User's Manual

64-28
Chapter 64 VRRP Configuration
64.1 Introduction to VRRP
VRRP (Virtual Router Redundancy Protocol) is a fault tolerant protocol designed to enhance connection
reliability between routers (or L3 Ethernet switches) and external devices. It is developed by the IETF for local
area networks (LAN) with multicast/broadcast capability (Ethernet is a Configuration Example) and has wide
applications.
All hosts in one LAN generally have a default route configured to specified default gateway, any packet
destined to an address outside the native segment will be sent to the default gateway via this default route.
These hosts in the LAN can communicate with the external networks. However, if the communication link
connecting the router serving as default game and external networks fails, all hosts using that gateway as the
default next hop route will be unable to communicate with the external networks.
VRRP emerged to resolve such problem. VRRP runs on multiple routers in a LAN, simulating a "virtual" router
(also referred to as a "Standby cluster") with the multiple routes. There is an active router (the "Master") and
one or more backup routers (the "Backup") in the Standby cluster. The workload of the virtual router is actually
undertaken by the active router, while the Backup routers serve as backups for the active router.
The virtual router has its own "virtual" IP address (can be identical with the IP address of some router in the
Standby cluster), and routers in the Standby cluster also have their own IP address. Since VRRP runs on
routes or Ethernet Switches only, the Standby cluster is transparent to the hosts with the segment. To them,
there exists only the IP address of the Virtual Router instead of the actual IP addresses of the Master and
Backup(s). And the default gateway setting of all the hosts uses the IP address of the Virtual Router.
Therefore, hosts within the LAN communicate with the other networks via this Virtual Router. But basically,
they are communicating with the other networks via the Master. In the case when the Master of the Standby
cluster fails, a backup will take over its task and become the Master to serve all the hosts in the LAN, so that
uninterrupted communication between LAN hosts and external networks can be achieved.
To sum it up, in a VRRP Standby cluster, there is always a router/Ethernet serving as the active router
(Master), while the rest of the Standby cluster servers act as the backup router(s) (Backup, can be multiple)
and monitor the activity of Master all the time. Should the Master fail, a new Master will be elected by all the
Backups to take over the work and continue serving the hosts within the segment. Since the election and
take-over duration is brief and smooth, hosts within the segment can use the Virtual Router as normal and
uninterrupted communication can be achieved.