An ISP provides VoIP and internet services to its customers. For security reasons, these services must be transported in different MPLS Layer 3 VPNs over the ISP core network. The customer CEs do…

Question

Options

• AUse a secondary interface IP address to differentiate between VoIP and Internet traffic
• BExtend the Layer 3 VPN toward the CE
• CEnable NBAR on the PE to direct the traffic into the correct VRF
• DUse a subinterface on the PE for each service, VoIP and Internet, with different subnets
• EUse policy-based routing to direct traffic into the correct VRF

• A
  4% (1)
• B
  4% (1)
• C
  12% (3)
• D
  4% (1)
• E
  77% (20)

Explanation

Policy-based routing (PBR) is correct because it lets the PE inspect incoming traffic characteristics - such as destination IP ranges or Layer 4 port numbers (e.g., UDP 5060/5004 for SIP/RTP) - and forward matching flows into the appropriate VRF, all on a single physical interface without requiring any CE-side configuration changes.

Why the distractors fail:

• A (Secondary IP): A secondary address still binds to the same interface/VRF; it gives the interface two IPs but provides no mechanism to steer traffic into separate VRFs based on traffic type.
• B (Extend L3 VPN to CE): This would require the CE to run VRF-aware routing or support MPLS, capabilities the CE explicitly lacks.
• C (NBAR): NBAR classifies applications for QoS marking but has no native ability to assign traffic to a VRF; you would still need PBR to do the actual steering.
• D (Subinterfaces): Subinterfaces depend on 802.1Q VLAN tags from the CE to demultiplex traffic - exactly what the CE cannot provide.

Memory tip: When the CE is "dumb" (one interface, no VLANs), the intelligence must move to the PE - remember PBR = "PE takes the wheel", matching traffic with an ACL and setting the VRF (via ip vrf forwarding in a route-map) before the routing table is consulted.

No community discussion yet for this question.