I've got a problem of configuring the preferred path via EIGRP. I have two circuits one set with "bandwidth 20000" - link1 and the other with "bandwidth 10000" - link2. EIGRP is naturally choosing link1 as the preferred path to the destination of 10.25.10.0. What I want to do is make link2 the preferred link, how can I achieve this in eigrp? The two links are at two separate sites.
You should never use the bandwidth command to artificially make EIGRP prefer one route to another. By setting a bandwidth parameter on an interface, you are influencing the QoS operations (if there are any configured), the Tx queue length and also, the EIGRP is allowed to reserve as much as 50% of the interface's preconfigured bandwidth for its traffic. By setting the bandwidth to an inappropriate (i.e. untrue) value, you can cause the EIGRP to saturate the interface. If possible, the bandwidth setting should always be set to the true bandwidth of the interface.
If you want to influence EIGRP metrics according to your needs, use the delay command to set up a delay value that is used in the EIGRP metric calculation. The delay parameter does not influence any other interface or router operations, as opposed to the bandwidth setting.
So, set the delay on your link1 to be significantly higher that the delay on link2. The router should then start to prefer the link2 - of course, if a particular destination is reachable through both links.
Thanks for that detailed reply, very much appreciated. As it happens Link1 just got upgraded to 20Mb and Link2 is 10Mb (Still to be upgraded). The problem is that I'm having slight issues with link1 so I want to make link2 the primary line.
If I use delay won't that affect my whole eigrp routing as it's a router command. Can you please give me an example of how to do this on the interface?
If the bandwidth statements are an accurate reflection of the capacity of the links I would leave them. If they are not accurate then I would remove them.
If you want to modify the metric of some routes but not all routes on an interface then there is an alternative to consider beside using the delay (which will impact all routes through that interface). You can configure an offset list to manipulate the metric. You can apply the offset list to inbound or to outbound traffic as is appropriate. And the offset list can use an access list to specify the specific routes whose metric you want to increase. All routes not permitted in the access list will continue to have their original metric.
A great response indeed, thank you! Yes, the offset lists allow to perform metric manipulation on a more fine-grained level. Spot on!
Personally, I try to never omit the bandwidth command from interfaces. The built-in IOS values may be far away from the actual capacity of the interface and using them might be even worse than, say, make a safe underestimated guess.
[toc:faq]The ProblemOn traditional switches whenever we have a trunk
interface we use the VLAN tag to demultiplex the VLANs. The switch needs
to determine which MAC Address table to look in for a forwarding
decision. To do this we require the switch to do...
[toc:faq]Introduction:Netdr is a tool available on a RSP720, Sup720 or
Sup32 that allows one to capture packets on the RP or SP inband. The
netdr command can be used to capture both Tx and Rx packets in the
software switching path. This is not a substitut...
IntroductionOSPF, being a link-state protocol, allows for every router
in the network to know of every link and OSPF speaker in the entire
network. From this picture each router independently runs the Shortest
Path First (SPF) algorithm to determine the b...