EIGRP failover problems

Mar 21, 2005 2 Replies
EIGRP failover problems open original image

I need help solving an eigrp problem/design problem. I have an eigrp network for redundancy but it is not working the way I need it to. We have 2 CORE routers (1 and 2) at two different locations that connect back to vendors mainframe. We have no control over the CORE routers (none eigrp) or the mainframe. Our routers (A and B) are connect the core routers via Ethernet so they are on the same network. ROUTERA has a 256K frame circuit to ROUTERB and to ROUTERC and is considered a hub router. ROUTERB has 256K frame to both ROUTERA and ROUTERC as well and is considered a hub as well. We have 3 other STUB routers like ROUTERC but to make the drawing simple I'm not listing them. ROUTERA has a static route 10.160.2.0 255.255.255.0 out to CORE1's ethernet (this network is some network on the other side of the core routers). ROUTERB has a same static router to CORE2's ethernet but with a 250 administrative distance. If ROUTERA goes down, everything switches over to ROUTERB and works fine. OK, Here is my problem. If I lose connection to CORE1 from ROUTERA but ROUTERA stays up then the static routes redistributed by eigrp are still valid so the eigrp will not learn a new router to ROUTERB to get to the



10.160.2.0 network. I don't think the vendors who own the CORE will not config eigrp on them.

How can I get around this? Can I setup a loopback and monitor that somehow?



Please help?


CORE1----------ROUTERA--------------| | | | WAN LINK | ROUTERC/STUB | | | CORE2----------ROUTERB--------------|



router eigrp 100 network 10.17.16.0 0.0.3.255 network 172.16.53.0 .0.0.0.255 no auto-summary redistribute static


Hi Michael,

I had a similar problem a while ago and one of the recommendations was to look at Object Tracking although I am not sure how reliable the technology is. Do a search on

formatting link
for Ping Based Routing, there should be a few documents.

The document I looked at was:

formatting link
Good luck

Regards

Darren

Michael Letchworth wrote: [snip]

It's a common problem. Static routes on shared media have a tough time of determining that it's next hop is not reachable.

Ping based routing may help, but it's pretty bleeding edge. The right solution is to run a routing protocol that you two can mutally agree upon (like BGP). But if that is not possible, then adding the next hop interface and address may help you.

On router-A ip route 10.160.2.0 255.255.255.0 FA0/0 172.16.1.1

where FA0/0 is the interface facing CORE01 router and 172.16.1.1 is the partners IP on CORE01 facing router A.

See

formatting link
for more info.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required