Failover Using Gre - Avaya G250 Administration

Configuring IPSec VPN
Using the G250/G350's peer-group entity (see
●
a. Define a peer-group. A peer-group is an ordered list of redundant remote peers, only one
of which is active at any time. When the active peer is considered dead, the next peer in
the list becomes the active remote peer.
b. When configuring a crypto map, point to the peer-group instead of to a single peer.

Failover using GRE

A branch with a G250/G350 can connect to two or more VPN hub sites, in a way that will
provide either redundancy or load sharing.
In this topology, the G250/G350 is connected through its 10/100 WAN Ethernet port to a DSL
modem.
Define two GRE tunnel interfaces:
●
GRE1 that leads to a Primary Main Office GRE End Point behind the VPN Hub Gateway
●
GRE2 that leads to a Backup Main Office GRE End Point behind the VPN Hub Gateway
●
Define two VPNs.
●
Connectivity to the networks in Primary/Backup Main Office is determined through GRE
●
keep alives. If network connectivity is lost due to failures in the WAN, in the Primary Main
Office, the GRE keep-alive will fail and the GRE interface will transition to a "down" state.
The two GRE tunnels can then be used for branch to Primary/Backup Main Office in either
Redundancy or Load sharing mode:
Redundancy – GRE2 is configured as a backup interface for GRE1, and is activated only
●
when GRE1 is down.
Load sharing – Both tunnel interfaces are active. Routing protocols (RIP or OSPF) route
●
traffic to destinations based on route cost & availability, as follows:
For two routes of equal cost to the same destination, one through the Primary Main Office
and one through the Backup Main Office, OSPF will automatically distribute traffic through
both routes, effectively sharing the load between routes.
416 Administration for the Avaya G250 and Avaya G350 Media Gateways
Failover using a peer-group on page 431):