Post date: Oct 29, 2012 4:03:34 PM
I talked with a customer today who was using VRRP to provide Layer3 Default gateway redundancy. The problem was each server is plugged into a single switch that is also providing the layer 3 routing and VRRP. So if the default gateway fails the layer 2 port will also fail so the VRRP will not be able to help as when the VRRP can do the job the port the server is attached to also goes down.
This is a good example of where you need to do some detailed failure analysis to determine what type of redundancy is really necessary and if it is really solving a problem. Removing the VRRP from the configuration makes the switches simpler to configure and troubleshoot, and this simplicity does not cost anything, (no real functionality is lost)