Redistribution Done Wrong – Top 5 Mistakes in Mixed Routing Environments

by

You’ve got OSPF running in one part of the network, EIGRP in another, and a handful of static routes thrown in for good measure.

What do you do?

You set up redistribution — the magical bridge between protocols.
Everything works… until it doesn’t.
Suddenly, your routes are flapping, loops appear, traffic drops, and someone says:

“Who touched the redistribution config?”

If that sounds familiar, this blog is for you.

Let’s walk through the Top 5 most common (and painful) mistakes engineers make while configuring route redistribution — and how to fix or avoid them

First — What is Route Redistribution, Really?

Route redistribution is when you inject routes from one routing protocol into another.

For example:

  • Injecting OSPF routes into EIGRP
  • Redistributing static routes into BGP
  • Mixing RIP and OSPF (hopefully not in production)

It allows networks using different protocols to share reachability — but it also breaks things quickly if done wrong.

Mistake #1: Redistributing Without Route Filtering

The Problem:
You type:

router ospf 1

redistribute eigrp 100

…and boom — you’ve now dumped every route from EIGRP into OSPF.
This leads to:

  • Routing table bloat
  • Loop risk
  • Unstable behavior

Unfiltered redistribution is like opening a fire hydrant — all or nothing.

 How to Fix It:

Use route-maps or distribute-lists to filter only what you want.

Example:

route-map OSPF-IN permit 10 

 match ip address 10

router ospf 1 

redistribute eigrp 100 route-map OSPF-IN

And create an access-list:

access-list 10 permit 10.10.0.0 0.0.255.255

Be intentional. Redistribute only what the receiving protocol needs.

Mistake #2: Forgetting to Set a Default Metric

Some protocols (like RIP or EIGRP) require a default metric to be assigned when redistributing routes. If you don’t, they won’t accept the routes — or worse, they’ll get default/zero metrics and misbehave.

For example:

router eigrp 100 

redistribute ospf 1

Now the routes may appear — but without usable metrics, EIGRP may ignore them.

Fix It:

Set the metric explicitly:

router eigrp 100 

default-metric 10000 100 255 1 1500 

redistribute ospf 1

Or apply per-route metrics via route-maps.

Think of metrics as “trust scores” — every protocol speaks a different language, so you need a translator.

Mistake #3: Two-Way (Mutual) Redistribution Without Route Tagging

This is the most dangerous mistake — and the one that causes routing loops.

Scenario:

  • You redistribute OSPF into EIGRP
  • Then redistribute EIGRP back into OSPF
  • Now routes keep bouncing back and forth
  • Suddenly, your routing table explodes, and traffic breaks

Welcome to the routing loop nightmare.

Solution: Use Route Tagging and Filters

Example:

route-map OSPF-to-EIGRP permit 10 

 set tag 100

 !

route-map EIGRP-to-OSPF deny 10 

 match tag 100 

route-map EIGRP-to-OSPF permit 20

Then apply:

router eigrp 100 

redistribute ospf 1 route-map OSPF-to-EIGRP

router ospf 1 

redistribute eigrp 100 subnets route-map EIGRP-to-OSPF

Tag routes as they enter a domain, and prevent re-importing them back in.

Mistake #4: Assuming All Protocols Treat Routes the Same

Routing protocols use different:

  • Metrics (OSPF = cost, EIGRP = bandwidth/delay, RIP = hop count)
  • Summarization behavior
  • Loop prevention mechanisms

If you redistribute blindly, you’ll create:

  • Asymmetric paths
  • Sub-optimal routing
  • Unpredictable behavior

Example:

  • OSPF auto-summarizes nothing
  • RIP summarizes at class boundaries by default
  • EIGRP might redistribute with a much higher metric than expected

Tip:

  • Know the default behavior of each protocol
  • Use show ip route to verify received metrics

Adjust with metric, metric-type, or route-map as needed

Mistake #5: Ignoring Administrative Distance and Failing to Control Route Preference

Let’s say:

  • You redistribute OSPF into EIGRP
  • EIGRP now has two routes to the same destination:
    • One from internal EIGRP
    • One from redistributed OSPF (external)

If you don’t manage AD (Administrative Distance) and route source preference, the router may choose the wrong path.

Redistributed routes often have higher AD (e.g., EIGRP external = 170)

Fix:

  • Use distance command to adjust preference
  • Use summarization or filtering to avoid overlapping prefixes
  • Monitor with:

show ip route <prefix> 

show ip protocols

Summary – Avoiding Redistribution Disasters

Mistake Result Fix
No route filtering Route flooding, loops Use route-map, access-list
No default metric set Routes ignored or broken Use default-metric or set metric
Two-way redistribution w/o tags Loops, duplicate routes Use set tag and deny re-import
Ignoring metric differences Poor routing decisions Normalize metrics, use type settings
Not controlling AD Wrong route wins Adjust distance values

Final Thoughts — Redistribution Is Powerful, but Dangerous

Redistribution is like connecting two different languages. You need a translator — and a set of rules to prevent chaos.

It’s easy to make it look like it’s working, but very hard to make it work cleanly and predictably.

Take it slow. Filter carefully. Tag intelligently. And monitor everything.

Leave a Comment