Linux Operations - OSPF Networking Implementation Based on Bird for New OpenWrt

Introduction

When configuring studio/enterprise networks, it is often necessary to implement interconnection between multiple network segments. These segments are typically at the same level or different levels, not in a hierarchical relationship, so they cannot directly communicate with each other. In such cases, we need to configure OSPF on the routers. This article will teach you how to implement OSPF networking on routers running OpenWrt.

Example Environment

• Main Router OS: X-WRT 25.04_b202510240128 Plucky
• Secondary Router OS: X-WRT 26.04_b202601250827 Plucky

Following this guide requires you to have a basic understanding of the OSPF protocol

---

Getting Started

1. Install Bird

• Navigate to System --> Software page and install the two software packages as shown in the figure

Q. Why not choose Bird 3?
A. Bird 2 has been long-term verified by the community and is considered a stable and reliable choice, with a large number of actual configuration cases and troubleshooting experience. Moreover, Bird 2 almost meets the vast majority of network configuration requirements, without needing to use the new features of Bird 3.

2. Configure Bird

• Open the Bird configuration in init.d

vim /etc/init.d/bird

• Change BIRD_CONF="/etc/bird.conf to

BIRD_CONF="/etc/bird/bird.conf"

• Create configuration folder and main configuration file

mkdir /etc/bird
vim /etc/bird/bird.conf

• Fill in the configuration according to the following format

define INTRA_ROUTER_ID      = <your_router_id>;
define INTRA_PREFIX_V4      = [ <your_network_segments>/<netmask>+ ];
define INTRA_PREFIX_V6      = [ <your_network_segments>/<netmask>+ ];

protocol device {
    scan time 10;
};

ipv4 table intra_table_v4;
ipv6 table intra_table_v6;

include "config/defs.conf";
include "config/kernel.conf";
include "config/ospf.conf";

Configuration Item Explanation

• INTRA_ROUTER_ID: The Router-ID of the current device
• INTRA_PREFIX_V4: Macro defining IPv4 internal network prefix

• INTRA_PREFIX_v6: Macro defining IPv6 internal network prefix

  Prefix Naming Convention

  • define INTRA_... # Internal routing related
  • define INTER_... # External/boundary routing related
  • define CUSTOMER_... # Customer routing related
  • define PEERING_... # Peering routing related

• scan time 10: Interface scanning interval is 10s
• ipv4 table: IPv4 routing table name
• ipv6 table: IPv6 routing table name

• include: Specifies which configuration files to include, you can add your own configuration files. The following are recommended configuration files to add

  • defs.conf: Can store functions such as routing entry filtering rules
  • kernel.conf: Stores filtering rules to be written into the kernel routing table
  • ospf.conf: Configuration file for OSPF configuration

Example Configuration

define INTRA_ROUTER_ID      = 100.64.0.163;
define INTRA_PREFIX_V4      = [ 100.64.0.0/16+, 192.168.0.0/16+ ];
define INTRA_PREFIX_V6      = [ fd18:3e15:61d0::/48+ ];

protocol device {
    scan time 10;
};

ipv4 table intra_table_v4;
ipv6 table intra_table_v6;

include "intra/defs.conf";

include "intra/kernel.conf";
include "intra/ospf.conf";

• Create sub-configuration files and fill in the configuration according to the example format

The configuration files provided in this article are examples, it is highly recommended to write your own, you can use AI

mkdir infra
cd infra

• defs.conf

vim defs.conf

function is_intra_net4() {
    return net ~ INTRA_PREFIX_V4;
}

function is_intra_net6(){
    return net ~ INTRA_PREFIX_V6;
}

• kernel.conf

vim kernel.conf

protocol kernel intra_kernel_v4 {
    kernel table 254;
    scan time 20;
    ipv4 {
        table intra_table_v4;
        import none;
        export filter {
            if source = RTS_STATIC then reject;
            accept;
        };
    };
};

protocol kernel intra_kernel_v6 {
    kernel table 254;
    scan time 20;
    ipv6 {
        table intra_table_v6;
        import none;
        export filter {
            if source = RTS_STATIC then reject;
            accept;
        };
    };
};

• ospf.conf

vim ospf.conf

protocol ospf v3 intra_ospf_v4 {
    router id INTRA_ROUTER_ID;

    ipv4 {
        table intra_table_v4;
        import where is_intra_net4() && source != RTS_BGP;
        export where is_intra_net4() && source != RTS_BGP;
    };

    include "ospf/*";
};

protocol ospf v3 intra_ospf_v6 {
    router id INTRA_ROUTER_ID;

    ipv6 {
#       table dn42_table_v6;
        table intra_table_v6;
        import where is_intra_net6() && source != RTS_BGP;
        export where is_intra_net6() && source != RTS_BGP;
    };

    include "ospf/*";
};

Configure OSPF

• Create OSPF related directories and configuration files

mkdir intra/ospf
vim intra/ospf/0.conf

OSPF configuration files can be named arbitrarily, but it is recommended to follow the naming convention <AreaID>.conf

• Write the OSPF configuration file according to the following format

area 0.0.0.0 {
    interface "<lan_interface>" { stub; };
    interface "<ospf_interface>" {
        type <type>;
        cost <cost>;
        hello <time>;
        retransmit <time>;
        priority <num>;
        authentication <type>;
        ecmp weight <num>;
        bfd <yes/no>;
    };
};

Configuration Item Explanation
area: OSPF AreaID
interface: Interface-specific configuration

• stub interface is a stub interface, which only accepts routing information but does not perform OSPF-related activities
• Interfaces without any annotation are OSPF active interfaces, which will normally perform neighbor discovery and handshake

• type: Interface network type

  • broadcast (default/recommended): Broadcast network. Suitable for broadcast media such as Ethernet/Token Ring. This mode elects DR/BDR and supports multiple routers
  • pointopoint: Point-to-point network. Suitable for PPP, HDLC, point-to-point tunnels, does not elect DR/BDR, establishes neighbors directly
  • nonbroadcast: Non-Broadcast Multiple Access (NBMA). Suitable for Frame Relay, ATM, X.25, requires manual neighbor configuration
  • pointomultipoint: Point-to-multipoint. Suitable for some mesh connections, simulates NBMA as multiple point-to-point links
  • stub: Stub network, suitable for networks with only a single router, does not send Hello, does not form neighbors
• cost: Interface cost. For internal networks, you can directly set it to 1, for tunnel links such as VPN, it is recommended to set a slightly larger value, such as 100

• hello: Hello packet transmission interval, in seconds

  If the hello package transmission interval is different between neibours, then they won't be handshake successfully, so this value must be same between routers.

• retransmit: Retransmission interval, in seconds
• authentication: OSPF authentication type
• ecmp weight: ECMP weight
• bfd: Whether to enable BFD support

Example Configuration

area 0.0.0.0 {
    interface "br-lan" { stub; };
    interface "zta7oqfzy6" {
        type broadcast;
        cost 100;
        hello 20;
    };
};

• After completing all the above configurations, restart Bird

/etc/init.d/bird restart

Verification

• Use the following command to check the Bird status

birdc s p

• If you see the OSPF protocol status as Running, it indicates successful activation

intra_ospf_v4 OSPF       intra_table_v4 up     16:49:39.948  Running
intra_ospf_v6 OSPF       intra_table_v6 up     16:49:39.948  Running

---

Related Link

• [Fun Experiment] A LAN Spanning 20km: Seamlessly Merging Remote Networks on OpenWrt Using ZeroTier + OSPF

---

Conclusion

• I am not a professional in the networking field, if there are any errors, please point them out!
• If you find this article helpful, please share it with those who need it. Thank you!