As we all know, choosing the right routing protocol is crucial for achieving efficient data transmission and network resource utilization.

The Open Shortest Path First (OSPF) protocol, as an Interior Gateway Protocol (IGP), holds a significant position in modern network architectures due to its high flexibility, scalability, and reliability.

OSPF Protocol

OSPF uses a link-state algorithm (based on Dijkstra’s SPF algorithm) to quickly exchange routing information and update routing tables in large network environments, ensuring that data packets are forwarded along the optimal path.

We’ve covered various aspects of OSPF in previous discussions. Today, we’ll provide an overall in-depth explanation of the OSPF protocol. Make sure to bookmark this article for later reading.

Today’s Article Reading Bonus: “Classic OSPF Study Notes (62 Pages)”
Speaking of OSPF, I have a super classic study note to share with you. Send me a private message with the code “Study Notes” to receive a set of OSPF classic study notes, a treasure trove from veteran network engineers.

01 Overview of OSPF

Since its development in the late 1980s, the OSPF protocol has become a widely adopted dynamic routing protocol.

Originally designed to address issues in traditional distance-vector routing protocols (like RIP) in large-scale network environments, such as slow convergence speed and routing loops, OSPF uses a link-state routing algorithm, enabling it to respond more quickly to network changes and maintain consistent routing information.

Currently, there are two main versions of OSPF:

OSPFv2: Primarily used in IPv4 networks, defined in RFC 2328. It is the first widely deployed version of OSPF, supporting unicast and multicast traffic.

OSPFv3: Developed with the advent of IPv6, defined in RFC 5340. OSPFv3 not only supports IPv6 address families but also introduces improvements such as more flexible multicast support and enhanced security.

01 Comparison of OSPF with Other Routing Protocols
Compared to other dynamic routing protocols, OSPF has the following characteristics:

Loop-Free Routing: By using the link-state algorithm, OSPF can effectively avoid routing loop issues.

Fast Convergence: When the network changes, OSPF can quickly update routing information to ensure continuous network services.

Hierarchical Design: By dividing the network into different areas (Area), OSPF can reduce the size of the routing table, simplify routing calculations, and improve network manageability.

Flexible Metric Standards: OSPF allows administrators to configure different metric standards based on network needs, such as bandwidth, delay, etc., to achieve more intelligent routing selection.

02 Basic Concepts of OSPF

01 Area
OSPF organizes and manages the network by dividing it into multiple logical areas (Area).

An area is an important component of the OSPF network, used to segment the network into multiple logical groups to simplify routing management and reduce the number of LSAs.

Main area types include:

Backbone Area (Area 0): This is the central area to which all other non-backbone areas must connect. All areas must be directly or indirectly connected to the backbone area to ensure that routing information can be propagated throughout the OSPF domain.

Normal Area: These areas are directly connected to the backbone area and can contain routers, network interfaces, and subnets.

Stub Area: This type of area does not receive external autonomous system routing updates, which can reduce the size of the routing table. It is typically used at the edge of the network to reduce the impact on the core network.

Totally Stubby Area: This area does not accept any external autonomous system routing updates except for the default route.

NSSA (Not-So-Stubby Area): Similar to a stub area, but allows the injection of external autonomous system routing information while also blocking certain types of LSAs from entering the area.

02 Link-State Database (LSDB)
Each router running the OSPF protocol maintains a local Link-State Database (LSDB).

The LSDB contains all the state information of the network links known to the router, including adjacency relationships, network states, and metric values. The LSDB is populated through the exchange of Link-State Advertisements (LSAs) between routers. All routers will eventually have the same LSDB, ensuring routing consistency.

03 Routing Table
Based on the information in the LSDB, each router independently runs the Shortest Path First (SPF) algorithm to calculate the best path to each destination network and stores this information in the routing table.

The routing table determines how the router forwards data packets to the target network.

03 OSPF Network Types

The OSPF protocol supports multiple network types, each affecting how OSPF neighbor relationships are formed and how link-state information is propagated.

Main OSPF network types include:

01 Broadcast Multi-Access
In broadcast multi-access networks, all devices share the same broadcast domain. In such networks, OSPF elects a Designated Router (DR) and a Backup Designated Router (BDR).

The DR is responsible for collecting information from other routers in the area and disseminating this information to other routers. The BDR acts as a backup to the DR and takes over its duties if the DR fails.

02 Non-Broadcast Multi-Access (NBMA)
Non-broadcast multi-access networks also support communication between multiple devices, but they lack broadcast capabilities.

In NBMA networks, OSPF also needs to elect a DR and BDR. Unlike broadcast networks, devices in NBMA networks need to manually configure adjacency relationships rather than automatically forming them.

03 Point-to-Point (e.g., PPP links)
Point-to-point networks are those where two devices communicate directly without any other devices intervening. In this network type, there is no need to elect a DR or BDR because there are only two nodes, and they can directly form an adjacency relationship.

04 Point-to-Multipoint

Point-to-multipoint networks are a special type of network where a central node communicates with multiple end nodes. End nodes do not communicate directly with each other.

In this network, the central node is responsible for establishing adjacency relationships with the end nodes and maintaining the LSDB. Point-to-multipoint networks can be configured in normal mode or full mode (Full Point-to-Multipoint), the latter meaning that all end nodes will also form adjacency relationships.

05 Impact of Network Types on OSPF Operation

Different network types affect OSPF neighbor discovery mechanisms and the efficiency of link-state information propagation.

For example, in broadcast networks, the election of DR and BDR reduces LSA flooding across the entire network; in point-to-point networks, the absence of DR/BDR election processes makes the establishment of adjacency relationships simpler and faster.

04 DR & BDR Election
In the OSPF protocol, to improve network efficiency and reduce the number of link-state information transmissions, Designated Router (DR) and Backup Designated Router (BDR) need to be elected in broadcast multi-access and non-broadcast multi-access networks.

01 Role of DR
The main responsibilities of the Designated Router (DR) include:

Collecting link-state information from other routers on the same network segment.

Encapsulating the collected information into Link-State Advertisements (LSAs) and flooding them to other routers.

Reducing the number of LSA floods in the network, thereby reducing network load.

02 Role of BDR
The Backup Designated Router (BDR) acts as a backup to the DR, maintaining the same link-state information as the DR. Once the DR fails, the BDR can immediately take over the DR’s responsibilities, ensuring high network availability.

03 Election Process
The election of DR and BDR is conducted by sending Hello packets, following these steps:

Initial Phase: All routers start by default considering themselves as DR and BDR and begin sending Hello packets.

Router ID Comparison: Routers determine each other’s existence through information carried in Hello packets and decide which router will become DR and BDR based on the Router ID (a unique identifier, usually the highest IP address on the router).

Highest Router ID: The router with the highest Router ID becomes the DR, and the second highest becomes the BDR.

Priority Setting: Administrators can influence the election results by configuring the router’s priority. A priority of 0 means the router will not become DR or BDR.

Election Stability: Once the election is complete, the DR and BDR do not change unless the network conditions change (such as DR or BDR failure).

04 Importance of DR/BDR
The DR and BDR mechanism improves network stability and efficiency, mainly in:

Reducing the number of LSA transmissions, as the DR acts as a central point for information collection and dissemination.

Increasing system reliability, with the BDR as a backup to the DR, enabling seamless switching in case of DR failure.

Reducing network maintenance costs by minimizing unnecessary LSA exchanges.

05 OSPF Protocol Packet Types
The OSPF protocol uses five types of packets to achieve neighbor discovery, link-state information exchange, and acknowledgment functions.

01 Hello Packet
Hello packets are mainly used to discover neighbors and maintain neighbor relationships. They contain basic information about the sender, such as Router ID and priority.

Trigger Condition: Sent periodically (interval determined by Hello interval parameter).

Content:

Router ID

Hello/Dead interval

Options field (indicating supported features)

Neighbor list (list of known neighbor Router IDs)

Function: Helps routers discover neighbors and maintain neighbor relationships, serving as the basis for DR/BDR election.

02 Database Description (DD) Packet
DD packets are used to describe the contents of the sender’s Link-State Database (LSDB) to negotiate LSDB synchronization between both parties.

Trigger Condition: Sent during the process of establishing adjacency relationships between two routers.

Content:

Most recent LSA sequence number

Summary information of LSAs in the LSDB

Function: Helps routers identify the link-state information possessed by the other party and decide which LSAs need to be synchronized.

03 Link State Request (LSR) Packet
When a router lacks certain LSAs, it sends an LSR packet to request the missing LSAs.

Trigger Condition: Sent when discovering LSDB inconsistency.

Content:

List of missing LSA LSIDs

Function: Requests missing link-state information to achieve LSDB synchronization.

04 Link State Update (LSU) Packet
LSU packets contain complete LSA information, used to respond to LSR packets or actively update link-state information.

Trigger Condition: Sent when receiving an LSR packet or when link-state changes occur.

Content:

One or more complete LSAs

Function: Sends complete LSAs to update the receiver’s LSDB.

05 Link State Acknowledgment (LSAck) Packet
LSAck packets are used to acknowledge received LSU packets, ensuring reliable transmission of link-state information.

Trigger Condition: Sent after receiving an LSU packet.

Content:

List of received LSA LSIDs

Function: Confirms the receipt of LSU packets, ensuring accurate synchronization of link-state information.

06 OSPF Protocol State Machine
The OSPF protocol uses a state machine model to manage different stages of neighbor relationships.

The state machine model describes how routers establish and maintain relationships with neighboring routers through a series of state transitions. Below are the various states of the OSPF neighbor state machine and their transition processes:

01 Down State
This is the initial state of the neighbor relationship, indicating that the current router has not detected the neighbor’s existence.

Transition Condition: When the router receives the first Hello packet, the state transitions to Init state.

02 Init State
In this state, the router has received the neighbor’s Hello packet and recorded the neighbor’s existence. However, the neighbor has not listed the current router as a neighbor in its Hello packet.

Transition Condition: When the neighbor’s Hello packet includes the current router’s Router ID, the state transitions to Two-Way state.

03 Two-Way State
This state indicates that bidirectional communication has been established, i.e., both routers list each other as neighbors in their Hello packets. At this point, both parties can further negotiate whether to establish an adjacency relationship.

Transition Condition: If both parties agree to establish an adjacency relationship, the state transitions to ExStart state.

04 ExStart State
In this state, routers begin to negotiate the format of DD packets to determine how they will exchange the contents of their Link-State Databases (LSDB). This step is to determine who will send the DD packet first.

Transition Condition: When the negotiation is complete, the state transitions to Exchange state.

05 Exchange State
In this stage, routers begin to exchange DD packets to share the summary contents of their respective LSDBs. This process is to determine which party’s LSDB is more complete or up-to-date.

Transition Condition: When the DD packet exchange is complete, the state transitions to Loading state.

06 Loading State
In this state, routers send LSR packets to request missing LSAs based on the previous DD packet exchange results and send complete LSA information via LSU packets.

Transition Condition: When all missing LSAs have been exchanged and both parties’ LSDBs are synchronized, the state transitions to Full state.

07 Full State
This is the final state of the neighbor relationship, indicating that both routers have established a full adjacency relationship, their LSDBs are synchronized, and they are ready to jointly maintain network routing information.

08 Role of the State Machine
The OSPF state machine ensures the establishment and maintenance of neighbor relationships through orderly state transitions, thereby achieving efficient and reliable link-state information exchange.

Each state has a clear meaning and transition conditions, helping routers maintain good neighbor relationships in complex network environments and ensuring the consistency of routing information.