EIGRP Lab Loopback Issue

[ALI12]

Hi Everyone,

I recently set up a lab in GNS3 to simulate the EIGRP routing protocol, connecting two ISPs across two sites.
However, I’m encountering a loopback issue based on the trace outputs (the Config files in attachments).

I would appreciate it if anyone could help identify what configuration might be missing or incorrect that’s causing this loopback issue.

Thank you in advance for your support!

========================================
Core-A Trace:
Core-A#traceroute 172.16.4.1 so lo 4
Type escape sequence to abort.
Tracing the route to 172.16.4.1
VRF info: (vrf in name/id, vrf out name/id)
1 10.40.40.1 60 msec
10.40.40.2 60 msec
10.40.40.1 44 msec
2 10.222.0.2 80 msec
10.111.0.2 60 msec
10.222.0.2 92 msec
3 172.18.0.3 136 msec 132 msec 140 msec
Core-A#
=================================================
Core-B Trace:
Core-B#traceroute 10.66.1.1 so lo 1
Type escape sequence to abort.
Tracing the route to 10.66.1.1
VRF info: (vrf in name/id, vrf out name/id)
1 172.18.0.1 72 msec
172.18.0.2 60 msec
172.18.0.1 68 msec
2 10.222.0.1 84 msec
10.111.0.1 84 msec
10.222.0.1 128 msec
3 10.40.40.3 100 msec 88 msec 140 msec
Core-B#
====================================================
====================================================
Core-A
router eigrp 10
network 10.66.4.0 0.0.0.255
network 10.66.3.0 0.0.0.255
network 10.66.2.0 0.0.0.255
network 10.66.1.0 0.0.0.255
network 10.40.40.0 0.0.0.7
----------------------------------
ISP-1-A:
router eigrp 10
address-family ipv4 vrf vpn
autonomous-system 10
network 10.40.40.0 0.0.0.7
network 10.111.0.0 0.0.0.3
--------------------------------
ISP-2-A:
router eigrp 10
address-family ipv4 vrf vpn
autonomous-system 10
network 10.40.40.0 0.0.0.7
network 10.222.0.0 0.0.0.3
++++++++++++++++++++++++++++++++++++
Core-B:
router eigrp 10
network 172.16.4.0 0.0.0.255
network 172.16.3.0 0.0.0.255
network 172.16.2.0 0.0.0.255
network 172.16.1.0 0.0.0.255
network 172.18.0.0 0.0.0.7
--------------------------------
ISP-1-B:
router eigrp 10
address-family ipv4 vrf vpn
autonomous-system 10
network 172.18.0.0 0.0.0.7
network 10.111.0.0 0.0.0.3
----------------------------------
ISP-2-B:
router eigrp 10
address-family ipv4 vrf vpn
autonomous-system 10
network 172.18.0.0 0.0.0.7
network 10.222.0.0 0.0.0.3

[Flavio Miranda]

@ALI12 

 Not sure if you got your problem but I see that the traffic is taking both path. I would try to interfere and send the traffic to one path only. You can do this using more than one command. You can start by using

ISP-1-A:
router eigrp 10
address-family ipv4 vrf vpn
autonomous-system 10
network 10.40.40.0 0.0.0.7
network 10.111.0.0 0.0.0.3

variance 2

[ALI12]

thanks @Flavio Miranda for your input,

I followed your suggestion, but unfortunately, there seems to be no change.

I'm still getting the same result, as shown in the PrintScreen

[Flavio Miranda]

@ALI12 

 There are other ways to influence the packet path

Use this method instead

ISP-1-A:
router eigrp 10
address-family ipv4 vrf vpn
autonomous-system 10
network 10.40.40.0 0.0.0.7
network 10.111.0.0 0.0.0.3

maximum-paths 1

ISP-1-B:
router eigrp 10
address-family ipv4 vrf vpn
autonomous-system 10
network 172.18.0.0 0.0.0.7
network 10.111.0.0 0.0.0.3

maximum-paths 1

You can also change the bandwidth on the interface of the ISP router. Just do it on both sides.

[paul driver]

Hello
Append the following and test again.

CoreA-B 
Int ethx/x
no ip split-horizon eigrp x

[ALI12]

Thank you @paul driver for your suggestion ,

I’ve implemented the changes, but unfortunately, I’m still seeing the same result.

[paul driver]

Hello
tbh i do t see an issue apart each core is using the same interface to route in/out so you disabled split horizon which was required - now what i now see is ecmp occurring.
which you would see if the path metric between each core loopbacks is calculated the same - which you influence if you desire to accommodate one path over another by increasing the interface delay  on the isp links facing each core 

on a side note depending on what ios you are using you may need to disable auto summarisation under the eigrp process
Router eigrp x
No auto summary  

[ALI12]

@paul driver 

I’ve disabled auto summarization, but the issue persists.

The loopback problem was resolved by adjusting the distance values on both Core-A and Core-B as follows:

Core-A:
router eigrp 10
distance 90 10.40.40.1 0.0.0.0
distance 91 10.40.40.2 0.0.0.0

Core-B:
router eigrp 10
distance 90 172.18.0.1 0.0.0.0
distance 91 172.18.0.2 0.0.0.0

But all traffic is now routed through only one ISP, and there’s no auto load balancing between Site A and Site B across both ISPs.

[MHM Cisco World]

Disable link between ISP-2-A and ISP-2-B and check ping between two core 

Issue of asymmetric with vrf I think

MHM

[ALI12]

Thank @MHM Cisco World for your input

when the link between ISP-2-A and ISP-2-B was disabled, traffic routed through ISP-1, and when the link between ISP-1-A and ISP-1-B was disabled, traffic routed through ISP-2 as expected.

However, after adjusting the distance values, all traffic is now routed through only one ISP, and there’s no auto load balancing between Site A and Site B across both ISPs.

I don't believe it's related to VRF since all routers are in the same VRF.

[MHM Cisco World]

It issue of asymmetric as I mention before 

Use redistrubte connect under eigrp and check.

Sorry from now until 1/1/2025 

I will be unavailable 

Hope other help you

Goodluck 

MHM

[ALI12]

@MHM Cisco World  Wishing you all the best, and good luck with your plans!

[paul driver]

Hello
what was the issue before you change the distances you had ecmp ?

[ALI12]

@paul driver 
there was loopback issue

[paul driver · Accepted solution]

Hello
Thats expected as you have equal cost multipath  (ecmp) - in this case meaning you have two different paths to the same destination that have the same metric calculation so they are equal - when you iniciate say two or more separate traceroutes it is then being shared between those two equal alternate paths