Eigrp : holding time expired and it is taken approx. 2 minutes to re-establish connection.

[jomo frank]

Hello Expert,

I am experiencing an issue with my eigrp hold time, when tunnel goes down it takes move 1min 30 s to re-establish 

connection.

I would welcome any guidance to help resolve this issue.

Undermentioned is a subset of the log file obtain for the subject router.

Mar 23 10:45:18 Caracas: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.137.10 (Tunnel2) is down: holding time expired
*Mar 23 10:45:25 Caracas: %LINEPROTO-5-UPDOWN: Line protocol on Interface Tunnel2, changed state to down
*Mar 23 10:47:20 Caracas: %LINEPROTO-5-UPDOWN: Line protocol on Interface Tunnel2, changed state to up
*Mar 23 10:47:20 Caracas: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.137.10 (Tunnel2) is up: new adjacency

Regards

jomo

[paul driver]

Hello

The reason for that the spoke rtr is stcuk in active (SIA) due to  dual process iniciating
Change the eigrp spokes to eigrp stubs.

router eigrp xx
eigrp stub

[jomo frank]

Hello Paul,

I unsure how to do this 

I include the eigrp networks statement for guidance 

router eigrp 100
network 10.179.192.10 0.0.0.255
network 192.168.19.0
network 192.168.27.0
network 192.168.124.0
network 192.168.137.0

Regards

[Georg Pauwen]

Hello,

what kind of network topology is that (e.g. DMVPN) ? What are the hello and hold timers currently configured for EIGRP ?

[jomo frank]

Hello Greorg

The network topology is site to site vpn.

EIGRP-IPv4 Neighbors for AS(100)
H Address Interface Hold Uptime SRTT RTO Q Seq
(sec) (ms) Cnt Num
1 192.168.137.10 Tu2 10 01:53:08 2 100 0 93429
0 192.168.124.10 Tu1 14 04:11:22 9 132 0 93422
Amazonia_Mall#

[Georg Pauwen]

Hello,

make sure the hello and hold timers on both end routers match (not only for the tunnels, but also for the underlying physical interfaces).

Actually, can you post the full running configurations of both routers ?

[Georg Pauwen]

Hello,

--> The network topology is site to site vpn.

Does that mean a site to site VPN with SVTIs, or a DMVPN ? As requested, post the full configs of both routers. What underlying WAN technology is being used ?

[paul driver]

Hello
I don’t think your issue has anything to do with eigrp timers its sounds like its due to the eigrp dual process initiating as a feasible successor cannot be found as such each rtr will query its neighbours for a better feasible path until a reply is received and until then basically the rtr would be stuck in active (SIA) so convergence will be slower than expected?

in a properly designed eigrp network you shouldn’t really expect to see dual 

sh ip eigrp topology <x.x.x.x >  check for feasible successors

Upon a failover post the output from:
sh ip eigrp topology active  you should a “r” next to the prefix meaning its waiting for a reply 

So as its seem you have a hub- spoke eigrp dmvpn topology if you set your spoke eigrp rtrs as stubs then they won’t participate in the dual process as such you should see a much better convergence.

[paul driver]

Upon a failover post the output from:
sh ip eigrp topology active  you should a “r” next to the prefix meaning its waiting for a reply 

So as its seem you have a hub- spoke eigrp dmvpn topology if you set your spoke eigrp rtrs as stubs then they won’t participate in the dual process as such you should see a much better convergence.