Routing Configuration Guide, Cisco Catalyst SD-WAN Releases 17.x

PDF

OMP route redistribution

Updated: February 6, 2026

Overview

Details which protocols are automatically redistributed into OMP (like Static and Connected) and which require explicit configuration (like BGP and EIGRP).

OMP route redistribution determines which routes OMP automatically learns and advertises, and which routes require explicit configuration for redistribution to prevent routing issues.

OMP automatically redistributes these routes after learning it either locally or from its routing peers:

  • Connected

  • Static

  • OSPF intra-area routes

  • OSPF inter-area routes

  • OSPFv3 intra-area routes (Address-Family IPv6)

  • OSPFv3 inter-area routes (Address-Family IPv6)

To avoid routing loops and less than optimal routing, redistribution of these routes require explicit configuration:

  • BGP

  • EIGRP

  • LISP

  • IS-IS

  • OSPF external routes

  • OSPFv3 external route (Address-Family IPv6)

  • OSPFv3 all routes (Address-Family IPv4)

Advertise network

The advertise network <ipv4-prefix> command advertises a specific prefix when a non-OMP route corresponding to the prefix is present in the VRF IPv4 routing table.

Note

This command is only supported for address-family ipv4 .

This is an example for advertise network configuration:

omp
  no shutdown
  graceful-restart
  address-family ipv4 vrf 1
   advertise connected
   advertise static
   advertise network X.X.X.X/X
  !

To avoid propagating excessive routing information from the edge to the access portion of the network, the routes that devices receive via OMP are not automatically redistributed into the other routing protocols running on the routers. If you want to redistribute the routes received via OMP, you must enable this redistribution locally on each device.

OMP route origin type and sub-type

OMP sets the origin and sub-origin type in each OMP route to indicate the route's origin. The Cisco Catalyst SD-WAN Controller and the router consider the origin type and subtype when selecting routes.

OMP route origin type

OMP route origin subtype

BGP

External Internal

Connected

—

OSPF

Intra-area, Inter-area, External-1, External-2, NSSA-External-1 and NSSA-External-2

OSPFv3

Intra-area, Inter-area, External-1, External-2, NSSA-External-1 and NSSA-External-2

Static

—

EIGRP

  • EIGRP Summary

  • EIGRP Internal

  • EIGRP External

LISP

—

IS-IS

Level 1 and level 2

OMP also carries the metric of the original route. A metric of 0 indicates a connected route.

Configure the redistribution of OSPF routes using CLI commands

Configure the redistribution of OSPF routes into OMP for VRF1.

Procedure

1.

Configure advertise ospf route-map <route-map-name> external .

The OSPF internal routes are redistributed into OMP by default without any explicit configuration.

Example:

This example shows the redistribution of OSPF external routes on all VRFs:

omp      
  no shutdown
  ecmp-limit       6
  graceful-restart
  no as-dot-notation
  timers  
   holdtime               300
   graceful-restart-timer 120
  exit    
  address-family ipv4
   advertise ospf external <-- This configuration implies OSPF Inter-Area/Intra-Area routes & External routes are redistributed into OMP
   advertise connected
   advertise static
  !

The following example shows the redistribution of OSPF external routes for a specific VRF:

Example:

omp      
  no shutdown
  ecmp-limit       6
  graceful-restart
  no as-dot-notation
  timers  
   holdtime               300
   graceful-restart-timer 120
  exit    
  address-family ipv4 vrf 1
   advertise ospf external
   advertise ospf route-map RLB
  !
2.

Use the external keyword in the configuration to apply the supplied route-map to both external and internal OSPF routes (Intra-Area/Inter-Area).

Example:

This example shows the redistribution of OSPFv3 external routes:

omp      
  no shutdown
  ecmp-limit       6
  graceful-restart
  no as-dot-notation
  timers  
   holdtime               300
   graceful-restart-timer 120
  exit    
   address-family ipv6  
   advertise ospfv3
   advertise ospf external
  
  !

Starting from Cisco IOS XE Catalyst SD-WAN Release 17.7.2, the real-time display of omp routes received and advertised in SD-WAN Manager are limited to only 4001 routes to avoid excessive CPU usage.

Administrative distance

Administrative distance is a metric that

  • is used to select the best path when there are two or more different routes to the same destination from multiple routing protocols, and

  • is used by Cisco Catalyst SD-WAN Controller or the router to prefer the OMP route to a destination with the lowest administrative distance value.

This table lists the default administrative distances used by the Cisco IOS XE Catalyst SD-WAN devices:

Protocol

Administrative distance

Connected

0

Static

1

NAT (NAT and static routes cannot coexist in the same VPN; NAT overwrites static routes)

1

Learned from DHCP

1

EIGRP Summary

5

EBGP

20

EIGRP

Internal: 90, External: 170

OSPF

110

OSPFv3

110

IS-IS

115

IBGP

200

OMP

251

OMP best path algorithm

Cisco IOS XE Catalyst SD-WAN devices advertise their local paths to the Cisco Catalyst SD-WAN Controller using OMP. Depending on the network topology, some paths might be advertised from multiple devices.

Cisco IOS XE Catalyst SD-WAN devices use this algorithm to choose the best path:

Table 1. Best path algorithm

Step

Applies to

Description

1

Edge devices

Cisco Catalyst SD-WAN Controller

Path validity

Checks whether the OMP path is valid. If not, ignores it.

2

Edge devices

Cisco Catalyst SD-WAN Controller

Active vs. stale paths

Prefers an active path over a stale path.

An active path is the one from a peer with which an OMP session is up. A stale path is one from a peer with which an OMP session is in Graceful Restart mode.

Note

A stale path is only advertised if the stale version is similar to the Route Information Base (RIB) version. Otherwise, the stale path is dropped.

3

Edge devices

Administrative distance

Selects the OMP path with the lower administrative distance.

Example: A path that the device learns locally via BGP would be preferred over a path that it learns from a Cisco Catalyst SD-WAN Controller via OMP. For information about administrative distance, see Administrative distance.

4

Edge devices

Cisco Catalyst SD-WAN Controller

OMP path preference

Selects the OMP path with the higher OMP path preference value.

5

 Cisco Catalyst SD-WAN Controller

Access region

Cisco Catalyst SD-WAN Controller drops advertisement from border router (BR) to BR in the same region.

6

Edge devices

Core region

Cisco Catalyst SD-WAN Controller allows advertisement between BRs in the same access region, but receives BR drops advertisement.

7

Multi-Region Fabric scenario only

Edge devices

Region path length

Compares region-path-length and prefers lower. If region-path-length-ignore is configured, then skips this step. (This addresses secondary regions in Multi-Region Fabric.)

8

Multi-Region Fabric scenario only

Border routers

Access region vs. core region

Prefers access region paths over core region paths.

9

Edge devices

Direct vs. transport gateway path

Prefers a direct path over a transport gateway path.

This step can be modified by the transport gateway path preference options, which can (a) cause the transport gateway path to be preferred, or (b) result in the paths to be considered equal. See Configure the Transport Gateway Path Preference in the Cisco Catalyst SD-WAN Multi-Region Fabric (also Hierarchical SD-WAN) Configuration Guide.

10

Multi-Region Fabric scenario only

Edge devices

Multi-Region Fabric subregion comparison

  • Prefers paths from the router's own subregion.

  • When comparing two paths that are not from the router's subregion, prefers a path that is not part of any subregion.

11

Multi-Region Fabric scenario only

Edge devices

Border router preference

Prefers a path with a higher border router preference value.

12

Edge devices

Derived affinity

Prefers a path with a lower derived affinity value.

13

Edge devices with an affinity preference configured

Affinity preference

Depending on the affinity preference configured on the device, prefers a path whose affinity is earlier in the preference list (higher priority). If the device uses affinity-preference-auto, then it prefers a path with a numerically lower affinity group.

Note

When comparing two paths with similar reorigination types, one with an affinity value and one without, prefers the path with an affinity value.

14

Edge devices

TLOC preference

Select an OMP path with a higher TLOC preference value.

Note

With TLOC preference and AAR policy configured, outbound and inbound traffic may follow different paths when the preferred TLOC goes out of SLA. For outbound traffic, tunnels in SLA will be preferred regardless of TLOC preference; however, TLOC preference still dictates inbound path selection.

15

Edge devices

Cisco Catalyst SD-WAN Controller

Origin type and subtype

Compares the origin type and subtype, and selects the first match from this list:

  • Connected

  • Static

  • EIGRP Summary

  • BGP External

  • EIGRP Internal

  • OSPF/OSPFv3 Intra-area

  • OSPF/OSPFv3 Inter-area

  • IS-IS Level 1

  • EIGRP External

  • OSPF/OSPFv3 External (External OSPF Type1 is preferred over External OSPF Type2)

  • IS-IS Level 2

  • BGP Internal

  • Unknown

16

Edge devices

Cisco Catalyst SD-WAN Controller

Origin metric

Selects an OMP path that has a lower origin metric.

17

Cisco Catalyst SD-WAN Controller

Path source

Prefers a path sourced from an edge router over the same path coming from a Cisco Catalyst SD-WAN Controller.

18

Edge devices

Cisco Catalyst SD-WAN Controller

Private IP address

If the router IDs are equal, a Cisco IOS XE Catalyst SD-WAN device selects the OMP path with the lower private IP address. If a Cisco Catalyst SD-WAN Controller receives the same prefix from two different sites and if all attributes are equal, it chooses both of them.

Note

From all equal cost multi-paths for a given prefix that are selected as best-paths and accepted by policy, advertise no more than the number of paths specified in send-path-limit.

Choosing the best path

Examples for choosing the best path:

Control connection failover and route advertisement

In a setup with two WAN Edge devices and four Cisco SD-WAN Controllers:

WAN Edge 1 forms control connections with two controllers by default (e.g., Controller 1 and Controller 2). If one of these controllers fails (e.g., Controller 1), WAN Edge 1 automatically reconnects to a backup controller (e.g., Controller 3) while maintaining its session with Controller 2.

After failover, route advertisements may change. If WAN Edge 2 originates prefix A and is connected to Controller 2, Controller 2 may not advertise prefix A to WAN Edge 1 if WAN Edge 1 now learns this prefix through Controller 3.

Thus, WAN Edge 1 only receives prefix A from Controller 3.

Day 1 Expected Behavior: A Cisco SD-WAN Controller does not re-advertise a route learned from another Cisco SD-WAN Controllerto a WAN Edge device if that device already receives the same route directly from a Cisco SD-WAN Controller.

However, this behavior is not always deterministic. Sometimes, the route may still be advertised, especially after you run the clear sdwan omp all command or disable graceful restart (GR).

Best path selection

When a Cisco SD-WAN Controller receives an OMP path (for example, 10.10.10.0/24) via OMP from a Cisco IOS XE Catalyst SD-WAN device (with an origin code of OSPF) and also from another Controller (with the same origin code), the best-path algorithm selects the path that the Cisco IOS XE Catalyst SD-WAN device sends directly, assuming all other factors are equal.

When a Controller learns the same OMP path (such as 10.10.10.0/24) from two Cisco IOS XE Catalyst SD-WAN devices at the same site, it chooses both paths and advertises them to other OMP peers, as long as all parameters are equal. By default, the Controller advertises up to four equal-cost paths.

Note

A Cisco IOS XE Catalyst SD-WAN device installs an OMP path in its forwarding table (FIB) only if the TLOC to which it points is active. For a TLOC to be active, an active BFD session must be associated with that TLOC. BFD sessions are established by each device which creates a separate BFD session with each of the remote TLOCs. If a BFD session becomes inactive, the Cisco SD-WAN Controller removes from the forwarding table all the OMP paths that point to that TLOC.