OSPF E1 and E2

[vishalpatil86]

Dear All,

What difference does it make in using metric type E1 or E2 while redistributing(OSPF) .

If the E2 is only external cost, why would network admin use E1 which is additional cost to external to reach the destination?

Regards,

Vishal

[InayathUlla Sharieff]

Vishal,

Simple answer;

Routes are redistributed in OSPF as either type 1 (E1) routes or type 2 (E2) routes, with type 2 being the default. A type 1 route has a metric that is the sum of the internal OSPF cost and the external redistributed cost. A type 2 route has a metric equal only to the redistributed cost, as shown in

Figure 14-2

. If routes are redistributed into OSPF as type 2 then every router in the OSPF domain will see the same cost to reach the external networks. If routes are redistributed into OSPF as type 1, then the cost to reach the external networks could vary from router to router.

Ref: http://www.ciscopress.com/articles/article.asp?p=27573&seqNum=4

http://blog.ine.com/2011/04/04/understanding-ospf-external-route-path-selection/

http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a00800943c5.shtml

HTH

Regards

Inayath

*Plz rate if this is helpfull.

[antonio.guirado]

Hello,

sometimes a practical example can be useful. Please, think in an enterprise with 2 internet conections A and B.

In your design A is a active connection and B is a backup connection. Each connection has a router and you

want to announce a default route in OSPF domain. If you use metric-type 1 internal routers forward traffic to the closest

router (A or B) basing in the lowest cost (redistribute cost + link costs). Notice, the trafic path can not be optimal because some packets can arrive router B. If you use metric-type 2 all routers have the same cost (only redistribute

cost) for default route injected by A and the same cost (of course different to A cost) for the default route injected by B.

That way you can have a real active-backup internet connection with optimal path.

Instead if your design is active-active is better use metric-type 1 to use the closest router.

I hope it can be useful.

Regards.

[Aninda Chatterjee]

This is how I remember when and why E1 or E2 routes should be used:

Think of yourself in a house with two doors (essentially giving you two exit points). You'd naturally want to go out the door that is closest to you. In this situation, do you need to know which door is closer to you? Yes. In this situation, would it help if you were told that each door is X away from you? No, because that gives you no way of distinguishing which door is closer. So, in this situation, you need a more real and calculated value to each door being told to you (as opposed to the same value being told for both doors) so that you can take an informed decision of which door to go to based on which is closer. This is when you'd want to use E1 routes.

On the other hand, if you are in a house with just one door (one exit point), does it matter if the value is correctly accumalated and sent to you? Not really, because you have no real choices. It's just that one exit point and you need to take that no matter what. So, in this situation, you can use E2 routes.

The entire reasoning behind this is that one router might be closer to your destination than another router, so it makes sense to use that one. E1 redistributed routes help in achieveing that.

Hope this helps.

Message was edited by: Aninda Chatterjee

[cadet alain]

Hi,

I like this way of explaining.

Regards

Alain

Don't forget to rate helpful posts.

[vishalpatil86]

me too....

thanks for the explanation

[ChrisKrause36323]

very illuminating :)