cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
714
Views
0
Helpful
4
Replies

EIGRP Tunnels dropped - Reason Unknown

Ricky S
Level 3
Level 3

Hi everyone,

We have a DMVPN environment running EIGRP between all routers. Each spoke builds a Tunnel0 and Tunnel1 to HUB0 and HUB1 respectively with Tunnel0 being the primary path for everyone to communicate and Tunnel1 as Secondary to be used if HUB0 fails. Also each router has Tunnel2 that connects them to HUB2 over 3G cellular network in case of a local high speed internet connection failure.

Sunday morning around 3AM EST one of our routers dropped both Tunnel0 and Tunnel1, which caused Tunnel2 to come up. Then 5 hours later, Tunnel1 came up followed by Tunnel0 which came up after 16 hours. Since Tunnel0 and Tunnel1 are tethered to separate HUB routers, I assumed the issue must've been with the spoke router which is a Cisco 2911 running c2900-universalk9-mz.SPA.150-1.M4.bin IOS image.

Also no other spokes experienced a similar issue at that time and both HUB0 and HUB1 were connected to all other sites at all times.

Any ideas as to what might have been the reason for DUAL to drop both EIGRP neighbors all of a suddent?

Thanks

1 Accepted Solution

Accepted Solutions

Are there logs available from the remote router during that time? If so they might have helpful information.

Also are the logs from hub0  and hub1 during this time available. They might have some entries that might shed light on why the remote router lost its neighbor relationship.

Am I correct in understanding that the remote router lost both upstream neighbors at the same time and that one neighbor came back after 5 hours and the other neighbor came back after 16 hours? A pattern like that makes me wonder if there were some issue in the provider of the local high speed connection.

HTH

Rick

HTH

Rick

View solution in original post

4 Replies 4

hello
Sounds like that router had no feasible successor for the hubs 0/1 and that's why it failed over to hub2

Sh log
Sh ip eigrp neighbour
Sh ip route eigrp

Res
paul


Sent from Cisco Technical Support iPad App


Please rate and mark as an accepted solution if you have found any of the information provided useful.
This then could assist others on these forums to find a valuable answer and broadens the community’s global network.

Kind Regards
Paul

Hub1 (10.10.201.1) is the feasible successor for all spokes. See sample eigrp topology output below.

RTRTAL001#sh ip eigrp top

P 172.18.171.0/24, 1 successors, FD is 1333760

       10.10.200.71 via 10.10.200.1 (1333760/1308160), Tunnel0

       10.10.201.71 via 10.10.201.1 (1335040/1308160), Tunnel1

P 192.168.226.0/24, 1 successors, FD is 1331456

       10.10.200.223 via 10.10.200.1 (1331456/1305088), Tunnel0

       10.10.201.223 via 10.10.201.1 (1332736/1305088), Tunnel1

P 10.10.101.0/28, 1 successors, FD is 1331456

       10.10.200.204 via 10.10.200.1 (1331456/1305088), Tunnel0

       10.10.201.204 via 10.10.201.1 (1332736/1305088), Tunnel1

P 192.168.155.207/32, 1 successors, FD is 1459200

       10.10.200.207 via 10.10.200.1 (1459200/1432832), Tunnel0

       10.10.201.207 via 10.10.201.1 (1460480/1432832), Tunnel1

Spokes use Tunnel1 as a secondary path if Tunnel0 drops, which should only happen if Hub0 (Tunnel0 Source) dies. Both Tunnel0 and Tunne1 share a single high speed internet connection and both should go down at the same time ONLY if local high speed connection goes down.

Are there logs available from the remote router during that time? If so they might have helpful information.

Also are the logs from hub0  and hub1 during this time available. They might have some entries that might shed light on why the remote router lost its neighbor relationship.

Am I correct in understanding that the remote router lost both upstream neighbors at the same time and that one neighbor came back after 5 hours and the other neighbor came back after 16 hours? A pattern like that makes me wonder if there were some issue in the provider of the local high speed connection.

HTH

Rick

HTH

Rick

Hi Rick, I actually checked the logs yesterday and it seems the hello timer ran out on both sides which made me wonder if there was an outage on the internet. I then called the provider and they said they (ofcourse) didn't have any issues on their side. But you might be right as it does smell like something took a bump on the internet side.

Thanks