MPLS overview

Overview

The NFM-P supports the configuration and provisioning of MPLS paths and LSPs.

MPLS is a data-carrying mechanism that emulates some properties of a circuit-switched network over a packet-switched network. MPLS prepends a packet with an MPLS header that is comprised of labels in a stack. As an NE forwards a packet through an MPLS network, it examines the top label for routing instructions and pops it off the stack; the contents below the label stack are not examined. The label stack is not removed until it reaches the egress NE.

MPLS uses one or more IGPs, such as IS-IS, OSPF, or RIP, to forward packets. Regardless of the IGP chosen, the MPLS configuration is the same. An advantage of MPLS is that if more than one IGP is enabled, it continues to work when another protocol fails. See Chapter 28, Routing protocol configuration for more information on IS-IS, OSPF, and RIP.

MPLS can be used as the underlying transport mechanism for service tunnels. For MPLS to be used as such, an MPLS mesh and an LSP mesh must be created before the tunnel is created. Since LSPs and service tunnels are unidirectional, they must be created in both directions. See Chapter 33, Service tunnels for more information about service tunnels.

When MPLS is enabled on an NFM-P-managed NE routing instance, the NFM-P displays an MPLS instance below the routing instance in the network view of the navigation tree. Using this view, you can perform various functions on the MPLS instance, for example, assign an L3 interface to an MPLS instance.

If a link, NE, or path fails, MPLS can determine a redundant path by using fast reroute. Fast reroute uses an alternative NE to complete the failed LSP. Fast reroute provides the following types of route protection.

• Many-to-one—one backup route is maintained for multiple protected LSPs on an NE.

• One-to-one—a separate backup route is maintained for each protected LSP on an NE algorithm.

You can list and view MPLS objects such as MPLS and RSVP instances and interfaces, static and dynamic LSPs, cross connections, and MPLS paths using the NFM-P Manage MPLS Objects form.

MPLS forwarding policies

The data model of a forwarding policy represents each pair of next hops (primary and backup) as a group and models the ECMP set as the set of next-hop groups. Flows of prefixes can be switched on a per next-hop group basis from the primary next-hop, when it fails, to the backup next-hop without disturbing the flows forwarded over the other next-hop groups of the policy. The same can be performed when reverting from a backup next-hop to the restored primary next-hop of the same next-hop group.

An endpoint MPLS forwarding policy allows the user to forward unlabeled packets over a set of user-defined direct (with option to push a label stack), indirect, or LSP next-hops. Routes are bound to an endpoint policy when their next-hop matches the endpoint address of the policy.

A label-binding MPLS forwarding policy provides the same capability for labeled packets. In this case, labeled packets matching the ILM of the policy binding label are forwarded over the set of next-hops of the policy.