Wrap the whole inner Ethernet frame
02The inner Ethernet frame becomes data inside VXLAN and the User Datagram Protocol (UDP). The outer Internet Protocol (IP) addresses identify tunnel endpoints, while outer Ethernet addresses serve the current physical link.
After receiver checks, VTEP B removes the outer headers and passes the inner Ethernet frame to the destination VM. The original inner FCS was not carried across the tunnel.
Source for this drawing: RFC 7348 · Read RFC section 5 ↗
New to this? Start hereA quick picture + helpful background
A message needs an address.
Think of a packet as a small addressed envelope. Devices pass it onward; transport protocols decide what delivery promises an application gets.
Helpful first, if these ideas are unfamiliar
These pictures are analogies to help you learn. The linked RFC explains the actual rules.
Terms in this drawing
Related RFCs and changes
“Updates” changes part of an earlier RFC. “Obsoletes” replaces it. This list shows selected related documents.
Inner addresses describe VM communication; outer addresses carry the wrapper across the underlay.
- The outer stack is Ethernet, IP, UDP and the 8-byte VXLAN header, followed by the inner Ethernet header and its payload. The original inner Frame Check Sequence (FCS) is not included; the outer Ethernet frame gets its own FCS.
- The outer IP source/destination identify VTEPs for known unicast. The outer destination MAC may identify an intermediate router rather than the remote VTEP.
- The registered VXLAN destination UDP port is 4789 and SHOULD be used by default; the destination port SHOULD be configurable for interoperability with early implementations.
All numbered byte examples use untagged Ethernet and a 20-byte IPv4 header with no options. Layer boxes are schematic, not bit-scaled. IPv6 transport has a separate layout in Figure 2; these examples do not assert that zero UDP checksums are universally valid for IPv6.
Read the RFC section ↗