What is GLBP and why to use it?
Gateway Load Balancing Protocol (GLBP) is a Cisco proprietary protocol that attempts to overcome the limitations of existing redundant router protocols such as HSRP and VRRP by adding basic load balancing functionality. ARP requests sent by arp clients will be answered with MAC addresses pointing to different routers per GLBP load balancing configurations that can be deployed. Thus, by default, load balancing is not based on traffic load, but rather on the number of hosts that will use each gateway router. By default GLBP load balances in round-robin fashion.
In regards to using GLBP with dynamic routing protocols, routers do not use the virtual IP addresses of GLBP to form adjacencies. GLBP is designed to provide end-users devices and servers default gateway/router redundancy and load balancing. For instance, EIGRP neighbor relationship are not formed with the GLBP Virtual IP Address but with the physical interface IP address. As such if you are using GLBP in a active passive mode or with specific weighting along with a dynamic routing protocol you should ensure your routes prefer the applicable routers unless offline to avoid router issues.
In the network below there are a total of two routers, two end-user computers and one server. In the example below we will configure two GLBP groups, one for the LAN and one for the WAN network. Both R1 and R2 will be the gateways for the LAN and the WAN network by way of the round-robin load balancing protocol. This will make use of both routers to prevent network capacity from being underutilized. In the event of a router failure failover and failback must occur automatically. GLBP groups will also make use of authentication to prevent any rogue routers from interfering with GLBP operations.
Verify the configuration
Now that the configuration is finished lets verify our GLBP deployment. Using the ping, traceroute and show ip route on PC1 we can verify connectivity, the path used to reach the destination and default gateway. You will find ping is successful from PC1 to Server and the path may be PC1 to R1 to Server or PC1 to R2 to Server depending on which router answered the arp request first.
On R1 issue the show glbp brief and show glbp commands to review the GLBP groups. This will confirm R1 and R2 are actively forwarding traffic for groups 1 and 2.
Now let's shutdown R1 and perform the same steps from PC1. As you can see everything functions the same with the only exception your default gateway, 10.1.0.3, is routed through the backup router R2 which is now the active GLBP router for groups 1 and 2.
For the last verification step we will test load balancing. Turn R1 back on and clear the arp cache on PC1 or PC2. Your default gateway, 10.1.0.3, will be routed again through R1. If you clear the arp cache a few more times and you will see your default gateway will be routed through R2 again. This is the round-robin load balancing policy in action.
