Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

OSPF E1 & E2 comparison

Hi Experts,

I learned the functionalities of OSPF E1 & E2 Routes and the concept of Forwarding Metric.

But I donot understand when to use which Metric type?

That is my understanding is, OSPF always uses the Lowest Metric to calculate the path and populate it in the Routing table whether it is E1 or E2 (considering the Forwarding metric).

Then were the difference lies in design perspective?

Can you shade some light on this and help us

sairam

2 ACCEPTED SOLUTIONS

Accepted Solutions
Hall of Fame Super Bronze

Re: OSPF E1 & E2 comparison

But I donot understand when to use which Metric type?

When the router needs to compare same routes with same route-type. For instance; multiple ASBRs sending the same route (E2, for instance). You can design one ASBR applying a forwarding metric of 30 while the other ASBR applying a forwarding metric of 20. You can then perform traffic engineering of redistributed routes to draw traffic towards a specified router.

So, you have several choices on the design. Say you have 3 ASBR and you want ASBR Router A to draw the routes and ASBR Router B and C to act as secondaries.

ASBR Router A will redistribute the routes as E1 and ASBR Router B and C will be redistributing as E2 with forwarding metric 20 and 30 respectively. This design will meet your requirements.

Regards

Edison.

Hall of Fame Super Silver

Re: OSPF E1 & E2 comparison

Hello Sairam,

first O E2 seed metric should be considered.

then, when both routes have the same O E2 seed metric lowest forwarding metric wins.

see

Type 2

external paths advertising equal type 2 metrics are

compared by looking at the distance to the forwarding

addresses.

http://www.ietf.org/rfc/rfc2328.txt

section 16.4.0

this is logical and explains the difference between O E1 and O E2

Hope to help

Giuseppe

7 REPLIES

Re: OSPF E1 & E2 comparison

Hi Sairam

E2 which is the default external type will give you metric calculation upto the point of redistributrion and no further, ie the metric will not increase as is traverses new links. In otherwords, it is a metric from the ASBR towards the destination. E1 however will include all additional hops within the path so is a more accurate measure. If at the point of redistribution, there are a lot more hops, then adding the metric as it passes the new routers is a wise choice, otherwise further down the line, a 64K serial link will be the same as a 10G link which is quite obviously not what you want.

Also type 1 is prefered over type 2 when metrics match

1. Intra-area routes.

2. Inter-area routes.

3. External Type-1 routes.

4. External Type-2 routes.

Hall of Fame Super Silver

Re: OSPF E1 & E2 comparison

Hello Sairam,

O E1 are recommended when multiple ASBRs exist and you want to be sure that the best path to the exit point is used.

O E1 routes are always preferred over O E2 routes regardless of seed metric.

O E1 are seen as more complete information because the seed metric is added to the path cost to reach ASBR.

if you have only one exit point for example for internet access using O E2 for the default route is enough.

So usually the design discriminator is the number of ASBRs / exit points in the network.

I personally prefer O E1 routes to have more control, but when inter-area routing is involved to reach the ASBR a potential for suboptimal paths still exists. (again if multiple ABRs connect the local area to the backbone).

Hope to help

Giuseppe

New Member

Re: OSPF E1 & E2 comparison

Hi Giuseppe & adam,

Thanks. But the query is not about the order of preference between E1 & E2. I am cent percent sure that E1 is prefered over E2. Also I know how the metric is calculated in both the cases.

But the confusion started when i noticed the parameter "forwarding metric" in E2 route (in E1 it is not found as it is taking complete path from destination to host including cost between ASBR & destination prefix)

Putting in simple words, when 2 different path exist I found

1. E1 takes the least cost calculated from HOST to DESTINATION

2. E2 also takes the least cost calculated from HOST to DESTINATION using the FORWARD METRIC even though the metric shown as 20 by default.

I feel both does the same job (E1 uses composite metric & E2 uses forwarding metric).

If so what is the difference. I hope I managed to express the doubt clearly.

Hope you will help me. Thanks for your valuable time

sairam

Hall of Fame Super Silver

Re: OSPF E1 & E2 comparison

Hello Sairam,

first O E2 seed metric should be considered.

then, when both routes have the same O E2 seed metric lowest forwarding metric wins.

see

Type 2

external paths advertising equal type 2 metrics are

compared by looking at the distance to the forwarding

addresses.

http://www.ietf.org/rfc/rfc2328.txt

section 16.4.0

this is logical and explains the difference between O E1 and O E2

Hope to help

Giuseppe

Hall of Fame Super Bronze

Re: OSPF E1 & E2 comparison

But I donot understand when to use which Metric type?

When the router needs to compare same routes with same route-type. For instance; multiple ASBRs sending the same route (E2, for instance). You can design one ASBR applying a forwarding metric of 30 while the other ASBR applying a forwarding metric of 20. You can then perform traffic engineering of redistributed routes to draw traffic towards a specified router.

So, you have several choices on the design. Say you have 3 ASBR and you want ASBR Router A to draw the routes and ASBR Router B and C to act as secondaries.

ASBR Router A will redistribute the routes as E1 and ASBR Router B and C will be redistributing as E2 with forwarding metric 20 and 30 respectively. This design will meet your requirements.

Regards

Edison.

New Member

Re: OSPF E1 & E2 comparison

Hall of Fame Super Bronze

Re: OSPF E1 & E2 comparison

Brandon,

I recommend opening a new thread.

Regards

Edison.

2779
Views
4
Helpful
7
Replies