First-Hop Redundancy Protocols (FHRP)

---

• end systems are typically served by a single router configured as the default gateway
  • possible to configure hosts with multiple default gateways for fault tolerance
    • does not work well in practice
    • requires a greater degree of complexity in the hosts’ routing algorithms than is typically implemented on an end system host
• To address this problem, various types of first hop redundancy protocols (FHRP) have been developed

First hop redundancy protocols (FHRP) provision failover routers to serve as the default gateway for a subnet.

• two kinds:
  • Virtual Router Redundancy Protocol (VRRP)
  • Hot Standby Router Protocol (HSRP)

Hot Standby Router Protocol (HSRP)

Hot Standby Router Protocol (HSRP) allows multiple physical routers to serve as a single default gateway for a subnet.

• proprietary, developed by Cisco
• how it works:
  • each router must have an interface connected to the subnet
    • with its own unique MAC address and IP address
    • need to be configured to share a common:
      • virtual IP address
      • MAC address
    • group of routers configured in this way is known as a standby group
      • communicate among themselves using IP multicasts
      • choose an active router based on priorities configured by an administrator
  • The active router responds to any traffic sent to the virtual IP address
  • Of the remaining routers in the standby group,
    • router with the next highest priority is chosen as the standby router
      • monitors the status of the active router
      • takes over the role if the active router becomes unavailable
        • triggers selection of a new standby router from the remaining routers in the group

Tip

Cisco has the Gateway Load Balancing Protocol (GLBP) which allows for an active/active load-balanced configuration.

Virtual Router Redundancy Protocol (VRRP)

Virtual Router Redundancy Protocol (VRRP) is an open standard protocol similar to HSRP.

• differences:
  • terminology
  • packet formats
• active router is known as the master
• all other routers in a group are called backup routers
• no specific standby router
  • instead, all backup routers monitor the status of the master
• in event of failure, a new master is selected from the available backup routers based on priority

Advantage

• does not require each router interface to be assigned a unique IP address
  • can configure routers to use only the virtual IP address
    • useful on subnets where address space utilization is high