Multi-Pod: Can not ping Spine TEP address

[daz_efx_2000]

Help Me

I have a LAB with 4 IPNs. With 2 IPNs per pod. And 2 Spines per pod however the spines are only connected to 1 IPN (see image)

Should there only be 1 IPN TEP address in each pod that's shared by spine 1 and 2 ?

Or should there be a IPN TEP address for each spine.

I am asking this because policy is not being pushed properly to pod 2 (APIC is in POD 1, only one APIC available for now)

Thanks

[Rick1776]

See the below statement form the Multi-POD white paper....

IPN Control Plane

As previously mentioned, the Inter-Pod Network represents an extension of the ACI infrastructure network, ensuring VXLAN tunnels can be established across Pods for allowing endpoints communication.

Inside each ACI Pod, IS-IS is the infrastructure routing protocol used by the leaf and spine nodes to peer with each other and exchange IP information for locally defined loopback interfaces (usually referred to as VTEP addresses). During the auto-provisioning process for the nodes belonging to a Pod, the APIC assigns one (or more) IP addresses to the loopback interfaces of the leaf and spine nodes part of the Pod. All those IP addresses are part of an IP pool that is specified during the boot-up process of the first APIC node and takes the name of ‘TEP pool’.

In a Multi-Pod deployment, each Pod is assigned a separate and not overlapping TEP pool.

https://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/application-centric-infrastructure/white-paper-c11-737855.html#IPNControlPlane5

[Rick1776]

Checkout this Cisco Live Presentation as well..

http://d2zmdbbm9feqrf.cloudfront.net/2017/usa/pdf/BRKACI-2003.pdf

https://www.ciscolive.com/online/connect/sessionDetail.ww?SESSION_ID=95448&tclass=popup

[nvermand]

Well it depends what TEP address you are referring to:
The control-plane TEP address or CP-ETEP, is the address used for BGP peering. It has to be unique per spine.
In a particular pod, the data-plane TEP or ETEP, represents the anycast next-hop for traffic originated from remote pod and destined to local pod.
If you're not able to ping a spine, on the leaf where you've initiated iping, you can check the routing table for overlay-1 VRF. You should see the subnet corresponding to your TEP network there. If not, there's a routing distribution issue and that probably means that either your inital Multi-pod configuration doesn't have the right information, or that your IPN routing configuration has some errors.

Nicolas

[daz_efx_2000]

Thanks guys for the info,

@nvermand I the TEP I am referring to is the one used for BGP. I am presuming CP-ETP stands for Control Plane Evpn Tunnel end point.

I can now ping the CP-ETEP at both ends. FYI I fixed this by simply removing and re-adding the setting at the following location.

Fabric-> Inventory -> Topology -> click "CONFIG" in the  interpod interactive map -> POD Connection Profile.

In this section you have to specify a POD ID and a Dataplane TEP (CP-ETEP) for each POD. I removed and re-added and I can now ping the CP-ETEP from the IPN's. However in this section I have only 2 entries. From what I have understood in your reply I should have 4 entries in here as i have 4 spines??

Current settings:

POD ID               Dataplane TEP

======              ===========

1                         10.10.10.1/32

2                         10.10.10.2/32

What my settings presumably should look like.??

POD ID               Dataplane TEP

======              ===========

1                         10.10.10.1/32

2                         10.10.10.2/32

1                         10.10.10.3/32

2                         10.10.10.4/32

[shehreyarkhan]

Hi,

~The post seems old, but how did you manage to solve the issue?

Regards