EIGRP Unequal Cost Traffic Sharing Problem

[Ali Razavi]

Hey everyone,

I've configured a router to prioritize among its three paths to the 150.1.1.0/24 network.  I can see that the topology table reflects the proper metrics after the changes were made.  However, only two routes are selected.  Initially after configuring the traffic-share min across-interfaces command, I received the following warning message:

%FIB-4-UNEQUAL: Range of unequal path weightings too large for prefix 150.1.1.0/24. Some available paths may not be used.

I've pasted the configs and routing table information below.  Please let me know if I'm on the right track.

R3#show run | section eigrp

router eigrp 200

variance 2

traffic-share min across-interfaces

offset-list 0 in 1000000 Serial1/1

network 10.0.0.4 0.0.0.3

network 10.0.0.8 0.0.0.3

network 10.0.0.12 0.0.0.3

network 150.3.3.0 0.0.0.255

no auto-summary

R3#show ip route 150.1.1.0

Routing entry for 150.1.1.0/24

  Known via "eigrp 200", distance 170, metric 2172416, type external

  Redistributing via eigrp 200

  Last update from 10.0.0.5 on Serial1/1, 00:00:08 ago

  Routing Descriptor Blocks:

  * 10.0.0.14, from 10.0.0.14, 00:00:08 ago, via Serial1/2

      Route metric is 2172416, traffic share count is 1

      Total delay is 20100 microseconds, minimum bandwidth is 1544 Kbit

      Reliability 255/255, minimum MTU 1500 bytes

      Loading 1/255, Hops 1

    10.0.0.5, from 10.0.0.5, 00:00:08 ago, via Serial1/1

      Route metric is 3172416, traffic share count is 0

      Total delay is 59162 microseconds, minimum bandwidth is 1544 Kbit

      Reliability 255/255, minimum MTU 1500 bytes

      Loading 1/255, Hops 1

R3#show ip eigrp topology 150.1.1.0 255.255.255.0

IP-EIGRP (AS 200): Topology entry for 150.1.1.0/24

  State is Passive, Query origin flag is 1, 1 Successor(s), FD is 2172416

  Routing Descriptor Blocks:

  10.0.0.14 (Serial1/2), from 10.0.0.14, Send flag is 0x0

      Composite metric is (2172416/28160), Route is External

      Vector metric:

        Minimum bandwidth is 1544 Kbit

        Total delay is 20100 microseconds

        Reliability is 255/255

        Load is 1/255

        Minimum MTU is 1500

        Hop count is 1

      External data:

        Originating router is 4.4.4.4

        AS number of route is 100

        External protocol is EIGRP, external metric is 0

        Administrator tag is 0 (0x00000000)

  10.0.0.5 (Serial1/1), from 10.0.0.5, Send flag is 0x0

      Composite metric is (3172416/1028160), Route is External

      Vector metric:

        Minimum bandwidth is 1544 Kbit

        Total delay is 59162 microseconds

        Reliability is 255/255

        Load is 1/255

        Minimum MTU is 1500

        Hop count is 1

      External data:

        Originating router is 1.1.1.1

        AS number of route is 100

        External protocol is EIGRP, external metric is 0

        Administrator tag is 0 (0x00000000)

  10.0.0.9 (Serial1/0), from 10.0.0.9, Send flag is 0x0

      Composite metric is (2684416/2172416), Route is External

      Vector metric:

        Minimum bandwidth is 1544 Kbit

        Total delay is 40100 microseconds

        Reliability is 255/255

        Load is 1/255

        Minimum MTU is 1500

        Hop count is 2

      External data:

        Originating router is 1.1.1.1

        AS number of route is 100

        External protocol is EIGRP, external metric is 0

        Administrator tag is 0 (0x00000000)

Thanks,

Ali

[Richard Burts · Accepted solution]

Ali

One of the requirements for variance to work in EIGRP is that the reported distance from a neighbor must be less than the Feasible Distance of the successor route. Looking in your output from show ip eigrp topology we see

FD is 2172416

So this is the value to compare with the reported distance from the neighbors.

The first neighbor reports the lowest distance and its route is the successor route using Serial 1/2

The second neighbor reports 1028160 as the distance. This is less than the FD and so the route through Serial 1/1 is also used.

The third neighbor reports 2172416. This is not less than FD and so this route is not used by variance.

So I believe that you are on the right track. It is just that one of the routes does not qualify to be used under variance.

HTH

Rick

[Ali Razavi]

That was right on the money Rick.  A soon as I decreased the interface delay value on 10.0.0.9 serial1/0, the route showed up in the RIB.  I apologize for the late replay as I've been ramping up for my ROUTE exam coming up in the next couple of weeks. 

[Richard Burts]

Ali

I am glad that my suggestion pointed you in the right direction and that you were able to resolve the issue. Thank you for using the rating system to mark the question as answered. It makes the forum more useful when people can read a question and can know that a solution was found. Your marking has contributed to this process. And this is a pretty subtle issue about variance which may be of help to other readers.

HTH

Rick

[paul driver]

Hello

I see you have both  the variance and offset configured! I was on the understanding one of the other could be used when L/B, with the offset utilised when the variance isnt.

In any case Your composite metric is calculated fine but in your offset statement you are not pointing it to the best lowest interface which according to your eigrp topology is Serial1/2 not Serial1/1

res

Paul

Please don't forget to rate any posts that have been helpful.

Thanks.