MPLS Configuration Guide for Cisco 8000 Series Routers, IOS XR Release 24.1.x, 24.2.x, 24.3.x, 24.4.x
Bias-Free Language
The documentation set for this product strives to use bias-free language. For the purposes of this documentation set, bias-free is defined as language that does not imply discrimination based on age, disability, gender, racial identity, ethnic identity, sexual orientation, socioeconomic status, and intersectionality. Exceptions may be present in the documentation due to language that is hardcoded in the user interfaces of the product software, language used based on RFP documentation, or language that is used by a referenced third-party product. Learn more about how Cisco is using Inclusive Language.
Introduced in this release on: Fixed Systems (8700 [ASIC:K100])(select variants only*).
MPLS Static Labeling allows for manual assignment of labels to specific routes, providing fine-grained control over label
distribution and traffic engineering. This feature is ideal for network scenarios where predictable and consistent routing
paths are required. By configuring static labels, network operators can optimize resource allocation and ensure precise traffic
management without relying on dynamic label distribution protocols.
*Previously this feature was supported on Q200 and Q100. It is now extended to Cisco 8712-MOD-M routers.
The MPLS static feature enables you to statically assign local labels to an IPv4/Ipv6 prefix. Also, Label Switched Paths (LSPs)
can be provisioned for these static labels by specifying the next-hop information that is required to forward the packets
containing static label.
If there is any discrepancy between labels assigned statically and dynamically, the router issues a warning message in the
console log. By means of this warning message, the discrepancy can be identified and resolved.
The advantages of static labels over dynamic labels are:
Improves security because the risk of receiving unwanted labels from peers (running a compromised MPLS dynamic labeling protocol)
is reduced.
Gives users full control over defined LSPs.
Utilize system resources optimally because dynamic labeling is not processed.
Restrictions
The router does not prevent label discrepancy at the time of configuring static labels. Any generated discrepancy needs to
be subsequently cleared.
Equal-cost multi-path routing (ECMP) is not supported.
Interfaces must be explicitly configured to handle traffic with static MPLS labels.
The MPLS per-VRF labels cannot be shared between MPLS static and other applications.
Forwarding Labeled Packets
This section describes how labeled packets are forwarded in MPLS networks, how forwarding labeled packets are different from
forwarding IP packets, how labeled packets are load-balanced, and what a LSR does with a packet with an unknown label.
Top Label Value
When a labeled packet is received, the label value at the top of the stack is looked up. The LSR sees the 20-bit field in
the top label, which carries the actual value of the label. As a result of a successful lookup, the LSR learns:
the next hop to which the packet is to be forwarded.
what label operation to be performed before forwarding - swap, push, or pop.
The processing is always based on the top label, without regard to the possibility that in the past some other number of
another label may have been "above it", or at present that some other number of another label may be below it. An unlabeled
packet can be thought of as a packet whose label stack is empty (that is, a packet whose label stack has depth zero).
IP Lookup Versus Label Lookup
When a router receives an IP packet, an IP lookup is done. This means that the packet is looked up in the Cisco Express Forwarding
(CEF) table. When a router receives a labeled packet, the label forwarding information base (LFIB) of the router is looked
up. The router knows by looking at the protocol field in the Layer 2 header what type of packet it receives: a labeled packet
or an IP packet.
Load Balancing Labeled Packets
If multiple equal-cost paths exist for an IPv4 prefix, Cisco IOS XR Software can load-balance labeled packets. When labeled
packets are load-balanced, they can have the same or different outgoing labels. The outgoing labels are the same if the two
links are between a pair of routers and both links belong to the platform label space. If multiple next-hop LSRs exist, the
outgoing label for each path is usually different, because the next-hop LSRs assign labels independently.
Unknown Label
In regular operations, an LSR should receive only a labeled packet with a label at the top of the stack that is known to
the LSR, because the LSR would have previously advertised that label. However, it is possible, in some cases, when something
goes amiss in the MPLS network, the LSR starts receiving labeled packets with a top label that the LSR does not find in its
LFIB. In such cases, the LSR drops the packet.
Define Label Range and Enable MPLS Encapsulation
By default, MPLS encapsulation is disabled on all interfaces. MPLS encapsulation has to be explicitly enabled on all ingress
and egress MPLS interfaces through which the static MPLS labeled traffic travels.
Also, the dynamic label range needs to be defined. Any label that falls outside this dynamic range is available for manually
allocating as static labels. The router does not verify statically-configured labels against the specified label range. Therefore,
to prevent label discrepancy, ensure that you do not configure static MPLS labels that fall within the dynamic label range.
You have to accomplish the following to complete the MPLS static labeling configuration. Values are provided as an example.
Define a dynamic label range, which in this task is set between 17000 and 18000.
Enable MPLS encapsulation on the required interface.
Setup a static MPLS LSP for a specific ingress label 24035.
Specify the forwarding information so that for packets that are received with the label, 24035, the MPLS protocol swaps labels
and applies the label, 24036. After applying the new label, it forwards the packets to the next hop, 10.2.2.2, through the
specified interface.
Router# show mpls interfaces
Interface LDP Tunnel Static Enabled
-------------------------- -------- -------- -------- --------
HundredGigE 0/0/0/25 No No Yes Yes
Verify that the status is "Created" for the specified label value.
Router#show mpls static local-label all
Label VRF Type Prefix RW Configured Status
------- --------------- ------------ ---------------- --------------- --------
24035 default X-Connect NA Yes Created
Check the dynamic range and ensure that the specified local-label value is outside this range.
Router#show mpls label range
Range for dynamic labels: Min/Max: 17000/18000
Verify that the MPLS static configuration has taken effect, and the label forwarding is taking place.
During configuring or de-configuring static labels or a label range, a label discrepancy can get generated when:
A static label is configured for an IP prefix that already has a binding with a dynamic label.
A static label is configured for an IP prefix, when the same label value is dynamically allocated to another IP prefix.
Verification
Identify label discrepancy by using these show commands.
Router#show mpls static local-label discrepancy
Tue Apr 22 18:36:31.614 UTC
Label VRF Type Prefix RW Configured Status
------- --------------- ------------ ---------------- --------------- --------
24000 default X-Connect NA Yes Discrepancy
Router#show mpls static local-label all
Tue Apr 22 18:36:31.614 UTC
Label VRF Type Prefix RW Configured Status
------- --------------- ------------ ---------------- --------------- --------
24000 default X-Connect N/A Yes Discrepancy
24035 default X-Connect N/A Yes Created
Router#show log
Syslog logging: enabled (0 messages dropped, 0 flushes, 0 overruns)
Console logging: level warnings, 199 messages logged
Monitor logging: level debugging, 0 messages logged
Trap logging: level informational, 0 messages logged
Buffer logging: level debugging, 2 messages logged
Log Buffer (307200 bytes):
RP/0/RSP0/CPU0:Apr 24 14:18:53.743 : mpls_static[1043]: %ROUTING-MPLS_STATIC-7-ERR_STATIC_LABEL_DISCREPANCY :
The system detected 1 label discrepancies (static label could not be allocated due to conflict with other applications).
Please use 'clear mpls static local-label discrepancy' to fix this issue.
RP/0/RSP0/CPU0:Apr 24 14:18:53.937 : config[65762]: %MGBL-CONFIG-6-DB_COMMIT : Configuration committed by user 'cisco'.
Use 'show configuration commit changes 1000000020' to view the changes.
Rectification
Label discrepancy is cleared by allocating a new label to those IP prefixes that are allocated dynamic label. The static label
configuration takes precedence while clearing discrepancy. Traffic can be affected while clearing discrepancy.
Router# clear mpls static local-label discrepancy all
Verify that the discrepancy is cleared.
Router# show mpls static local-label all
Wed Nov 25 21:45:50.368 UTC
Label VRF Type Prefix RW Configured Status
------- --------------- ------------ ---------------- --------------- --------
24000 default X-Connect N/A Yes Created
24035 default X-Connect N/A Yes Created
Associated Commands
show mpls static local-label discrepancy
clear mpls static local-label discrepancy all
Configuring Backup within a Forwarding Set
Various types of FRR backups can be configured between links with in a forwarding path set. You can configure the following
types of FRR backups:
Pure FRR Backup
Reciprocal FRR backup
One-way FRR backup
In pure FRR backup, there will be separate primary paths and backup paths. In reciprocal FRR backup, each path can act as
both primary and backup. In one-way FRR backup, some paths act as both primary and backup while other paths may be just primary
paths or backup paths.
Configuration Example: Pure FRR Backup
This example shows how to configure pure FRR backup with in a forwarding path set.
The following table describes the forwarding behavior for pure FRR backup. Here P1-F and P2-F are the forwarding paths and
P1-B and P2-B are the backup paths.
Action
Transient State
Interface Steady State
Forward Steady State
N/A
N/A
P1-F: Up P2-F: Up
P1-B: Up P2-B: Up
P1-F: Flow P2-F: Backup
P1-B: N/A P2-B: N/A
P1-F Down
P1-F FRR to P2-F
P1-F: Up P2-F: Down
P1-B: Up P2-B: Up
P1-F: Down P2-F: Flow
P1-B: Backup P2-B: N/A
P2-F Down
P2-F FRR to P1-B
P1-F: Down P2-F: Down
P1-B: Up P2-B: Up
P1-F: Down P2-F: Down
P1-B: Flow P2-B: Backup
P1-B Down
P1-B FRR to P2-B
P1-F: Down P2-F: Down
P1-B: Down P2-B: Up
P1-F: Down P2-F: Down
P1-B: Down P2-B: Flow
Configuration Example: Reciprocal FRR Backup
This example shows how to configure reciprocal FRR backup with in a forwarding path set.
The following table describes the forwarding behavior for one-way FRR backup.
Action
Transient State
Interface Steady State
Forward Steady State
N/A
N/A
P1-F: Up P2-F: Up
P1-B: Up P2-B: Up
P1-F: Flow P2-F: Flow
P1-B: N/A P2-B: N/A
P2-F Down
P2-F NO-FRR to P1-F
P1-F: Down P2-F: Up
P1-B: Up P2-B: Up
P1-F: Flow P2-F: Down
P1-B: Backup P2-B: N/A
P1-F Down
P1-F FRR to P1-B
P1-F: Down P2-F: Down
P1-B: Up P2-B: Up
P1-F: Down P2-F: Down
P1-B: Flow P2-B: Flow
P1-B Down
P1-B FRR to P2-B
P1-F: Down P2-F: Down
P1-B: Down P2-B: Up
P1-F: Down P2-F: Down
P1-B: Down P2-B: Flow
Configuring Static LSP Next Hop Resolve
You can specify the outgoing next hop instead of explicitly specifying the outgoing path while configuring static LSPs. This
next hop is resolved using the routing information base (RIB) which provides a list of paths to auto-configure. While specifying
the next hop for the incoming label in a static LSP, you can specify the next hop address with out the interface using the
resolve-nexthop command.
The following restrictions apply for this feature:
Only supports a single next hop address which may resolve to multiple paths.
Non-default VRFs are not supported.
Configuration Example
This example shows how to configure the static LSP next hop without specifying the interface using the resolve-nexthop command.
Configuring Static LSP Next Hop Resolve with Recursive Prefix
When a routing table entry references to another IP address and not to a directly connected exit interface, the next-hop IP
address is resolved using another route with an exit interface. This is known as a recursive look up because multiple lookups
are required to resolve the next-hop IP address. Static LSP next hop resolve with recursive prefix feature supports resolution
of recursive routes for static LSPs. In this feature, you can specify a next hop which is not directly connected using the
resolve-nexthop command for a static LSP.
Restrictions
The following restrictions apply for this feature:
Only eBGP routes are supported.
Configuration Example
This example shows how to configure the static LSP next hop resolve with recursive prefix. Here 192.168.2.1 is a recursive
route learnt through eBGP.
Introduced in this release on: Fixed Systems (8700 [ASIC:K100])(select variants only*).
MPLS Static Labeling allows for manual assignment of labels to specific routes, providing fine-grained control over label
distribution and traffic engineering. This feature is ideal for network scenarios where predictable and consistent routing
paths are required. By configuring static labels, network operators can optimize resource allocation and ensure precise traffic
management without relying on dynamic label distribution protocols.
*Previously this feature was supported on Q200 and Q100. It is now extended to Cisco 8712-MOD-M routers.
The MPLS static feature enables you to statically assign local labels to an IPv4/Ipv6 prefix. Also, Label Switched Paths (LSPs)
can be provisioned for these static labels by specifying the next-hop information that is required to forward the packets
containing static label.
If there is any discrepancy between labels assigned statically and dynamically, the router issues a warning message in the
console log. By means of this warning message, the discrepancy can be identified and resolved.
The advantages of static labels over dynamic labels are:
Improves security because the risk of receiving unwanted labels from peers (running a compromised MPLS dynamic labeling protocol)
is reduced.
Gives users full control over defined LSPs.
Utilize system resources optimally because dynamic labeling is not processed.
Restrictions
The router does not prevent label discrepancy at the time of configuring static labels. Any generated discrepancy needs to
be subsequently cleared.
Equal-cost multi-path routing (ECMP) is not supported.
Interfaces must be explicitly configured to handle traffic with static MPLS labels.
The MPLS per-VRF labels cannot be shared between MPLS static and other applications.
MPLS Static Forwarding Over A BVI
Table 3. Feature History Table
Feature Name
Release Information
Feature Description
MPLS Static Forwarding Over A BVI
Release 7.3.1
The router can receive MPLS L2VPN traffic from an L2 bridge domain, and forward the L3 (customer) traffic over an egress BVI,
using an MPLS static LSP. For the incoming L2VPN traffic, the BVI serves as an L3 gateway.
Since the router can perform switching for L2 traffic and routing for incoming L3 MPLS traffic, it enhances flexibility for
transporting MPLS traffic.
Consider this sample topology, connecting a CE router to a PE router.
Figure 1. MPLS Static Forwarding Over A BVI
Pointers
L2VPN packets are attached to specific bridge domains, and correspond to a VLAN (or 802.1Q tag). In turn, the VLANs are associated
with specific bundle or physical interfaces for sending traffic between the CE and PE routers. These associations have to
be configured on the CE and PE routers for transporting L2VPN traffic.
The L2VPN traffic encapsulates (IPv4 or IPv6) customer payload, and is sent from the CE router to the PE router.
The PE router does an MPLS label lookup on the incoming MPLS traffic, and removes the VLAN (or 802.1Q) header. In general,
the router can perform a label operation like swap, PHP, or pop. After removing the VLAN header, the (previously) encapsulated
IP traffic is sent towards the bridge-group virtual interface (BVI).
With BVI support for the MPLS static function, the incoming labelled traffic can be resolved using a static LSP. The BVI resolves
the nexthop to an L3 interface.
BVI pointers
A BVI next hop can be a static route, a directly connected route, or a route resolved through BGP or an IGP.
Only an MPLS static LSP can use a BVI as a next hop.
Configuration
The configurations explain how to enable forwarding of (incoming) L2VPN traffic over a (outgoing) BVI, through an MPLS static
LSP.
Interfaces Configuration
The l2transport keyword indicates that the interface is an L2 interface, and the L2 traffic belongs to the VLANs specified in the dot1q tags.
The rewrite ingress tag pop command form instructs the router to remove the 802.1Q (or VLAN) tag and forward the payload.
Corresponding configurations should also be enabled on the CE router.
Router# configure terminal
Router(config)# interface Bundle-Ether101.101 l2transport
Router(config-if)# encapsulation dot1q 2001
Router(config-if)# rewrite ingress tag pop 1 symmetric
Router(config-if)# mtu 2000
Router# configure terminal
Router(config)# interface Bundle-Ether102.101 l2transport
Router(config-if)# encapsulation dot1q 3001
Router(config-if)# rewrite ingress tag pop 1 symmetric
Router(config-if)# mtu 2000
Router# configure terminal
Router(config)# interface HundredGigE0/11/0/25.101 l2transport
Router(config-if)# encapsulation dot1q 101
Router(config-if)# rewrite ingress tag pop 1 symmetric
Router# configure terminal
Router(config)# interface HundredGigE0/11/0/31.101 l2transport
Router(config-if)# encapsulation dot1q 1001
Router(config-if)# rewrite ingress tag pop 1 symmetric
Router(config-if)# exit
Router(config)# commit
BVI Configuration
The BVI acts as an L3 gateway for the ingress L2VPN traffic. The customer IP traffic is sent to the BVI IP address that is
configured in this task.
L2VPN is configured, and a bridge domain (bd1) is associated with it.
A bundle interface and BVI are associated with the BD. The instructions enable VLAN traffic received at the bundle interface
(Bundle-Ether101.101) to be routed through the L3 gateway, BVI BVI101. VLAN-to-interface association was done in an earlier
step.
The MPLS over generic routing encapsulation (MPLSoGRE) provides a mechanism for tunneling MPLS packets over a non-MPLS network.
This feature uses MPLS over GRE to encapsulate MPLS packets inside GRE tunnels to create a virtual point-to-point link across
non-MPLS networks.
With this feature, the core network can be configured with IPv4 addresses to interconnect the MPLS networks through the IP
network.
MPLS over GRE supports MPLS static forwarding over a GRE tunnel at line rate, which is the normal speed at which the traffic
is sent through networks. For more information on line rate, see the Cisco 8000 Series Routers Data Sheet.
You can configure a provider router to send incoming customer traffic over the GRE tunnel, addressed to a set of load-balancing
servers.
The GRE tunnel is configured with encapsulation that adds an MPLS packet and a GRE header to the incoming packet at the starting
point of the tunnel. When the packet reaches the endpoint of the GRE tunnel, the GRE header is removed and the payload is
forwarded to the destination based on the MPLS label.
Figure 2. MPLS Over GRE Tunnel
In the image, you can see that the GRE tunnel begins at router R1. R1 uses the policy based routing (PBR) process for GRE
tunnel encapsulation, adds an MPLS label to the incoming packet, and then adds a GRE header. Then it sends the traffic towards
router R2.
R2 uses the PBR process for GRE tunnel decapsulation, and based on the MPLS label, it forwards the traffic towards its destination.
Configuration Example
This example shows how to enable MPLS static forwarding over GRE tunnel.
Configuration on router R1.
Configure a tunnel interface.
Configure the mode of encapsulation as GRE for the tunnel interface.
Configure a policy map and redirect the traffic to next hop IP.
Assign the policy map to a VLAN subinterface.
Configure MPLS out labels to be applied to the incoming packets.
Configuration on router R2.
Configure a tunnel interface.
Configure the mode of decapsulation as GRE for the tunnel interface.
Configure a class map with match criteria for the source and destination IP addresses.
Configure a policy map and specify the class name configured in the class map.
Configure GRE decapsulation.
GRE Tunnel Configuration on R1
The GRE tunnel starts on R1.
The GRE tunnel destination must be a valid IPv4 address.
Configure MPLS out labels. Ensure that the out label is the same for all paths of an MPLS static label switch path(LSP). You
can configure up to a maximum of 16 paths. After applying the labels, the packets are forwarded to the specified next hop.
PBR Configuration for GRE Tunnel Decapsulation on R2
Router(config)# class-map type traffic match-all test_gre1
Router(config-cmap)# match protocol gre
Router(config-cmap)# match destination-address ipv4 50.0.0.1 255.255.255.255
Router(config-cmap)# match source-address ipv4 10.0.0.1 255.255.255.255
Router(config-cmap)# end-class-map
Router(config)# policy-map type pbr P1-test
Router(config-pmap)# class type traffic test_gre1
Router(config-pmap-c)#decapsulate gre
Router(config-pmap-c)# end-policy-map
Router(config)# vrf-policy vrf default address-family ipv4 policy type pbr input P1-test
Running Configuration
Use the following show commands to view the configuration.
PBR Configuration for GRE Tunnel Encapsulation on R1
Router# show running-config policy-map type pbr *
policy-map type pbr PBR_ENCAP_1
class type traffic class-default
redirect ipv4 nexthop 111.0.0.1
!
end-policy-map
!
Router# show running-config int tunnel-ip 1
interface tunnel-ip1
ipv4 address 112.0.0.2 255.255.255.0
tunnel mode gre ipv4 decap
tunnel source Loopback 0
tunnel destination 10.0.0.1
!
PBR Configuration for GRE Tunnel Decapsulation on R2
Router# show running-config class-map type traffic match-all
class-map type traffic match-all test_gre1
match protocol gre
match destination-address ipv4 50.0.0.1 255.255.255.255
match source-address ipv4 10.0.0.1 255.255.255.255
end-class-map
!
policy-map type pbr P1-test
class type traffic test_gre1
decapsulate gre
!
class type traffic class-default
!
end-policy-map
!
vrf-policy
vrf default address-family ipv4 policy type pbr input P1-test
!
Feedback