What is VRRP; how does it provide VIP failover in TR7?
Short answer
VRRP (Virtual Router Redundancy Protocol, RFC 5798) is a redundancy protocol that shares a virtual IP address among more than one device: the Master device that carries the address at that moment sends advertisements at regular intervals; if the backup device does not receive the advertisements for about three intervals, it takes over the address. In TR7, cluster VIPs work with VRRP (VIP transition mechanisms). VRRP relies on the Interface IP pair selected in the VIP Communication field of the interface where the VIP is; for IPv6 VIPs, the VIP Communication (v6) pair is selected separately, and the VIPs on the interface do not become active until a pair is selected. While VIP Transition Method = Only VRRP (default), the advertisements are sent as unicast between the two addresses, and the interval is one third of the VIP Transition Duration (s) value (default 3, range 1–1000), rounded up; thus the Slave takes over the VIP when the advertisements have stopped for about this duration. In TR7 Link Check and in the methods that include GW check, the interval is 0.2 seconds, and the decision is additionally made according to the check results of the two devices.
Where It Is Used in TR7
| Where | Field / label | What for |
|---|---|---|
| Network → Interfaces → Edit Interface (in a cluster) | VIP Communication, VIP Communication (v6) (ip/cidr - peer ip/cidr pairs and None in the list) | Selecting the Interface IP pair over which the VRRP advertisements are sent |
| The same form (visible when a pair is selected) | VIP Transition Duration (s), VIP Transition Method | The failover time and the condition that triggers the failover |
| Network → Route Tables → route table preview (Preview) | Reload VIP Service, Restart VIP Service (per device in a cluster) | Reloading the VIP configuration or restarting the VIP service |
| Settings → General Settings → Notification Management | Notification Type = VIP status | Notification when the VIP moves to the Master or Slave device and when it is active on both devices at once |
| WEB-CLI | network cluster-ip-service status list, network cluster-ip-service restart (network cluster-ip-service) | Querying the status of the VIP service and restarting it |
VRRP in TR7
| Only VRRP | TR7 Link Check · TR7 GW Check · TR7 Link and GW Check | |
|---|---|---|
| Advertisement interval | VIP Transition Duration (s) ÷ 3, rounded up (default 3 → 1 second) | 0.2 seconds; VIP Transition Duration (s) is not taken into account |
| Sending the advertisement | Unicast between the two addresses of the VIP communication pair | The pair is a prerequisite for the VIP configuration to be generated |
| Failover decision | When the advertisements from the Master stop, the Slave takes over | Additionally, the result of the interface link and/or default gateway check on the two devices |
The virtual router ID of VRRP is not entered in the form; it is assigned by TR7 when the VIP communication pair is selected. Selecting the VIP role and the return: What is the Master/Slave role on a VIP? · When a failed device comes back, does the VIP automatically move back to it?
Common Misconceptions
- "VRRP advertisements are sent from the VIP address." VRRP relies on the Interface IP pair selected in VIP communication (in Only VRRP, the advertisements go as unicast between these two addresses); therefore the pair must have an address defined on each of the two devices.
- "The VIP works even if VIP communication is not selected." The VIPs on the interface do not become active until a pair is selected; an IPv4 pair does not activate IPv6 VIPs.
- "VIP transition duration (s) is the advertisement interval." In Only VRRP, the advertisement interval is one third of this value; the value determines the approximate time until the Slave takes over the VIP.
How it is done in the documentation
