Configuring BFD-BGP4+ Interworking

BFD for BGP4+ speeds up fault detection and therefore increases the route convergence speed.

Prerequisites

Before you configure BFD-BGP4+ interworking, complete the following tasks on devices at both ends:

Setting the IP addresses of interfaces to ensure reachable adjacent nodes.
Configuring basic BGP4+ functions to enable neighbor relationship in Full state. For details, see Configuring BGP4+.
Configure global BFD functions, including enabling BFD and adjusting global parameters. For details, see Configuring Global BFD Functions.

Context

BFD can rapidly detect IPv6 forwarding failures. By adopting the BFD fast detection mechanism, an IPv6 network can transmit voice services, video services, and VoD services with high QoS. This enables service provides to provide their customers with highly available and reliable voice over IP (VoIP) and other real-time services.

BGP periodically sends Keepalive messages to the peer to detect faults on the neighbor. This mechanism, however, takes more than one second to detect a fault. When the data rate is up to Gbit/s, the detection mechanism causes a great packet loss. This mechanism fails to meet the requirement on the reliability of core networks.

BGP introduces BFD for BGP4+. The fast detection mechanism of BFD can faster detect faults on the links between BGP peers. The convergence of networks therefore speeds up.

Procedure

Access the system view.
```
system-view
```
Access the BGP view.
```
bgp { as-number-plain | as-number-dot }
```
(Optional) Access the BGP-VPN instance IPv6 address family view.
```
ipv6-family vpn-instance vpn-instance-name
```
BFD for BGP can be configured for the VPN in this view. To configure BFD for BGP for the public network, skip this step.
Enable BFD for the peer or peer group and create a BFD session using default parameters.
```
peer { group-name | ipv6-address } bfd enable [ single-hop-prefer ] 
```
single-hop-prefer takes effect only on IBGP peers. By default, if single-hop-prefer is not specified, multi-hop sessions are established between direct IBGP peers (Huawei devices). To interconnect a Huawei device and a non-Huawei device that defaults the sessions between IBGP peers to single-hop, configure single-hop-prefer in the command.

After BFD is enabled for a peer group, BFD sessions will be created on the peers that belong to this peer group and are not configured with the peer bfd block command.
(Optional) Modify BFD session parameters.
```
peer { group-name | ipv6-address } bfd { min-tx-interval min-tx-interval | min-rx-interval min-rx-interval | detect-multiplier multiplier | wtr wtr-value } ^*
```
The BFD parameters of peers take precedence over those of peer groups. If BFD parameters are configured on peers, they will be used in BFD session establishment.

The default interval for transmitting BFD packets and the default detection multiplier are recommended. When changing the default values, pay attention to the network status and the network reliability requirement. A short interval for transmitting BFD packets can be configured for a link that has a higher reliability requirement. A long interval for transmitting BFD packets can be configured for a link that has a lower reliability requirement.
There are three formulas: Actual interval for the local device to send BFD packets = max {Locally configured interval for transmitting BFD packets, Remotely configured interval for receiving BFD packets}, Actual interval for the local device to receive BFD packets = max {Remotely configured interval for transmitting BFD packets, Locally configured interval for receiving BFD packets}, and Local detection period = Actual interval for receiving BFD packets x Remotely configured BFD detection multiplier.

For example:
- On the local device, the configured interval for transmitting BFD packets is 200 ms, the interval for receiving BFD packets is 300 ms, and the detection multiplier is 4.
- On the peer device, the configured interval for transmitting BFD packets is 100 ms, the interval for receiving BFD packets is 600 ms, and the detection multiplier is 5.
Then:
- On the local device, the actual interval for transmitting BFD packets is 600 ms calculated by using the formula max {200 ms, 600 ms}; the interval for receiving BFD packets is 300 ms calculated by using the formula max {100 ms, 300 ms}; the detection period is 1500 ms calculated by multiplying 300 ms by 5.
- On the peer device, the actual interval for transmitting BFD packets is 300 ms calculated by using the formula max {100 ms, 300 ms}; the interval for receiving BFD packets is 600 ms calculated by using the formula max {200 ms, 600 ms}; the detection period is 2400 ms calculated by multiplying 600 ms by 4.
wtr wtr-value can be specified in the command to suppress frequent BFD and BGP session flapping caused by link flapping. If a BFD session over a link goes Down, it does not go Up immediately after the link recovers. Instead, the BFD session waits for the WTR timer to expire before going Up. If the link fails again before the WTR timer expires, BFD does not send a link fault message to BGP, and the BGP session status is stabilized.

The default value of wtr-value is 0, which means that the WTR timer will not be started.
(Optional) Disable the peer from inheriting the BFD function of the peer group to which it belongs.
```
peer ipv6-address bfd block
```
If a peer joins a peer group enabled with BFD, the peer inherits the BFD configuration of the group and creates a BFD session. To prevent the peer from inheriting the BFD function of the peer group, perform this step.

The peer bfd block command and the peer bfd enable command are mutually exclusive. After the peer bfd block command is run, the BFD session is automatically deleted.