ACI Multi-Pod Data Traffic Flow
Let’s see how ARP packet is exchanged when ARP Flooding is enabled in BD1, and EP1 and EP2 are member of it.
- EP1 sends an ARP request to know EP2’s MAC address, EP1 and EP2 are in same subnet.
- When leaf receives the packet, learns EP1 information also the leaf does not find any information about EP2 in its local forwarding tables. In BD1 ARP flooding is enabled, the leaf encapsulates the packet into a multicast packet with external destination address is the GIPo associated to the specific BD While encapsulation, the leaf also adds to the VXLAN header the S_Class information relative to the End Point Group (EPG) that EP1 belongs to, Leaf send this packet to spine selected as AD for that GIPo.
- The designated spine sends the VXLAN ARP request across the IPN, where IPN replicate this packet toward all the remote Pods where this specific BD has been deployed.
- When designated spine nodes in Pod2 receives the packet it floods this frame locally using local multi-destination tree. Also that the spine has learned EP1 information from an MP-BGP update received from the spine in Pod1.
- The leaf where EP2 is connected receives the flooded ARP request, learns EP1 information (location and class-id) and forwards the packet to all the local interfaces part of the BD.
- EP2 receives the ARP request and plan for ARP reply.
- Now EP2 generates a unicast ARP reply destined to EP1 MAC address as EP2 know the MAC address of EP1.
- When Local leaf receives this ARP reply , this leaf has EP1 location information so it encapsulate the frame in VXLAN encapsulated and destined to Leaf 1 in Pod1. The local leaf also learns EP2 information informs the local spine nodes via COOP.
- The leaf 1 node in Pod1 receives the packet, decapsulates it, learns EP2 information and store it in its Local table along with class-id information and forwards the packet to the interface where EP1 is connected. EP1 is hence able to receive the ARP reply.
ARP Exchange via ARP Gleaning Method: