Eigrp over GRE IPSEC tunnels

nhaider · ‎04-24-2002

Just for clarification does Eigrp weigh down GRE IPSEC tunnels by default. I am using the tunnels as a backup to Frame T1's at our POP sites. The tunnels are directly connected as far as EIGRP is concerned and the Upstream connection is a DS3 to our ISP. In my testing I have tested this and without the need to weigh down anything the frame T1 route is prefered, which is exactly what I want. I have also introduced multiple hops in the test bed for the frame network just to be sure that I will not need to weigh down the tunnel. In my testing I introduced 7 hops over the frame network. And the tunnels where using an upstream Ethernet link that was directly connected (tunnel source and destination) to the router that the test network is off of. Eigrp would still prefer the Frame Route over the tunnel. Does anyone know if the tunnel is naturaly weighed down? The Cisco TAC engineer stated that they weren't, but that leaves me questioning why the tunnel isn't prefered in my test bed?

Thanks,

ciscomoderator · ‎05-02-2002

Since there has been no response to your post, it appears to be either too complex or too rare an issue for other forum members to assist you. If you don't get a suitable response to your post, you may wish to review our resources at the online Technical Assistance Center (http://www.cisco.com/tac) or speak with a TAC engineer. You can open a TAC case online at http://www.cisco.com/tac/caseopen

If anyone else in the forum has some advice, please reply to this thread.

Thank you for posting.

tmehok · ‎05-02-2002

If I understand your question correct. One thing that can account for this is the bandwidth statement under the interface. By default it is 9Kbps on a GRE tunnel interface. You can also issue the "show eigrp topolgy" command to see the metric of every router in the EIGRP topology table.

Hope this helps.

nhaider · ‎05-02-2002

I have no bandwidth statements issued on any of the interfaces, so It should be the default. I have checked the eigrp topology table and it doesn't even show the tunnel interfaces as feasible successors in the table, which is odd. But when I initiate the failure on the interface that the T1 is on (primary interface) Eigrp does converge and use's the tunnel interface as the new route. The configuration is working as I want it to; it is using the tunnel as purely a backup route. But out of curiosity, I am still in search of why or if Cisco weighs down tunnels by default? I have tried unsuccessfuly to get an answer out of Cisco's tech support, but none of them seem too sure about their answers, and the final answer that I have recieved is that as far as EIGRP is concerned it shouldn't weigh down tunnel interfaces by default. If that is true then why is my tests showing the opposite? Anyway's the closest theory that I can come up with is that IPSEC and GRE encapsulation add a considerable amount of overhead to traffic and affect EIGRP's routing decision's. Thanks for the help, if you find out any new info pass it my way. Much Appreciated.

Nabeel

jerry.roy · ‎05-02-2002

It seems to me you WOULD want the bandwidth statements on your Frame links. Doesn't the router automatically start to negotiate from the fastest speed available down to the slowest before it sends frames. If you are at maximum (I believe 1.536 on Frame) then there would be no issue. But if your speed is slower (say 384), it has to negotiate every frame to reach the appropriate speed before sending, which would slow you down. I am not sure this is relevant but if EIGRP only has 9 kbps, the tunnel interface would always be the preferred secondary path.