In an MPLS network with Segment Routing (SR), ensuring rapid recovery from link or node failures is crucial for maintaining high network availability. Cisco’s IOS-XR supports various Fast Reroute (FRR) mechanisms within MPLS Segment Routing environments, including Link Protection, Node Protection, and Topology Independent Loop-Free Alternate (TI-LFA). This article will provide a configuration guide and explanations for these MPLS-SR FRR mechanisms in an MPLS SR-enabled network running OSPF as the Interior Gateway Protocol (IGP).
What is TI-LFA in an MPLS network?
TI-LFA (Topology-Independent Loop-Free Alternate) is a fast reroute (FRR) mechanism in MPLS Segment Routing (MPLS-SR) designed to provide sub-50ms recovery from link or node failures in IP/MPLS networks. Unlike traditional LFA, which relies on specific topological conditions and may not cover all failure scenarios, TI-LFA is “topology-independent,” meaning it ensures protection for all traffic flows regardless of network topology. TI-LFA works by precomputing backup paths based on Segment Routing (SR) policies, allowing traffic to be quickly rerouted through alternate paths without complex recalculations. In the event of a failure, traffic is redirected via pre-established repair paths that adhere to the shortest path routing principles (SPF), minimizing packet loss and maintaining high availability across the network.
In short MPLS Segment Routing (MPLS-SR) is a modern approach to routing in MPLS (Multiprotocol Label Switching) networks. It allows for the efficient steering of traffic through predefined network segments. These segments are advertised by link-state routing protocols (IS-IS, OSPFv2, and OSPFv3) within IGP topologies. With MPLS-SR, path control and traffic engineering can be achieved without the need for protocols like LDP or RSVP-TE, which are typically used to set up traffic-engineered paths in traditional MPLS networks. These segments are stacked as labels in packet headers, allowing routers to follow predefined paths for traffic without this additional state tracking. This simplifies traffic engineering, enhances scalability, and enables efficient routing.
In this lab I will demonstrate the process of migrating from a traditional MPLS setup (using MPLS, OSPF, and LDP) to an MPLS-SR configuration on IOS-XR and IOS-XE. In this guide, you will see the configuration steps required and differences between both operating systems. Following this, I will configure a Segment Routing Mapping Server (SRMS) to map the prefixes of IOSv routers since they don’t support MPLS-SR.
Multiprotocol Label Switching (MPLS) is a data-forwarding technique that uses labels to route packets along predefined paths, rather than traditional IP routing, which relies on layer-3 addresses. By attaching labels to packets, MPLS allows for high-speed data transfers and efficient traffic engineering, making it a go-to for large-scale carrier and enterprise networks.
Segment Routing (SR), is an extension for link-state IGPs (OSPF and IS-IS). Traditional MPLS forms Label Switched Paths (LSPs) through label distribution protocols such as LDP or RSVP-TE. SR simplifies this by eliminating these protocols and allowing source-based routing. In SR, the source node attaches a list of segments (labels) to the packet, guiding it through the desired path without the need for intermediate nodes to compute routing decisions. This shift enables networks to be more adaptable and scalable.
MPLS Traffic Engineering (MPLS-TE) is a technology that enhances the capabilities of MPLS (Multiprotocol Label Switching) to enable more granular control over traffic flow within a network.
Traffic engineering refers to the practice of optimizing the flow of network traffic in a way that ensures efficient use of network resources, avoids congestion, and achieves better overall performance. In traditional IP networks, traffic generally follows the shortest path, which can lead to suboptimal usage of network capacity and congestion. MPLS-TE allows operators to move beyond shortest-path routing by explicitly setting up paths through the network that distribute traffic in a desired way.
In this lab I am going to configure a tunnel to overrule the IGP shortest path and chose a different path.
Using Cisco’s Modeling Labs (CML) I build the following MPLS lab using OSPF and LDP neighbor relationships using virtual routers running IOSv.
3 x P routers (Router1, Router2, Router3)
2 x PE router (Router4, Router5)
2 x CE router (Router6, Router7)
Default Behaviour The default traffic flow behaviour from PE Router4 towards PE Router5 will follow the IGP shortest path via P Router3. This path is one hop instead of traversing via Router1+Router2 being two hops away and double the cost. MPLS-TE With MPLS-TE we can define a different path via Router1+Router2. There can be many reasons why we would want to do this and many ways how we can achieve this. In this Lab I am going to enable MPLS-TE and simply exclude Router3 from our path.
For example the link between Router1 and Router2 gives on Router1: 10.1.2.1/24and on Router2: 10.1.2.2/24.
Verification on Router3 (P):
Router3#sh ip ospf neighborNeighbor ID Pri State Dead Time Address Interface4.4.4.40 FULL/-00:00:3510.3.4.4 GigabitEthernet0/01.1.1.10 FULL/-00:00:3810.1.3.1 GigabitEthernet0/12.2.2.20 FULL/-00:00:3510.2.3.2 GigabitEthernet0/25.5.5.50 FULL/-00:00:3310.3.5.5 GigabitEthernet0/3Router3#sh mpls interfacesInterface IP Tunnel BGP Static OperationalGigabitEthernet0/0 Yes (ldp) No No No YesGigabitEthernet0/1 Yes (ldp) No No No YesGigabitEthernet0/2 Yes (ldp) No No No YesGigabitEthernet0/3 Yes (ldp) No No No YesRouter3#sh mpls ldp neighbor Peer LDP Ident: 5.5.5.5:0; Local LDP Ident 3.3.3.3:0 TCP connection: 5.5.5.5.57381-3.3.3.3.646 State: Oper; Msgs sent/rcvd: 46/45; Downstream Up time: 00:26:54 LDP discovery sources: GigabitEthernet0/3, Src IP addr: 10.3.5.5 Addresses bound to peer LDP Ident:10.2.5.55.5.5.510.3.5.5 Peer LDP Ident: 4.4.4.4:0; Local LDP Ident 3.3.3.3:0 TCP connection: 4.4.4.4.42087-3.3.3.3.646 State: Oper; Msgs sent/rcvd: 44/46; Downstream Up time: 00:26:54 LDP discovery sources: GigabitEthernet0/0, Src IP addr: 10.3.4.4 Addresses bound to peer LDP Ident:10.3.4.44.4.4.410.1.4.4 Peer LDP Ident: 2.2.2.2:0; Local LDP Ident 3.3.3.3:0 TCP connection: 2.2.2.2.646-3.3.3.3.23943 State: Oper; Msgs sent/rcvd: 45/45; Downstream Up time: 00:26:51 LDP discovery sources: GigabitEthernet0/2, Src IP addr: 10.2.3.2 Addresses bound to peer LDP Ident:10.1.2.210.2.5.210.2.3.22.2.2.2 Peer LDP Ident: 1.1.1.1:0; Local LDP Ident 3.3.3.3:0 TCP connection: 1.1.1.1.646-3.3.3.3.22044 State: Oper; Msgs sent/rcvd: 45/45; Downstream Up time: 00:26:51 LDP discovery sources: GigabitEthernet0/1, Src IP addr: 10.1.3.1 Addresses bound to peer LDP Ident:10.1.2.110.1.3.110.1.4.11.1.1.1
Router Configurations
P Routers: (Router1, Router2, Router3)
The P routers are configured with the standard subnetting scheme from the table above in combination with OSPF area 0 and LDP as the labelling protocol. The Label range is based on the Router number.
Router1, Router2, Router2##---- MPLS ranges and LDP#---- Modify label range per routermpls label range 100199mpls label protocol ldpmpls ldp router-id Loopback0 force#---- Interface configuration with MPLS & OSPFinterface Loopback0 ip address 1.1.1.1255.255.255.255 ip ospf 1 area 0!interface GigabitEthernet0/0 ip address 10.1.2.1255.255.255.0 ip ospf network point-to-point ip ospf 1 area 0 mpls ip!interface GigabitEthernet0/1 ip address 10.1.3.1255.255.255.0 ip ospf network point-to-point ip ospf 1 area 0 mpls ip!interface GigabitEthernet0/3 ip address 10.1.4.1255.255.255.0 ip ospf network point-to-point ip ospf 1 area 0 mpls ip!
PE Routers: (Router4, Router5)
The PE routers are configured with the standard subnetting scheme from the table above in combination with OSPF area 0 and LDP as the labelling protocol. Each PE routers has an IBGP session to the other PE router (Router4 <-> Router5) for CE traffic.
Router4 (PE)# ============= MPLSmpls label range 400499mpls label protocol ldpmpls ldp router-id Loopback0 force# ===== Interfaces interface Loopback0 ip address 4.4.4.4255.255.255.255 ip ospf 1 area 0!interface GigabitEthernet0/0 ip address 10.3.4.4255.255.255.0 ip ospf network point-to-point ip ospf 1 area 0 mpls ip!interface GigabitEthernet0/1 ip vrf forwarding CUST ip address 10.4.6.4255.255.255.0!interface GigabitEthernet0/3 ip address 10.1.4.4255.255.255.0 ip ospf network point-to-point ip ospf 1 area 0 mpls ip# ============= OSPFrouter ospf 1 router-id 4.4.4.4!# =========== BGProuter bgp 65000template peer-session IBGP remote-as 65000 transport connection-mode active update-source Loopback0 exit-peer-session ! bgp router-id 4.4.4.4 bgp log-neighbor-changes no bgp default ipv4-unicast neighbor 5.5.5.5 inherit peer-session IBGP neighbor 5.5.5.5 transport connection-mode passive ! address-family ipv4 exit-address-family ! address-family vpnv4 neighbor 5.5.5.5 activate neighbor 5.5.5.5 send-community extended neighbor 5.5.5.5 next-hop-self exit-address-family ! address-family ipv4 vrf CUST neighbor 10.4.6.6 remote-as 65006 neighbor 10.4.6.6 activate neighbor 10.4.6.6 as-override exit-address-family
Router5 (PE)# ============= MPLSmpls label range 500599mpls label protocol ldpmpls ldp router-id Loopback0 force# ===== Interfaces interface Loopback0 ip address 5.5.5.5255.255.255.255 ip ospf 1 area 0!interface GigabitEthernet0/1 ip address 10.2.5.5255.255.255.0 ip ospf network point-to-point ip ospf 1 area 0 mpls ip!interface GigabitEthernet0/2 ip vrf forwarding CUST ip address 10.5.7.5255.255.255.0!interface GigabitEthernet0/3 ip address 10.3.5.5255.255.255.0 ip ospf network point-to-point ip ospf 1 area 0 mpls ip# ============= OSPFrouter ospf 1 router-id 5.5.5.5!# =========== BGProuter bgp 65000template peer-session IBGP remote-as 65000 transport connection-mode active update-source Loopback0 exit-peer-session ! bgp router-id 5.5.5.5 bgp log-neighbor-changes no bgp default ipv4-unicast neighbor 4.4.4.4 inherit peer-session IBGP ! address-family ipv4 exit-address-family ! address-family vpnv4 neighbor 4.4.4.4 activate neighbor 4.4.4.4 send-community extended neighbor 4.4.4.4 next-hop-self exit-address-family ! address-family ipv4 vrf CUST neighbor 10.5.7.7 remote-as 65006 neighbor 10.5.7.7 activate neighbor 10.5.7.7 as-override exit-address-family
Traceroute between CE routers before TE (R6->R7)
When performing a traceroute between CE routers we see the default IGP shortest path behaviour. R6 -> R4 -> R3 -> R5 -> R7. After MPLS-TE we will have created the following path: R6 -> R4 -> R1 -> R2 -> R5 -> R7.
MPLS, or Multiprotocol Label Switching, is a technique that enhances the speed and efficiency of data flow across complex networks. It operates by adding short path labels to network packets, directing them through a predetermined Label-Switched Path (LSP) rather than traditional IP-based routing. These labels contain all the forwarding information, allowing routers to forward packets based on the label rather than performing complex IP lookups. By simplifying the routing decision process, MPLS can reduce latency, optimize network performance, and enable quality-of-service (QoS) features that guarantee certain levels of bandwidth and prioritize critical applications like voice and video.
MPLS is widely used in service provider networks, supporting technologies like VPNs (Virtual Private Networks) and traffic engineering. In a typical MPLS setup, labels are assigned and stripped at the network’s edge, so the core network can process packets quickly without IP overhead. Additionally, MPLS is adaptable to various network protocols and media, enabling seamless interoperability across different types of infrastructure. By allowing network operators to manage traffic dynamically and reroute around congestion or failures, MPLS ensures greater reliability and robustness, making it a preferred choice for large-scale enterprise and ISP networks.
MPLS Traffic Engineering (MPLS-TE) is a technology that enhances the capabilities of MPLS to enable more granular control over traffic flow within a network. This is achieved by manipulating traffic paths to optimize resource usage, avoid congestion, and meet specific service requirements, like bandwidth guarantees or low latency. Here are key methods by which MPLS-TE can manipulate paths and traffic flow:
MPLS-TE Traffic manipulation options
Explicit Routing with Constraint-Based Routing (CBR)
Constraint-based routing allows MPLS-TE to create Label-Switched Paths (LSPs) that follow a specific path through the network, rather than relying on traditional routing protocols.
Explicit path setup enables network operators to define exact paths based on link attributes, resource availability, or even administrative preferences, avoiding congested or unreliable links.
Constraints can include bandwidth, latency, maximum hop count, and available resources.
! Define an explicit path list for the TE tunnelRouter(config)# ip explicit-pathname Path_R1_R3Router(config-ip-expl-path)# next-address 10.1.1.2 ! IP of Router2Router(config-ip-expl-path)# next-address 10.1.2.2 ! IP of Router3! Configure the TE TunnelRouter(config)# interface Tunnel1Router(config-if)# ip unnumbered Loopback0Router(config-if)# tunnel mode mpls traffic-engRouter(config-if)# tunnel destination 10.1.3.3 ! Destination (Router3)Router(config-if)# tunnel mpls traffic-eng path-option 1 explicit name Path_R1_R3Router(config-if)# tunnel mpls traffic-eng bandwidth 1000 ! Set bandwidth constraintRouter(config-if)# no shutdown
Traffic Engineering Database (TED)
The TED collects information on the state of the network, such as available bandwidth, link utilization, and link properties.
MPLS-TE uses the TED to make dynamic routing decisions based on real-time information, thus selecting paths that avoid congested areas and optimize resource use.
! Enable traffic engineering on OSPFRouter(config)# router ospf 1Router(config-router)# mpls traffic-eng router-id Loopback0Router(config-router)# mpls traffic-eng area 0! Ensure interfaces participate in TERouter(config)# interface GigabitEthernet0/1Router(config-if)# ip router ospf 1 area 0Router(config-if)# mpls traffic-eng tunnels
Resource Reservation with RSVP-TE
RSVP-TE (Resource Reservation Protocol with TE extensions) is used to signal and reserve resources along the selected path.
This protocol sets up traffic-engineered LSPs (TE LSPs) and reserves the necessary bandwidth to meet quality-of-service (QoS) requirements.
With RSVP-TE, MPLS-TE can ensure certain traffic flows (like voice or video) get dedicated resources, reducing packet loss and jitter.
! Enable RSVP globallyRouter(config)# mpls traffic-eng tunnelsRouter(config)# ip rsvp signaling hello! Enable RSVP on each interface used by the MPLS-TE tunnelRouter(config)# interface GigabitEthernet0/1Router(config-if)# ip rsvp bandwidth 100001000 ! Interface bandwidth in kbps, reserved bandwidth! Configure an MPLS-TE tunnel with RSVPRouter(config)# interface Tunnel2Router(config-if)# ip unnumbered Loopback0Router(config-if)# tunnel mode mpls traffic-engRouter(config-if)# tunnel destination 10.1.3.3Router(config-if)# tunnel mpls traffic-eng bandwidth 2000Router(config-if)# tunnel mpls traffic-eng path-option 1 dynamicRouter(config-if)# no shutdown
Fast Reroute (FRR)
Fast Reroute enables rapid path switching in case of a link or node failure, ensuring minimal disruption.
FRR pre-establishes backup LSPs so that traffic can be diverted almost instantaneously in case of an issue on the primary path, enhancing reliability and service continuity.
! Configure fast reroute on the tunnel interfaceRouter(config)# interface Tunnel2Router(config-if)# mpls traffic-eng fast-rerouteRouter(config-if)# tunnel mpls traffic-eng path-option 1 dynamicRouter(config-if)# no shutdown
Load Balancing and Path Diversity
MPLS-TE supports load balancing by distributing traffic across multiple LSPs. This is particularly useful for high-traffic routes that need more bandwidth than a single path can provide.
Path diversity ensures that critical data can be split across multiple paths, reducing the risk of a single point of failure and improving network redundancy.
Router(config)# interface Tunnel3Router(config-if)# ip unnumbered Loopback0Router(config-if)# tunnel mode mpls traffic-engRouter(config-if)# tunnel destination 10.1.3.3Router(config-if)# tunnel mpls traffic-eng path-option 1 dynamicRouter(config-if)# tunnel mpls traffic-eng path-option 2 explicit name Path_R1_R3Router(config-if)# no shutdown
Bandwidth Guarantees and Traffic Prioritization
MPLS-TE can allocate bandwidth to specific traffic flows, ensuring certain types of traffic, like real-time or high-priority data, meet their QoS requirements.
Differentiated services (DiffServ) can be implemented within MPLS-TE, allowing traffic prioritization at the LSP level and ensuring high-priority traffic gets preferential treatment.
! Set bandwidth requirement for TE tunnelRouter(config)# interface Tunnel4Router(config-if)# tunnel mode mpls traffic-engRouter(config-if)# tunnel destination 10.1.3.3Router(config-if)# tunnel mpls traffic-eng bandwidth 5000 ! 5000 kbps reservedRouter(config-if)# no shutdown
Administrative Policies and Affinity-Based Routing
Administrative policies (affinity or coloring) can be used to prefer or avoid certain links based on the type of traffic.
Affinity or link coloring allows paths to be marked for certain traffic types (e.g., customer A’s traffic can only use certain links), enabling more precise traffic segregation and adherence to SLA requirements.
! Define affinity on an interface (e.g., marking it with color 0x10)Router(config)# interface GigabitEthernet0/2Router(config-if)# mpls traffic-eng administrative-weight 0x10! Set affinity for the tunnelRouter(config)# interface Tunnel5Router(config-if)# tunnel mode mpls traffic-engRouter(config-if)# tunnel destination 10.1.3.3Router(config-if)# tunnel mpls traffic-eng path-option 1 dynamicRouter(config-if)# tunnel mpls traffic-eng attribute-flags affinity 0x10Router(config-if)# no shutdown
Dynamic Path Computation with Path Computation Element (PCE)
The Path Computation Element (PCE) is a centralized network component that dynamically computes paths for MPLS-TE LSPs based on network-wide data.
PCE enhances scalability and efficiency in large networks by providing real-time, optimized path computation and reducing computational strain on routers.
! Enable PCEP on the routerRouter(config)# pceRouter(config-pce)# address ipv4 10.1.4.4Router(config-pce)# source Loopback0Router(config-pce)# no shutdown! Configure the tunnel to use PCE forpath computationRouter(config)# interface Tunnel6Router(config-if)# tunnel mode mpls traffic-engRouter(config-if)# tunnel destination 10.1.3.3Router(config-if)# tunnel mpls traffic-eng path-option 1 dynamic pceRouter(config-if)# no shutdown
These examples demonstrate basic configurations for MPLS-TE features. Advanced setups may require customizations based on network architecture, device capabilities, and specific application needs.
