Route redistribution allows routes learned by one routing protocol to be advertised into another. It is commonly required during migrations, mergers and networks where different routing protocols must exchange reachability information.
Redistribution is more than entering a single command. Engineers must translate metrics, control route direction and prevent previously redistributed routes from returning to their original domain. These skills are covered through practical routing labs in the CCNP Enterprise course.
What is route redistribution?
Route redistribution is the process of taking routes from one routing source and injecting them into another routing protocol. The router performing this function is often called an Autonomous System Boundary Router, or ASBR, although the exact term depends on the protocols involved.
Consider this diagram in words:
OSPF domain Redistribution router EIGRP domain
R1 -------------------- R2 ---------------------------- R3
10.10.10.0/24 OSPF 10 and EIGRP 100 172.16.30.0/24R2 participates in both OSPF process 10 and EIGRP autonomous system 100. Without redistribution, R2 may know routes from both protocols, but R1 does not automatically learn EIGRP routes and R3 does not automatically learn OSPF routes.
Redistribution is configured under the destination routing protocol. For example, this command takes EIGRP routes and places them into OSPF:
R2(config)# router ospf 10
R2(config-router)# redistribute eigrp 100 subnetsThe reverse direction is configured under EIGRP:
R2(config)# router eigrp 100
R2(config-router)# redistribute ospf 10 metric 10000 100 255 1 1500Redistribution is normally directional. If both domains must learn each other's networks, redistribution must be configured in both directions, with suitable filtering and loop prevention.
Why do redistributed routes need a seed metric?
Routing protocols calculate path quality differently, so a metric from one protocol cannot normally be copied directly into another. A seed metric gives the imported route an initial metric that the destination protocol understands.
OSPF uses cost, EIGRP uses a composite metric, and RIP uses hop count. A route with an EIGRP metric of 28160 therefore cannot enter OSPF with the same value and retain equivalent meaning.
| Destination protocol | Required metric treatment | Common Cisco IOS approach |
|---|---|---|
| OSPF | Assign an OSPF external cost | Use metric or accept the protocol's applicable default |
| EIGRP | Supply bandwidth, delay, reliability, load and MTU | Use an inline metric or default-metric |
| RIP | Assign a valid hop count from 1 to 15 | Use metric or default-metric |
| BGP | Uses BGP path attributes rather than an IGP metric | Apply policy for attributes such as MED and communities when needed |
How is an EIGRP seed metric configured?
EIGRP requires five values when an explicit redistribution metric is supplied. In order, they represent bandwidth, delay, reliability, load and MTU.
router eigrp 100
redistribute ospf 10 metric 10000 100 255 1 1500The values mean:
| Value | Meaning |
|---|---|
| 10000 | Minimum bandwidth in Kbit/s |
| 100 | Delay in tens of microseconds |
| 255 | Reliability, where 255 is fully reliable |
| 1 | Load, where 1 represents minimal load |
| 1500 | Path MTU in bytes |
The metric should reflect the intended path rather than being copied blindly from a sample configuration. An alternative is to define one default metric for redistributed routes:
router eigrp 100
default-metric 10000 100 255 1 1500
redistribute ospf 10Routes imported into classic EIGRP are shown as external EIGRP routes, usually with the D EX code in the routing table.
How is an OSPF seed metric configured?
OSPF accepts redistributed routes as external routes. Engineers can explicitly set the external cost and choose whether the route is external type 1 or type 2.
router ospf 10
redistribute eigrp 100 subnets metric 20 metric-type 1The subnets keyword is important because it permits classless routes to be redistributed. It should normally be included in Cisco IOS OSPF redistribution configurations.
What is the difference between OSPF E1 and E2 routes?
An OSPF E1 metric includes both the external cost and the internal OSPF cost required to reach the ASBR. An E2 metric primarily uses the external cost, making E2 routes less sensitive to the internal distance from the redistributing router.
| OSPF route type | Metric calculation | Practical use |
|---|---|---|
| E1 | Internal cost to ASBR plus external cost | Useful when routers should prefer the nearest exit |
| E2 | External cost; internal cost is used as a tie-breaker | Suitable when the external path should have a consistent advertised cost |
In a network with two redistribution points, E1 routes often provide more meaningful exit selection because internal topology cost is included. The correct choice still depends on the routing policy and traffic design.
Example routing table entries may look like this:
R1# show ip route 172.16.30.0
Routing entry for 172.16.30.0/24
Known via ospf 10, distance 110, metric 30, type extern 1
Last update from 10.0.12.2 on GigabitEthernet0/0The route is identified as OSPF external type 1, has an administrative distance of 110 and an accumulated metric of 30.
How do redistribution loops happen?
A redistribution loop can occur when a route is imported into another protocol and later redistributed back into its original protocol. Multiple redistribution routers make this route feedback problem more likely.
Consider a dual-ASBR design described in words:
R2
OSPF domain -----|----- EIGRP domain
|
R4
R2 and R4 both redistribute in both directions.Suppose 10.10.10.0/24 originates in OSPF:
- R2 imports it from OSPF into EIGRP.
- R4 learns the route as an EIGRP external route.
- R4 may try to redistribute it back into OSPF.
- OSPF now receives another advertisement for a prefix that originally came from OSPF.
The result may be an unexpected path, unstable routing or traffic following a less suitable exit. Administrative distance alone is not a complete loop-prevention mechanism because it controls route installation, not the full redistribution policy.
How can route tags prevent redistribution loops?
A route tag is a policy marker carried with a route and checked during later redistribution. One protocol can tag imported routes, while the reverse redistribution policy rejects routes carrying that tag.
The following design uses tag 110 for routes originating in OSPF and tag 120 for routes originating in EIGRP.
First, prevent EIGRP-originated routes from being sent back into EIGRP, then tag genuine OSPF routes:
route-map OSPF-TO-EIGRP deny 10
match tag 120
route-map OSPF-TO-EIGRP permit 20
set tag 110
router eigrp 100
redistribute ospf 10 metric 10000 100 255 1 1500 route-map OSPF-TO-EIGRPNext, prevent OSPF-originated routes from returning to OSPF and tag genuine EIGRP routes:
route-map EIGRP-TO-OSPF deny 10
match tag 110
route-map EIGRP-TO-OSPF permit 20
set tag 120
router ospf 10
redistribute eigrp 100 subnets metric 20 metric-type 1 route-map EIGRP-TO-OSPFThe route-map sequence matters. The deny entry checks for routes that must not cross the boundary again, while the permit entry allows and tags the remaining routes. Route tags are not packet markings and do not change packet forwarding by themselves.
How can prefix lists filter redistributed routes?
Prefix lists allow only approved networks to be redistributed. This reduces the routing domain's size and prevents infrastructure, default or unintended routes from crossing a protocol boundary.
Assume OSPF should learn only 172.16.30.0/24 from EIGRP:
ip prefix-list EIGRP-EXPORT seq 5 permit 172.16.30.0/24
route-map EIGRP-TO-OSPF permit 10
match ip address prefix-list EIGRP-EXPORT
set tag 120
router ospf 10
redistribute eigrp 100 subnets metric 20 metric-type 1 route-map EIGRP-TO-OSPFA prefix list has an implicit deny at the end. Any EIGRP route not matching the permit entry is excluded from redistribution by this route-map.
For a range of more-specific networks, ge and le can be used carefully:
ip prefix-list BRANCHES seq 5 permit 172.16.0.0/16 ge 24 le 28This permits prefixes contained within 172.16.0.0/16 whose prefix lengths are between /24 and /28. It does not permit the aggregate /16 itself.
How does administrative distance affect redistribution?
Administrative distance selects between routes learned from different sources on the same router. It does not convert metrics or automatically stop a route from being advertised back into another protocol.
Common Cisco IOS administrative distances include:
| Route source | Administrative distance |
|---|---|
| Connected | 0 |
| Static | 1 |
| EIGRP summary | 5 |
| eBGP | 20 |
| Internal EIGRP | 90 |
| OSPF | 110 |
| RIP | 120 |
| External EIGRP | 170 |
| iBGP | 200 |
A redistribution router uses routes present in its routing information base and permitted by its policy. If the expected source route is not installed, the redistribution command may have nothing eligible to inject. Changing administrative distance can also create unexpected path selection, so filtering and tags are usually clearer policy tools.
How do you verify route redistribution?
Verification should confirm the source route, redistribution configuration, route-map matches and final route on a receiving router. Checking only the configuration is not enough.
Use this sequence:
show ip route 172.16.30.0
show ip protocols
show route-map
show ip prefix-list
show ip ospf database external
show ip eigrp topology 10.10.10.0/24show ip protocols displays routing processes, redistribution sources and configured filters. show route-map can show whether route-map entries are matching, while show ip prefix-list reports prefix-list hit counts on supported IOS versions.
On an OSPF router, verify external routes with:
R1# show ip route ospf
O E1 172.16.30.0/24 [110/30] via 10.0.12.2, 00:04:18, GigabitEthernet0/0On an EIGRP router, a redistributed OSPF route may appear as:
R3# show ip route eigrp
D EX 10.10.10.0/24 [170/25600256] via 10.0.23.2, 00:03:41, GigabitEthernet0/0D EX confirms that the route is external to EIGRP. The value 170 is the default administrative distance for an external EIGRP route, while the second value is its EIGRP metric.
Why is a redistributed route missing?
A missing redistributed route is commonly caused by an absent source route, an invalid seed metric, a route-map denial or incorrect process configuration. Troubleshoot from the redistribution router outward rather than repeatedly changing commands.
1. Confirm that the source route exists
show ip route 172.16.30.0The exact route must be present from an eligible source. A route learned by a different protocol than expected may not match the configured redistribute statement.
2. Check the destination routing process
show ip protocols
show running-config | section routerConfirm the correct OSPF process ID, EIGRP autonomous system number and address-family context.
3. Validate the seed metric
EIGRP and RIP need usable metrics for imported routes. Check for an inline metric or a default-metric statement.
4. Inspect route maps and prefix lists
show route-map EIGRP-TO-OSPF
show ip prefix-list EIGRP-EXPORTLook for incorrect prefix lengths, missing permit sequences and unexpected tag matches. Remember that both route maps and prefix lists have implicit deny behaviour when no permit statement matches.
5. Check the receiving protocol's database
A route may exist in the OSPF link-state database but not be installed in the routing table because a better route already exists.
show ip ospf database external
show ip route 172.16.30.06. Test forwarding, not only route presence
ping 172.16.30.1 source 10.10.10.1
traceroute 172.16.30.1 source 10.10.10.1A route in one direction does not guarantee a return route. Verify both sides and check ACLs or firewalls if routing is correct but traffic still fails.
What are the best practices for safe redistribution?
Safe redistribution starts with a clear route policy rather than unrestricted mutual redistribution. Advertise only required prefixes, assign deliberate metrics and define how routes are prevented from returning to their source.
- Prefer a single redistribution point when the design permits it.
- Use prefix lists to allow only approved networks.
- Use route tags in designs with multiple redistribution points.
- Set seed metrics based on intended path preference.
- Document route ownership and tag meanings.
- Avoid redistributing every connected or static route without a filter.
- Verify return-path reachability.
- Test failure scenarios by shutting one redistribution link in a lab.
Students preparing for advanced routing labs can also review the CCNP ENARSI practical study order. For scenario-based revision, see these CCNP interview questions on BGP and route redistribution.
Frequently asked questions
Is route redistribution required when a router runs two protocols?
No. A router can run two routing protocols and maintain routes from both without advertising those routes between the protocols. Redistribution is needed only when one routing domain must learn routes from the other.
Is two-way redistribution always necessary?
No. One-way redistribution may be sufficient when the return path is supplied by a default route or another controlled mechanism. Two-way redistribution should be used only when both domains need specific routes and loop prevention is in place.
What happens if an EIGRP seed metric is missing?
Routes requiring a seed metric may not be imported into EIGRP with a usable metric. Configure the five metric values on the redistribute command or define default-metric under the EIGRP process.
Does administrative distance prevent redistribution loops?
Not reliably. Administrative distance influences which route enters the local routing table, but it does not express route origin or provide complete feedback prevention. Route tags and explicit filtering are better controls.
Should OSPF redistributed routes use E1 or E2?
Use E1 when the internal OSPF cost to the ASBR should influence exit selection. Use E2 when the external metric should remain the main comparison, while recognising that internal cost can still act as a tie-breaker.
Can route redistribution cause asymmetric routing?
Yes. Different metrics, administrative distances or filters can make each side choose a different path. Asymmetry may be acceptable, but it must be checked carefully when stateful firewalls or path-sensitive applications are present.
Summary
Route redistribution connects separate routing domains, but the protocols do not share a common metric system. A reliable design supplies valid seed metrics, filters exported prefixes, uses route tags to stop route feedback and verifies both forward and return paths.
To practise OSPF, EIGRP, BGP, route maps and redistribution troubleshooting in structured labs, enquire about syllabus and batch details for the CCNP Enterprise course.
Reviewed by Network Rhinos networking trainers.