Problem solve Get help with specific problems with your technologies, process and projects.

SRLGs solve network reliability problems

How SRLGs can make your virtual networks more reliable.

For more IT articles and tips specific to small and midsized businesses, visit

Shared Risk Link Groups, SRLGs, increase the reliability of MPLS VPNs, but you must make sure your ISP has implemented them. Increased reliability is one of the advantages of MPLS over Frame Relay and ATM. IP networks quickly reroute packets when a link or node fails. Since MPLS is built on the IP networking foundation, it also reroutes quickly on a failure. SRLGs guard against situations where even the reroute doesn't resolve the failure.

The following example illustrates the use of SRLGs. (For details on how MPLS creates Label Switched Paths, or LSPs. see this article.)

In this example, router A allocates a label to define this LSP. The path goes from router A to router B and then C and D. The routers determine the route using the Interior Gateway Protocol, or IGP that the provider uses, usually OSPF or IS-IS. As a packet is forwarded down the path, each intermediate router uses the label to find out which interface to use to send the packet on its way. If a link or router fails anywhere on the route, the LSP must change to accomplish the reroute. A backup path is created at the time the initial path is created to provide for failover in milliseconds. To protect against a link failure between routers B and C or a failure of router C, a backup path can be created from B to router E and from E to D. Similarly, other backup paths are created to protect against other failures.

What happens if the link from B to C shares the same physical medium with the B to E link? Perhaps the two links are multiplexed on the same fiber at some point, and someone digs in the wrong place. Or maybe the links are on different fibers in the same conduit or running parallel in the same right of way. The result? Both links fail.

SRLGs provide a way to avoid this situation. An SRLG identifier is applied to each interface of each router in the network. The interface on router A that connects to B is given the identifier 1. The corresponding interface on B is also given identifier 1. The interface on B connecting to C and the interface on C connecting to B are given identifier 2. When two links share the same physical medium, same conduit, or same right of way anywhere, they share the same risk, so the same identifier is assigned to all of those router interfaces. If the link between B and E shares the same risk as the link from B to C, the interface on B connecting to E will also be assigned identifier 2. OSPF and IS-IS standards have been updated to propagate the risk identifiers and the routing algorithms modified so a backup route will never share a risk with the primary.

Implementing SRLGs is a time-consuming job that requires ISPs to assign risk identifiers. That means carefully examining all physical paths in the network to identify shared risks and updating identifiers every time the network changes. Major router vendors have updated their software to support SRLGs, but you must check whether your ISP has implemented SRLGs. Otherwise, your network runs the risk of failure whenever someone with a backhoe makes a mistake.

David B. Jacobs has more than twenty years of networking industry experience. He has managed leading-edge software development projects and consulted to Fortune 500 companies as well as software start-ups.

This was last published in October 2004

Dig Deeper on Network management and monitoring

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.