Config Router

You are here: Home / Cisco / CCNP SP MPLS FAQ: MPLS Traffic Engineering

CCNP SP MPLS FAQ: MPLS Traffic Engineering

by

CCNP SP MPLS FAQ: MPLS Traffic Engineering

Q1. Name the advantages of MPLS traffic engineering

Answer: The advantages of MPLS traffic engineering include the following:

  • Steering traffic so that all links in the network are used optimally
  • Routing traffic around hotspots in the network
  • Fast rerouting traffic around link and node failures

Q2. What are the components of MPLS traffic engineering?

Answer: Following are the components of MPLS traffic engineering:

  • OSPF or IS-IS with TE extensions enabled
  • RSVP
  • Link attributes
  • CSPF on the head end router
  • TE Tunnel (Trunk) attributes
  • Link manager

Q3. What are the attributes of the links enabled for traffic engineering?

Answer: The attributes of the links that are enabled for traffic engineering include these:

  • Maximum reservable bandwidth
  • Maximum reservable sub-pool bandwidth
  • SRLG
  • Attribute flags
  • Traffic engineering metric

Q4. Name the six ways how you can forward IP traffic onto a TE tunnel.

Answer: You can forward IP traffic onto a TE tunnel by using the following:

  • A static route
  • Policy-based routing
  • Forwarding adjacency
  • Autoroute announce
  • Direct mapping of AToM traffic onto TE tunnels
  • CBTS

Q5. Name four kind of path options you can specify.

Answer: You can specify the following path options:

  • Explicit
  • Explicit with exclude address
  • Explicit with loose next address
  • Dynamic

Q6. When you have an MPLS VPN network and TE tunnels that do not always have the PE routers as head/tail end routers, what do you need to have?

Answer: When you have an MPLS VPN network and TE tunnels that do not always have the PE routers as head/tail end routers, you need the following:

  • LDP enabled on the links
  • LDP targeted session on the TE tunnel if the tail end router is a P router

Q7. What is the LSR that is the head end router of a backup tunnel called?

Answer: The point of local repair (PLR) is the LSR that is the head end router of a backup tunnel.

Q8. Why do you need a link-state routing protocol for MPLS TE?

Answer: You need a link-state routing protocol for MPLS TE because the head end LSR needs to know the attributes and available bandwidth of all links.

Q9. What is the feature called that has the IGP advertise TE tunnels as links?

Answer: Forwarding adjacency is the feature that has the IGP advertise TE tunnels as links.

Q10. How does the head end LSR of a TE tunnel know that the tunnel is fast rerouted over a backup tunnel?

Answer: The head end LSR of a TE tunnel knows that the tunnel is fast rerouted over a backup tunnel because it receives a PathErr message from the PLR, indicating that Local Repair is active.

Q11. Is it possible to restrict which prefixes have labels assigned to them so that label switching occurs to only certain destination prefixes?

Answer: No. Labels are assigned to all internal prefixes, although it is possible to filter the distribution of these prefixes, which provides the desired result.

Q12. What is the maximum frame size permissible across an Ethernet segment, and how can introducing MPLS affect forwarding across this type of media?

Answer: 1518 bytes. By introducing MPLS, you can exceed the maximum permissible frame size through the addition of one or more MPLS labels.

Q13. Which mechanism can you deploy to help determine the maximum allowable Path MTU for a packet? Path MTU discovery as described in RFC 1191.

Answer: Path MTU discovery as described in RFC 1191.

Q14. What is the maximum size datagram, which does not support Path MTU discovery, that hosts can send?

Answer: 576 octets.

Q15. What is the purpose of the DF bit?

Answer: To determine whether a packet can be fragmented by a receiving LSR. If the bit is set, the packet cannot be fragmented and must be dropped if it exceeds the maximum MTU of the outgoing link. In a frame-mode MPLS implementation, how are loops detected in the forwarding plane?

Q16.In a frame-mode MPLS implementation, how are loops detected in the forwarding plane?

Answer: By examination of the TTL field within the MPLS header. If the TTL field reaches 0, the packet is dropped.

Q17. Is it possible to detect loops in the forwarding plane by using the TTL field in a cell-mode implementation of MPLS?

Answer: No. There is no concept of TTL within an ATM header.

Q18. Using the hop-count TLV, an ATM-LSR can detect a loop in the control plane. What is the maximum number of permissible hops within a Cisco implementation, and can you change this maximum?

Answer: 254 hops. You can change the maximum by using the mpls ldp maxhops command.

Q19. When using traceroute across an MPLS network, how can ICMP messages be forwarded back to the source of the traceroute?

Answer: By reusing the original label stack, the original destination can forward the packet back across the MPLS network to the traceroute’s source.

Q20. What action does an aggregation LSR take if no label exists on an incoming packet?

Answer: It routes the packet based on its destination IP address.