Overview

Generic Routing Encapsulation (GRE) is a tunneling protocol that can encapsulate almost any network layer protocol and delivers it over an IP network. GRE is defined by two tunnel endpoints, the GRE source and destination endpoint. From the common user perspective the two endpoints appear as if they are directly connected in a point-to-point configuration but in reality the packet may traverse a number of hops before it reaches the destination endpoint. The generic packet format of a GRE encapsulated packet is show in Figure 1.

Figure 1. GRE Packet Format

The original payload packet is first encapsulated with the GRE header and then encapsulated again with the delivery protocol header to transport it over the network. For example, if the network doesn't support routing IPv6 packets, the packet can be encapsulated with GRE header and then use IPv4 as the delivery protocol to transport the packet. The packet is decapsulated at the destination endpoint and the actual IPv6 packet is forwarded to the destination. One of the most common issue in enterprise networks is that on local subnets private IP addresses are used but routing packets between different subnets via private IP headers is not supported over public networks such as the internet. This issue is solved by using GRE which can encapsulate packets into a delivery protocol header that can be transported over the public network.

Application Scenario Limitation

In L2 GRE or VXLAN networks, only one next hop is allowed for the same egress interface. In the following figure, the same egress interface on Switch1 has two tunnels, that is, two next hops, which is not allowed.

However, multiple L2 GRE or VXLAN tunnels can exist from the same egress port on Switch1 if connected via the IP router, ensuring that one egress interface has only one next hop, as shown in the figure below.