Question 10
An east-west application works without tunneling but drops large payloads after VXLAN overlay migration. Which shortcut would be most likely to undermine this objective?
Explanation
Correct Answer
This is the unsafe shortcut to avoid: assume the underlay can carry the original maximum Ethernet frame unchanged after VXLAN headers are added. It violates the principle that VXLAN adds encapsulation overhead and requires adequate path MTU or effective payload sizing. An appropriate alternative is to configure sufficient supported underlay MTU along the fabric path or safely adjust payload MTU.
Incorrect Answers
That choice—“Replace the tenant VLAN ID repeatedly without checking VXLAN endpoint reachability”—addresses this other problem: An endpoint's remote leaf cannot deliver traffic even though the local access VLAN is assigned correctly because the tunnel endpoint is unreachable. Here the scenario is: An east-west application works without tunneling but drops large payloads after VXLAN overlay migration. VXLAN adds encapsulation overhead and requires adequate path MTU or effective payload sizing. The directly relevant unsafe shortcut is “Assume the underlay can carry the original maximum Ethernet frame unchanged after VXLAN headers are added”.
The option “Map every rack's VLAN 210 to an arbitrary shared VNI without checking tenant boundaries” is useful in a different setting: A rack migration assigns the correct switch access VLAN but places the VTEP into another tenant's virtual network identifier. The case being tested is: An east-west application works without tunneling but drops large payloads after VXLAN overlay migration. VXLAN adds encapsulation overhead and requires adequate path MTU or effective payload sizing. The directly relevant unsafe shortcut is “Assume the underlay can carry the original maximum Ethernet frame unchanged after VXLAN headers are added”.
“Treat all VXLAN BUM packets as harmless and expand replication to every VTEP in the data center” is relevant to a different condition: An overlay virtual network has the correct VNI but unknown destination traffic continues to flood farther than intended. The stated question instead concerns: An east-west application works without tunneling but drops large payloads after VXLAN overlay migration. VXLAN adds encapsulation overhead and requires adequate path MTU or effective payload sizing. The directly relevant unsafe shortcut is “Assume the underlay can carry the original maximum Ethernet frame unchanged after VXLAN headers are added”.