hello 10, dead 40), the timers actually match, but the DR election requirement on Router A causes it to expect a BDR, which Router B will never participate in. The adjacency may form but will never reach FULL, or it may form but the DR election will fail, leaving the link in a 2-WAY state. Another problem is using the wrong network type on a multi-access non-broadcast network. If you set a Frame Relay link to broadcast, the routers will try to multicast hellos, but the Frame Relay switch does not forward multicast, so no hellos are received, and no adjacencies form. The fix is to use NBMA or point-to-multipoint and configure neighbors manually.
To implement this in a lab, you can configure three routers connected to the same switch. Set the network type to broadcast on all three, observe the DR election, and then change one to point-to-point to see the behavior. This hands-on practice is essential for ENCOR exam preparation. Finally, remember that OSPF network types also interact with OSPF over MPLS and virtual environments. In cloud virtual networks, the underlying hypervisor may not support multicast, so you must explicitly configure the network type to match the environment. This practical knowledge separates a junior from a senior network engineer.
Memory Tip
For exam day, remember B-P-N as initials for the three main categories: Broadcast (with DR), Point-to-point (no DR), and NBMA (with DR but manual neighbors). Associate the timers: Broadcast and Point-to-point share 10s hello, while NBMA and Point-to-multipoint share 30s hello.
Covered in These Exams
Related Glossary Terms
802.1Q is the networking standard that allows multiple virtual LANs (VLANs) to share a single physical network link by tagging Ethernet frames with VLAN identification information.
802.1X is a network access control standard that authenticates devices before they are allowed to connect to a wired or wireless network.
5G is the fifth generation of cellular network technology, designed to deliver faster speeds, lower latency, and support for many more connected devices than previous generations.
Frequently Asked Questions
What is the default OSPF network type on an Ethernet interface?
The default OSPF network type on an Ethernet interface is broadcast. This means the interface uses multicast hello packets, elects a DR and BDR, and has default hello and dead timers of 10 and 40 seconds.
Do I need to configure the neighbor command on a point-to-point link?
No. On a point-to-point link, neighbors are discovered automatically through multicast hello packets. The neighbor command is only required on NBMA or point-to-multipoint non-broadcast network types.
What happens if two routers have different OSPF network types on the same link?
The routers will likely fail to form a full OSPF adjacency. The network type mismatch can cause differences in hello timers, multicast usage, and DR election expectations, leading to the neighbor state being stuck in INIT or EXSTART.
Which OSPF network type does not use a designated router?
Point-to-point and point-to-multipoint network types do not elect a designated router. They treat the link as a direct connection or a collection of direct connections, so no DR is needed.
Can I change the OSPF network type on a loopback interface?
Yes, but it is not recommended. The default loopback network type advertises the IP as a host route. Changing it to broadcast or point-to-point may cause OSPF to try to form adjacencies over the loopback, which is not a physical link, leading to instability.
What are the hello and dead timers for NBMA OSPF network type?
On an NBMA network type, the hello timer defaults to 30 seconds and the dead timer defaults to 120 seconds. These are longer than broadcast timers because NBMA networks often have slower or more complex connectivity.
Why would I use point-to-multipoint instead of NBMA?
Point-to-multipoint is simpler to configure because it does not require DR election or the manual neighbor command on each router if multicast is supported. It treats each remote router as a separate point-to-point link, making it ideal for hub-and-spoke topologies over non-broadcast clouds.
Does OSPF network type affect the routing table?
Indirectly, yes. The network type influences how LSAs are generated and flooded, which in turn affects the link-state database and the resulting routing table. For example, a broadcast network generates Type 2 Network LSAs, while point-to-point does not.
Summary
OSPF network types are a configuration setting that determines the operational behavior of the OSPF routing protocol on a given interface. They control everything from how routers discover each other and how often they exchange hello packets, to whether a designated router is elected and how routing updates are flooded. The five primary types are broadcast, NBMA, point-to-point, point-to-multipoint, and point-to-multipoint non-broadcast.
Each type is suited to a specific network medium, such as Ethernet, serial links, or Frame Relay clouds. For certification exams, especially the Cisco CCNP ENCOR, you must know the default network types for common interfaces, the hello and dead timer values, and which types require manual neighbor configuration. Common exam traps include assuming that a DR is always required, or that all Ethernet interfaces are always broadcast.
Understanding network types is not just about passing the exam; in real-world network engineering, getting this configuration right ensures that OSPF converges quickly, remains stable, and uses resources efficiently. Misconfiguration can lead to routing black holes and unnecessary support tickets. Remember that the network type is a per-interface parameter, and it must match between neighbors for adjacency to form.
By mastering this topic, you gain deeper insight into how OSPF adapts to different link-layer technologies, a skill essential for any enterprise networking professional.