VGMP Group

VRRP Group Management Protocol (VGMP) is a Huawei proprietary protocol. The VGMP group concept is defined in the VGMP protocol. The FWs manage the active/standby status based on the VGMP group.

Each FW has a VGMP group, which cannot be deleted, and no other VGMP group can be created. A VGMP group has two attributes: priority and status. The VGMP group priority determines the VGMP group status.

The VGMP group priority is not user-configurable. After the device starts up, it automatically generates a VGMP group priority based on the hardware configuration of the device. This priority is called the initial priority. The initial priority is related to the number of CPUs on the device. Table 1 lists the initial priorities of devices of different models. When a fault occurs on a device, the priority of the VGMP group on the device decreases.

**Table 1** Initial priorities
Type	Model	Initial Priority
Single-CPU models	Except the USG6635E/6655E, USG6680E and USG6712E/6716E	45000
Dual-CPU models	USG6635E/6655E, USG6680E and USG6712E/6716E	45002

A VGMP group can be in one of the following states: initialize, load-balance, active, and standby. initialize is the initial state when hot standby is not enabled on the device. The other three states are determined by comparing the VGMP group priority of the device with that of the peer device. The device receives VGMP packets from the peer through the heartbeat link to learn the VGMP group priority of the peer.

If the VGMP group priority of the local device is the same as that of the peer, the VGMP group state of the device is load-balance.
If the priority of the VGMP group on the local device is higher than that on the peer, the VGMP group on the device is in the active state.
If the VGMP group priority of the local device is lower than that of the peer, the VGMP group state of the device is standby.
The VGMP group on the local device is in the active state if the local device does not receive VGMP packets from the peer and cannot learn the priority of the peer VGMP group due to faults, for example, a heartbeat link fault.

Hot standby requires that the two devices must be of the hardware model and have the same number and types of boards. Therefore, in normal cases, the VGMP groups of the two devices have the same priority, and the VGMP group is in the load-balance state. If a device is faulty, the priority of the VGMP group on the device decreases. The priority of the VGMP group on the faulty device is lower than that on the normal device. Therefore, the VGMP group becomes standby on the faulty device and active on the normal device.

The FWs can adjust the status of the VRRP group status, costs of dynamic routes (OSPF, OSPFv3, and BGP routes), VLAN status, and interface status (mirroring mode) based on the VGMP group status to implement active/standby or load balancing mode of hot standby. For details, see:

Checking the Priority and Status of a VGMP Group

Run the display hrp state command to check the VGMP group priority and status.

HRP_M<sysname> display hrp state
 Role: active, peer: standby
 Running priority: 45000, peer: 44998
 Backup channel usage: 0.00%
 Stable time: 0 days, 2 hours, 15 minutes
 Last state change information: 2017-09-22 14:31:24 HRP core state changed, old_state = normal, new_state = abnormal(active), local_priority = 45000, peer_priority = 44998.

**Table 2** Description of the **display hrp state** command output
Item	Description
Running priority: 45000, peer: 44998	Local and peer VGMP group priorities. Running priority indicates the priority of the local VGMP group. peer indicates the priority of the peer VGMP group.
Last state change information	Information about the last status switching of the local VGMP group. old_state is the historical status of the VGMP group. new_state is the current status of the VGMP group. The values of old_state and new_state may be one of the following: Initial: Indicates that the VGMP group is in the initialize state. Normal: Indicates that the VGMP group is in the load-balance state. Abnormal (standby): Indicates that the VGMP group is in the standby state. Abnormal (active): Indicates that the VGMP group is in the active state. As shown in the command output, the VGMP group status changes because the peer device is faulty and the priority of the VGMP group decreases. The priority of the local VGMP group is higher than that of the peer VGMP group. The status of the local VGMP group changes from load-balance to active.