Eigrp bandwitch change behaviour

[Arie]

I have three routers connected each other. I tried to influence on eigrp metrics modificating the bandwith of all interfaces. I mark two of three path as 15000 and one for 64. However the situation is as before, nothing has been changed when i looked up into route tables of all routers. Is it now enough to change bandwith to force metric to be recalculated by eigrp? As i make big diffrence between paths bandwitch it should pick the one where 15000.

[willwetherman]

Hi

EIGRP only considers the smallest bandwidth along the path to a destination prefix when calculating its metric so increasing the bandwidth on an interface may not result in a metric change due to a smaller bandwidth somewhere else along the path. When you changed the bandwidth did you do it on all interfaces on one of the paths between the router's RIB that you are looking at and the destination prefix?

As changing the interface bandwidth can impact other router operations such as QoS, its advised not to modify it to try and influence EIGRP path selection. Instead it is advised to modify interface delay to manipulate EIGRP metrics.

Hope this helps

[Arie]

yes, i changed all bandwitch interfaces so i made one way very slow (obviously in theory) and second way  very fast. That didn't force eigrp to switch trace. Are you adicing to change delay as well on slowest path as well?

[willwetherman]

Hmm that is odd as EIGRP should automatically update the metric if thats the case.

Can you provide a diagram of the topology and the the output of 'show ip eigrp topology x.x.x.x/y' for one of the prefixes that you are looking at?

[Julio E. Moisa]

Hi Arie

Try to change the bandwidth on the outgoing interface, Could you please your diagram?

[Julio E. Moisa]

EIGRP Metric = 256*(Bandwidth + Delay)

Where

• Bandwidth = 10000000/bandwidth(i), where bandwidth(i) is the least bandwidth of all outgoing interfaces on the route to the destination network represented in kilobits.

• Delay = delay(i) where delay(i) is the sum of the delays configured on the interfaces, on the route to the
destination network, in tens of microseconds.

[Julio E. Moisa]

An example:

I have the following topology, R4 is advertising the loopback0 4.4.4.4/32, If the metrics (Bandwidth and delay) are not being modified I will have links with equal metric:

R1#sh ip eigrp topology 4.4.4.4 255.255.255.255
IP-EIGRP (AS 100): Topology entry for 4.4.4.4/32
State is Passive, Query origin flag is 1, 2 Successor(s), FD is 435200
Routing Descriptor Blocks:
10.12.0.2 (FastEthernet0/0), from 10.12.0.2, Send flag is 0x0
Composite metric is (435200/409600), Route is Internal
Vector metric:
Minimum bandwidth is 10000 Kbit
Total delay is 7000 microseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 2
10.13.0.3 (FastEthernet0/1), from 10.13.0.3, Send flag is 0x0
Composite metric is (435200/409600), Route is Internal
Vector metric:
Minimum bandwidth is 10000 Kbit
Total delay is 7000 microseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 2
R1#
R1#
R1#
R1#
R1#sh ip route 4.4.4.4
Routing entry for 4.4.4.4/32
Known via "eigrp 100", distance 90, metric 435200, type internal
Redistributing via eigrp 100
Last update from 10.13.0.3 on FastEthernet0/1, 00:05:00 ago
Routing Descriptor Blocks:
10.13.0.3, from 10.13.0.3, 00:05:00 ago, via FastEthernet0/1
Route metric is 435200, traffic share count is 1
Total delay is 7000 microseconds, minimum bandwidth is 10000 Kbit
Reliability 255/255, minimum MTU 1500 bytes
Loading 1/255, Hops 2
* 10.12.0.2, from 10.12.0.2, 00:05:00 ago, via FastEthernet0/0
Route metric is 435200, traffic share count is 1
Total delay is 7000 microseconds, minimum bandwidth is 10000 Kbit
Reliability 255/255, minimum MTU 1500 bytes
Loading 1/255, Hops 2

R1#

You can verify the path taken by R1 to reach R4's loopback using:

R1#sh ip cef exact-route 1.1.1.1 4.4.4.4
1.1.1.1 -> 4.4.4.4 =>IP adj out of FastEthernet0/0, addr 10.12.0.2
R1#
R1#sh ip cef exact-route 1.1.1.1 4.4.4.4
1.1.1.1 -> 4.4.4.4 =>IP adj out of FastEthernet0/0, addr 10.12.0.2

Now Let's modify the bandwidth value on the inteface facing to R4 on R2

* Note, you can see the bandwidth using: show interface Fa0/1  for example

interface Fa0/1
bandwidth 100  (decreased)

Now we can check the change of path from R1

R1#sh ip cef exact-route 1.1.1.1 4.4.4.4
1.1.1.1 -> 4.4.4.4 =>IP adj out of FastEthernet0/1, addr 10.13.0.3
R1#sh ip cef exact-route 1.1.1.1 4.4.4.4
1.1.1.1 -> 4.4.4.4 =>IP adj out of FastEthernet0/1, addr 10.13.0.3

R1#sh ip eigrp topology 4.4.4.4 255.255.255.255
EIGRP-IPv4 Topology Entry for AS(1)/ID(1.1.1.1) for 4.4.4.4/32
State is Passive, Query origin flag is 1, 1 Successor(s), FD is 435200
Descriptor Blocks:
10.13.0.3 (FastEthernet0/1), from 10.13.0.3, Send flag is 0x0
Composite metric is (435200/409600), route is Internal
Vector metric:
Minimum bandwidth is 10000 Kbit
Total delay is 7000 microseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 2
Originating router is 4.4.4.4
R1#

Hope it is useful 

:-)