As shown in Figure 5, when a packet is sent from VNI A to VNI B, the ingress VTEP routes the packet to the Layer-3 VNI. It rewrites the inner destination MAC address to the egress VTEP’s router MAC address and encodes the Layer-3 VNI in the VXLAN header. After the egress VTEP receives the encapsulated VXLAN packet, it first decapsulates the packet by removing the VXLAN header. Then it looks at the inner packet header. Because the destination MAC address in the inner packet header is its own MAC address, it performs a Layer-3 routing lookup. The Layer-3 VNI in the VXLAN header provides the VRF context in which this routing lookup is performed.
I'm confused with the explanation, I guess when they refer to Layer-3 VNI they are talking about the VNI we declared below the VRF context (in the same document they used 39000 and 39010). If is that the case the corresponding interface VLAN of those VNI has no IP address, so how the traffic is routed? My other question is, for instance if we have only two VTEPs in the topology, in the data-packet the destination IP address of the VXLAN header (outer header) is the same if we routed or bridged the traffic?
Hi,I'm learning vxlan right now,so I'll try to answer your questions.
From my understanding,when the VTEP doing IRB,it does not need an ip address for l3 vni,it only uses the l3 vni MAC to encapsulate the packet(inner packet),the inner packet is like this:
source mac:ingress vtep router mac dest mac:egress vtep router mac
source ip:source user ip dest ip:dest user ip
As for your other question,I think the answer is Yes.Vxlan uses the same egress vtep ip transfering either routed or bridged user traffic.Just consider Vxlan a tunnel, and the address of the tunnle endpoint always keeps the same no matter what's travelling in it.
Hi all,not sure this is the right place to raise my question..if not..beg you pardon.problem: I have to interconnect 2 datacenter with 2 links 10gb each so...10+10physical connection is DWDM (dark fiber) and each datacenter has is own nexus93108 (as ...
Coming soon - Be the first to know about the New Product/Release Series. Join Customer Connection to register for the briefings throughout the month of October!
Membership in the Cisco Customer Connection program is required to attend. Registra...
Description:Fault delegate: A Fabric Node Group (fabricNodeGrp) configuration was not deployed on the fabric node 152 because: Node Not Registered for Node Group Policies Type:Config Cause:configuration-failed Change Set:deplSt:not-register...
<original blog published on cisco @ https://blogs.cisco.com/datacenter/cisco-mds-32g-fibre-channel-fabric-switches-small-doesnt-mean-less >
When we think about fabric switches, we tend to consider low-end hardware. These typically do ...
Cisco Champion Radio · S7|E34 Cisco HyperFlex with iSCSI Helps Consolidate Workloads
IT organizations that get the most out of their technology investments tend to grow faster than their peers. As a result, they are always exploring enhancements to exist...