Route Priorities and Administrative Distance – Hybrid Network Routing and Connectivity – ANS-C01 Study Guide

Route Priorities and Administrative Distance

The administrative distance applies a priority value to a prefix in the routing table relative to other routing protocols that may be running on the router. A static route’s administrative distance configuration is an optional value that is used to determine the trustworthiness of the route. The lower the administrative distance, the more trustworthy the route. The default administrative distance for a static route is 1, which is the most trustworthy of all routes except for directly connected interfaces, which have a value of either 0 or 1 depending on the router vendor. Direct Connect routes are preferred over site-to-site VPN routes. So, if the eBGP routes have a lower administrative distance of 200, they would take priority over the static with a higher administrative distance of 250. Using a static route to back up a dynamic route is achieved by assigning a higher administrative distance that is referred to as a floating static route. Should the eBGP route go away, then the static may have the lowest administrative distance and be installed in the route forwarding table.

Route Summarization

AWS route summarization is a useful technique to reduce the number of route entries in the routing table by using a single, larger subnet instead of many smaller routing entries. Summarization reduces the number of routes that are advertised and stored in the routing table but does not affect reachability. Route summarization is often referred to as route aggregation.

Summarization is supported by either dynamic or static routing. Route summarization is implemented using the AWS web management console, APIs, the command-line interface (CLI), or SDKs, all of which make it easy to configure and manage route summarization.

Route summarization groups multiple contiguous subnets into a single, larger subnet, known as the summary route. By implementing summarization, the number of routes that need to be advertised and stored in the routing table is reduced, making the network more efficient and reducing the risk of routing loops.

An example of how route summarization works is to consider a block of IP addresses in use in your VPC that are in the 10.10.x.x range. You may have many subnets assigned in your availability zones using smaller chunks of this address space. For example, AZ-A may use the 10.10.1.x through 10.10.49.x blocks, AZ-B has 10.10.50.x through 10.10.99.x, AZ-C has 10.10.100.x through 10.10.149.x, and AZ-D has 10.10.150.x through 10.10.199.x. You also have the address space of 10.10.200.x through 10.10.255.x set aside for future expansion. The standard approach would be to advertise each of these subnets using a mask of 25.255.255.0, or /24, into the routing table, which would create up to 255 address blocks, or, as they are commonly called, prefixes. This would be inefficient and consume route table entries. An alternative, by using route summarization, would be to advertise 10.10.0.0 /16. This tells your remote routers that if they have a destination IP address of any device from 10.10.0.0 through 10.10.255.254 to send it to this VPC, and the local routing will know the more specific subnets to forward the packet to. Since all of your prefixes are a subset of the 10.10.0.0 /16 summarization, all packets in the 10.10.x.x range would be forwarded to your router, which would then have the more specific subnets in its route forwarding table and know how to deliver the packet to the correct subnet in your VPCs.

The AWS routed network, and all other routers, will select the route in its forwarding table that has the most specific prefix to determine the next hop for a packet. For example, if a route table has two routes to the same destination prefix, one with a /24 prefix and another with a /16 prefix, the /24 prefix route will be used. 10.10.16.0 /24 is a better, or more specific, route than a table entry of 10.10.0.0 /16. This is often called the longest match wins route selection. This means that the route with the longer prefix will be chosen.