The possible causes are as follows:

•  Configuration parameters such as the authentication mode and key, VRID, virtual address list, and version on devices in the VRRP group are different.

•  The link where heartbeat packets pass is unstable or disconnected.
•  The interface is blocked by loop prevention protocols such as STP and RRPP.
•  The CPU usage is high.

When devices in the VRRP group use the same priority:

•  If the VRRP master and slave are determined, the VRRP status is not negotiated again.
•  If two masters exist before devices in the VRRP group exchange heartbeat packets, the VRRP status is negotiated as follows:

        If the source IP address of a received heartbeat packet on the local end is larger than the primary IP address of the interface, the local device switches to the Backup state. Otherwise, the local end discards the heartbeat packet and retains in Master state.