In the diagram and configs it's not clear where exactly nw 1, 2 and 3 are, you cut that part of the config out. I know changing the bandwidth works well and quite fast. I have done so in the past for a customer running a 6M ATM pipe and T1 frame relay for redundancy. We needed to work on the router that terminated the ATM so I simply changed the bandwidth to 100 and with a couple seconds all the traffic reverted to frame.
If you really only want to test whether changing the bandwidth works simply connect two routers back to back with two serial connections, equal clocks, no distribute lists and then change the bandwidth. Dave V patankar wrote: > > Hi, > > I have a very simple three router scenario, using 12.0(10) IOS on 1200 > platform, and configured to run EIGRP. > > > RB > S0/ \ s1 > (nw 1) bw=512 / \ bw=1024 (nw 2) > s1 / \ s1 > RC ------- RA > bw=512 > nw 3 > > All the three links are serial links connected by using a back-to-back X.21 > cables. > > Looking at the above diagram you would expect the following > > RB > > all traffic will be send over interface S1 to reach nw 3, and no load > balancing should take place. > > RC > > Two routes, to reach network nw 2, load balancing should take place, and two > routes should be seen in the routing tabel. > > RA > > All traffic should leave via interface S1 to reach the nw 1 and no load > balancing should take place. > > Problem: > > All the routers are showing that two paths exist to reach the destination > networks. > > Please check the outputs: > ----------------------------------- > ra# > interface Serial0 > bandwidth 512 > ip address 172.16.91.21 255.255.255.252 > no fair-queue > ! > interface Serial1 > bandwidth 1024 > ip address 172.16.91.25 255.255.255.252 > ! > router eigrp 1 > passive-interface Ethernet0 > network 172.16.0.0 > distribute-list 1 out Serial0 > distribute-list 1 out Serial1 > auto-summary > no eigrp log-neighbor-changes > ! > D 172.16.91.28/30 [90/6023936] via 172.16.91.26, Serial1 > [90/6023936] via 172.16.91.22, Serial0 > C 172.16.91.24/30 is directly connected, Serial1 > ra# > > ------------------------------------ > interface Serial0 > bandwidth 512 > ip address 172.16.91.30 255.255.255.252 > no ip directed-broadcast > no ip mroute-cache > no fair-queue > clockrate 38400 > ! > interface Serial1 > bandwidth 1024 > ip address 172.16.91.26 255.255.255.252 > no ip directed-broadcast > clockrate 38400 > ! > router eigrp 1 > passive-interface Ethernet0 > network 172.16.0.0 > distribute-list 1 out Serial0 > distribute-list 1 out Serial1 > rb# > > 172.16.0.0/16 is variably subnetted, 4 subnets, 2 masks > C 172.16.15.0/24 is directly connected, Ethernet0 > D 172.16.91.20/30 [90/6023936] via 172.16.91.25, Serial1 > [90/6023936] via 172.16.91.29, Serial0 > C 172.16.91.28/30 is directly connected, Serial0 > C 172.16.91.24/30 is directly connected, Serial1 > ------------------------------------- > rc# > > interface Serial0 > bandwidth 512 > ip address 172.16.91.22 255.255.255.252 > no ip directed-broadcast > no ip mroute-cache > delay 2000 > no fair-queue > clockrate 38400 > ! > interface Serial1 > bandwidth 512 > ip address 172.16.91.29 255.255.255.252 > no ip directed-broadcast > ! > router eigrp 1 > passive-interface Ethernet0 > network 172.16.0.0 > distribute-list 1 out Serial0 > distribute-list 1 out Serial1 > ! > > Gateway of last resort is not set > > 172.16.0.0/16 is variably subnetted, 4 subnets, 2 masks > C 172.16.14.0/24 is directly connected, Ethernet0 > C 172.16.91.20/30 is directly connected, Serial0 > C 172.16.91.28/30 is directly connected, Serial1 > D 172.16.91.24/30 [90/6023936] via 172.16.91.21, Serial0 > [90/6023936] via 172.16.91.30, Serial1 > rc# > ------------------------------------- > P.S. By changing the delay parameter on rb for inteface s0 (20000 -> 20001) > will for the traffic to go via interface s1. > > Has anyone seen this behaviour before, is it normal for Eigrp to behave in > such a way ? > > Note: All the Ethernet networks are 172.16.15.0, 172.16.14.0 and 172.16.90.0 > all using a 24 bit mask, the distribute list was used to stop these updates > from entering the eigrp routing table. > > The main focus here is to find out why the bandwidth command when used in > EIGRP is not effective. I hope I have not confussed anyone. > > Thanks in Advance. > vp -- David Madland Sr. Network Engineer CCIE# 2016 Qwest Communications Int. Inc. [EMAIL PROTECTED] 612-664-3367 "Emotion should reflect reason not guide it" Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=38807&t=38744 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
