1.What is the purpose of the qos pre-classify command? 
A. to enable the IOS to copy the ToS field from the original IP header to the outer tunnel IP header 
B. to enable the IOS to copy the ToS field from the outer tunnel IP header back into the original IP header 
C. to enable the IOS to classify the packet based on the original IP header instead of the tunnel IP header 
D. to enable the IOS to classify the packet based on the outer tunnel IP header instead of the original IP header 
E. to enable class-based marking on tunnel interface 
F. to enable class-based marking on IPSec crypto maps 
Answer: C 
2.What are two benefits of WFQ? (Choose two.) 
A. WFQ is very easy to configure, and no manual traffic classification is necessary. 
B. WFQ can provide fixed-bandwidth and fixed-delay guarantees. 
C. WFQ can provide fixed-bandwidth guarantees. 
D. WFQ can provide fixed-delay guarantees. 
E. WFQ prevents the large-volume flows with large packet size from starving out the low-volume flows with
small packet size. 
F. Based on DSCP, WFQ allows weighted, random dropping of packets when the WFQ system is full. 
Answer: AE 
3.Based on the configuration, which two of the following statements are true? (Choose two.)  
A. This configuration will use a single token bucket. 
B. This configuration will drop all exceeding traffic. 
C. This is a dual-rate, class-based policing example. 
D. This is a percentage-based policing example. 
E. This is a multi-action, class-based policing example. 
Answer: CD
TestInside    642-642
4.Based on the configuration, which two statements are true? (Choose two.)  
A. The interactive traffic class will have a minimum bandwidth guarantee of 256 kbps. 
B. The interactive traffic class will have a maximum bandwidth guarantee of 256 kbps. 
C. If the interactive traffic class exceeds an average rate of 256 kbps, the traffic rate will be throttled down to 128
kbps. 
D. This configuration allows class-based traffic shaping to lower the traffic rate in response to the BECN bit. 
E. The interactive traffic class will have a min-rate (min-cir) of 128 kbps. 
Answer: DE 
5.Which technology is required when configuring FRF.12 on a Cisco device? 
A. FRF11.c 
B. FRF.8 
C. VoFR 
D. MLP with interleaving 
E. FRTS 
F. WRED 
Answer: E 
6.What are three methods of reducing delay for critical applications in a converged network? (Choose three.) 
A. Apply payload compression. 
B. Increase link capacities. 
C. Apply header compression. 
D. Increase LFI fragment size. 
E. Reduce inter-packet gaps. 
F. Increase all queue depths. 
Answer: ABC 
7.At the network layer, IP packets are typically classified based on which three items? (Choose three.) 
TestInside    642-642
A. packet length 
B. VLAN Identifier 
C. flow control bits 
D. source and destination IP addresses 
E. content of the ToS byte 
Answer: ADE 
8.What are three reasons link efficiency mechanisms are deployed on WAN links? (Choose three.) 
A. decrease delay 
B. decrease jitter 
C. increase link speed 
D. increase throughput 
E. decrease propagation delay 
Answer: ABD 
9.Which mechanism is used for the prioritizing, protection, and isolation of traffic based on marking? 
A. classification and marking 
B. congestion management 
C. congestion avoidance 
D. metering 
E. policing 
F. shaping 
Answer: B 
10.For which service is assured forwarding PHB used? 
A. best effort 
B. expedited forwarding 
C. guaranteed bandwidth 
D. class selector 
Answer: C 
TestInside    642-642
11.Which three factors will affect processing delay? (Choose three.) 
A. CPU speed 
B. router architecture 
C. queuing mechanism 
D. type of media 
E. IP switching method 
F. distance of media 
Answer: ABE 
12.which four of the following bit values are used for bits 5-7 of the DSCP field to select AF PHB? (Choose four.) 
A. 000 
B. 001 
C. 010 
D. 011 
E. 100 
F. 101 
Answer: BCDE 
13.Based on the following 2950 switch configurations, which statement is correct?  
A. Queue 1 is setup as the expedite queue. 
B. Queue 2 is setup as the expedite queue. 
C. Queue 3 is setup as the expedite queue. 
D. Queue 4 is setup as the expedite queue. 
E. No queue is setup as the expedite queue. 
Answer: E
TestInside    642-642
14.When LLQ is being configured, which IOS command is used to limit the traffic rate on the priority queue even
when the other class queues are not congested? 
A. priority 
B. bandwidth 
C. queue-limit 
D. police 
E. hold-queue 
Answer: D 
15.Where is the error in the following policy-map configuration?  
A. The bandwidth command is not a valid command for the class-default traffic class in this case. 
B. The voice traffic class is missing the random-detect command. 
C. The mission-critical traffic class bandwidth guarantee should be lower than the voice traffic class priority
bandwidth guarantee. 
D. The mission-critical traffic class is missing the queue-limit command. 
E. Fair-queue should be enabled for the mission-critical traffic class. 
Answer: A 
16.QoS policy propagation through BGP (QPPB) supports which two QoS markers? (Choose two.) 
A. DSCP 
TestInside    642-642
B. IP precedence 
C. QoS group 
D. COS 
E. MPLS EXP 
Answer: BC 
17.Which element is mandatory for QoS policy propagation through BGP operations? 
A. MPLS 
B. NBAR 
C. CEF 
D. QoS pre-classify 
E. policy-based routing 
F. MQC 
Answer: C 
18.Based on the following policy-map configuration, the bulk traffic class packets are not being ECN marked, yet
there are many random drops in the bulk traffic class. What is the most likely cause of the problem?  
A. The CB-WRED min threshold is set too low. 
B. The CB-WRED max threshold is set too high. 
C. The CB-WRED mark probability denominator is set too low. 
D. The endpoints generating the bulk traffic packets are not ECN capable. 
E. Some of the routers within the traffic path are not ECN capable. 
TestInside    642-642
Answer: D 
19.Which two of the following statements about CBWFQ are correct? (Choose two.) 
A. The CBWFQ scheduler provides a guaranteed amount of minimum bandwidth to each class. 
B. CBWFQ services each class queue using a strict priority scheduler. 
C. The class-default queue only supports WFQ. 
D. Each CBWFQ traffic class is policed using a congestion aware policer. 
E. Inside a class queue, processing is always FIFO, except for the class-default queue. 
Answer: AE 
20.Which of the following configurations requires the use of hierarchical policy maps? 
A. the use of class-based WRED within a CBWFQ class queue 
B. the use of a strict priority-class queue within CBWFQ 
C. the use of CBWFQ inside class-based shaping 
D. the use of nested class-maps with class-based marking 
E. the use of both the bandwidth and shape statements within a CBWFQ class queue 
Answer: C 
21.The following commands have been configured under the fa0/1 interface on the 2950 switch:  
Voice traffic from the IP phone that is directly connected to the fa0/1 interface is experiencing excessive delays.
What could be the cause of this problem? 
A. The wrr-queue bandwidth weightings are not correct. 
B. The default wrr-queue cos-map is being used. 
C. The default cos-to-dscp map is being used. 
D. The default dscp-to-cos map is being used. 
E. The trust boundary configuration is not correct. 
Answer: B
TestInside    642-642
22.Which QoS mechanism adds IP Precedence information for prefixes into the FIB table? 
A. QoS pre-classify 
B. AutoQoS 
C. QPPB 
D. Class-Based Marking 
E. LLQ 
F. Class-Based WRED 
Answer: C 
23.Refer to the partial router configuration. Which two of the following statements are true? (Choose two.)  
A. The configuration is invalid since it refers to a class map within a different class. 
B. The class-map class1 command will set the qos-group value to 4 for all IP packets. 
C. Regardless of destination IP address, all traffic sent to Mac address 1.2.3 will be subject to policing 
D. Only those packets which satisfy all of the matches in class1 and class2 will be subject to policing. 
E. All traffic from a server with the IP address of 147.23.54.21 will be subject to policing. 
F. Any IP packet will be subject to policing. 
Answer: CE 
24.In a typical converged campus network, which two of the following are considered QoS best practices?
(Choose two.) 
A. Traffic classification and marking is performed as close to the traffic source as possible. 
B. Traffic classification and marking is performed at the high speed core layer. 
TestInside    642-642
C. Ensure voice traffic is serviced by a weighted fair queue. 
D. Only a reasonable number of applications should be classified into the mission-critical traffic class. 
E. NBAR is used at the high speed core layer to discover and classify network applications. 
F. Catalyst switches should use weighted round robin (WRR) queuing giving the voice traffic the highest priority. 
Answer: AD 
25.Which three characteristics are drawbacks to the use of a best-effort with over-provisioning backbone design?
(Choose three.) 
A. The design costs more to implement than a DiffServ backbone. 
B. Denial of Service attacks on one service can affect all network traffic. 
C. The design uses a different over-provisioning ratio for the different traffic classes. 
D. It requires complex QoS mechanisms at the network edge. 
E. It requires complex QoS mechanisms at the network core. 
F. Unplanned network failures can cause unexpected congestion in the network. 
Answer: ABF 
26.Which two different traffic types have the most similar sensitivity to latency, jitter, and packet loss? (Choose
two.) 
A. SQL transactions 
B. voice 
C. voice signaling 
D. streaming video 
E. video conferencing 
F. peer-to-peer file sharing 
Answer: BE 
27.To determine the bandwidth requirement for each VoIP call, not including Layer 2 overhead, how much
bandwidth per call should be added to account for the voice signaling traffic? 
A. 20 bps 
B. 40 bps 
C. 150 bps 
TestInside    642-642
D. 240 bps 
E. 480 bps 
F. 640 bps 
Answer: C 
28.What does the following command accomplish?  
A. enables the router to lower the shaping rate when BECN bits are received 
B. enables the router to lower the shaping rate when FECN bits are received 
C. enables the router to respond to FECN bits by creating test frames in the opposite direction with the BECN bit
set 
D. enables the router to respond to BECN bits by creating test frames in the opposite direction with the FECN bit
set 
E. enables the router to increase the shaping rate when BECN bits are received 
F. enables the router to increase the shaping rate when FECN bits are received 
Answer: C 
29.What will be the peak shape rate based on the configuration that follows?  
A. 16000 bps 
B. 32000 bps 
C. 48000 bps 
D. 64000 bps 
E. 80000 bps 
F. 96000 bps 
Answer: D 
TestInside    642-642
30.Which QoS mechanism calculates the mean queue depth to determine its operation? 
A. WRED 
B. LLQ/CBWFQ 
C. WFQ 
D. class-based shaping 
E. class-based policing 
Answer: A 
31.When configuring WFQ, what is the default number of dynamic queues based on? 
A. hold-queue limit 
B. congestive discard threshold (CDT) 
C. interface bandwidth 
D. hash of the packet headers 
E. inter-packet arrival rate 
F. drop probability denominator 
Answer: C 
32.The qos pre-classify command can be configured under which two configuration modes? (Choose two.) 
A. router(config)# 
B. router(config-if)# 
C. router(config-pmap-c)# 
D. router(config-crypto-map)# 
E. router(config-cmap)# 
F. router(config-router)# 
Answer: BD 
33.Traffic classification using NBAR is configured using which IOS command? 
A. router(config-cmap)# match protocol {protocol-name} 
B. router(config-if)# ip nbar protocol-discovery 
C. router(config)# ip nbar port-map {protocol} [tcp|udp] {port-number} {port-number} .... 
D. router(config)# ip nbar pdlm {pdlm-file} 
TestInside    642-642
Answer: A 
34.The show policy-map interface command output is showing too many random drops for the mission-critical
traffic class. What can be changed to reduce the random drops? 
A. Increase the WRED max-threshold value for the mission-critical traffic class. 
B. Increase the WRED min-threshold value for the mission-critical traffic class. 
C. Decrease the WRED drop probability denominator for the mission-critical traffic class. 
D. Decrease the queue-limit for the mission-critical traffic class. 
E. Enable fair-queue within the mission-critical traffic class. 
Answer: B 
35.Based on the following Cisco Catalyst 2950 Series switch configurations, which additional command will
enable these actions?  
A. wrr-queue bandwidth 1 10 15 25 
B. wrr-queue bandwidth 1 20 30 50 
C. wrr-queue bandwidth 15 25 10 1 
D. wrr-queue bandwidth 10 15 25 0 
E. wrr-queue bandwidth 15 25 10 0 
Answer: D 
TestInside    642-642
36.Which command is needed to correct the following configurations in order to enable ppp multilink LFI on the
s0/0 interface?  
A. the no ip address command on the s0/0 interface 
B. the multilink-group 2001100114 command on the Multilink2001100114 interface 
C. the ppp multilink load-threshold command on the s0/0 interface 
D. the ppp multilink load-threshold command on the Multilink2001100114 interface 
E. the ppp multilink interleave command on the s0/0 interface 
Answer: A 
37.Where is the fragment size configured when using FRF.12 link fragmentation and interleaving? 
A. within the physical serial interface 
B. within the Frame Relay map-class 
C. within the MQC policy-map 
D. within the MQC class-map 
E. within the MQC service-policy 
F. within the logical multilink interface 
TestInside    642-642
Answer: B 
38.This policy-map is applied to a 128-kbps serial interface in the output direction. Which set of additional
commands within this policy-map will improve the link efficiency on the serial interface?  
A. class voice 
no priority 25 
bandwidth percent remaining 25 
B. class voice 
compression header ip tcp 
C. class telnet 
compression header ip tcp 
D. class telnet 
fair-queue 
E. class ftp 
air-queue 
F. class ftp 
compression header ip 
Answer: C 
39.Which command is typically applied to the voice traffic class within a policy-map? 
A. bandwidth {kbps} 
B. compress header ip rtp 
C. fair-queue 
D. shape average {bps} 
TestInside    642-642
E. shape peak {bps} 
F. random-detect dscp-based 
Answer: B 
40.What are two errors in this policy-map configuration? (Choose two.)  
A. The bandwidth command is missing for the class-default traffic class. 
B. WFQ cannot be configured for the class-default traffic class. 
C. WFQ cannot be configured for the bulk and interactive traffic classes. 
D. The bandwidth units for the bulk and interactive traffic classes are inconsistent. 
E. The priority command cannot be used with the bandwidth command within the same policy-map. 
F. The voice traffic class should have the no fair-queue command. 
Answer: CD 
41.Which two commands are typically applied to the voice traffic class within a policy-map? (Choose two.) 
A. shape peak {bps} 
B. priority {kbps} 
C. bandwidth {kbps} 
D. compress header ip rtp 
E. random-detect ecn 
TestInside    642-642
F. random-detect dscp-based 
Answer: BD 
42.Mission-critical traffic is not getting a minimum bandwidth of 192 kbps in this policy-map configuration. What
should be done to correct the problem?  
A. Change the bandwidth statement to the bandwidth 192000 command. 
B. Replace the bandwidth statement with the priority 192 command. 
C. Set the shape rate to a CIR that is higher than 192 kbps. 
D. Use the shape peak command instead of the shape average command. 
E. Increase the maximum queue size for the mission-critical traffic class using the queue-limit 128 command. 
F. Decrease the maximum queue size for the mission-critical traffic class using the queue-limit 40 command. 
Answer: C 
43.Refer to the exhibit. According to the show output, which statement is true?  
TestInside    642-642
A. NBAR protocol discovery has been enabled on the router through the use of the match protocol commands
within the class-map. 
B. The unknown protocol traffic statistics refer to all the traffic matched by the class-default traffic class. 
C. HTTP is the most active protocol on the Fa0/0 interface based on byte count. 
D. The 5-minute average bits-per-second rate for all traffic entering the Fa0/0 interface is 398 kbps. 
E. There is a total of 39,990 NetBIOS packets exiting the Fa0/0 interface. 
Answer: C 
44.Based on the Cisco Catalyst 2950 Series switch configuration shown, with no weights being assigned to the
egress queues, which type of queuing is being implemented?  
A. weighted round robin 
B. priority 
C. custom 
D. weighted fair 
E. weighted round robin with an expedite queue 
F. modified deficit round robin 
Answer: B 
45.Which algorithm (discard method) does WRED use to determine which packets will be dropped when the
average queue length becomes larger than the user-specified maximum threshold? 
A. tail discard 
B. random discard 
C. weighted early discard 
D. packet size based discard 
Answer: A 
TestInside    642-642
46.Refer to the exhibit. Why would applying the limit-interactive policy-map to the fa0/0 interface as shown
below cause an error?   
A. The interactive class-map has not been configured. 
B. The interactive traffic class is missing the bandwidth {kbps} command. 
C. Class-based policing can only be applied in the output direction. 
D. TCP header compression can only be applied in the output direction. 
E. There is already an output service-policy defined on the fa0/0 interface. 
Answer: D 
47.What is the purpose of using multiactions traffic policing? 
A. so that exceed traffic can be shaped and violate traffic can be policed 
B. so that conform, exceed, and violate traffic can be marked with different CLPs 
C. so that conform traffic from different flows can be marked with different DSCPs 
D. so that class-based policing can mark at Layer 2 and Layer 3 at the same time 
E. so that traffic can be policed using two separate rates 
Answer: D 
48.Refer to the exhibit. Which three statements are true about the configuration? (Choose three.)  
A. Traffic that is subject to shaping can burst up to 32,000 bps. 
B. IP traffic (DSCPs 5, 6, and 7) that is sent on fastethernet 0/0 will be traffic policed. 
C. RTP traffic (ports 2024 and 1000) that is sent on fastethernet 0/0 will be traffic shaped. 
D. Traffic that is subject to policing will have the DCSP set to 0 if the rate exceeds 1000 bps. 
E. IP traffic (DSCPs 1, 2, 3, and 4) that is sent on fastethernet 0/0 are considered to have a violate status and are
dropped. 
F. IP traffic (DSCP 0) that is sent on fastethernet 0/0 will be subject to fair queuing. 
Answer: ABF 
49.A Frame Relay interface has been configured for adaptive shaping with a minimum rate of 15 kbps. The
current maximum transmit rate is 56 kbps. 
If three FECNs are received over the next 4 seconds, what will be the maximum transmit rate after the last FECN
has been received? 
A. 7 kbps 
B. 10 kbps 
C. 15 kbps 
D. 28 kbps 
E. 37 kbps 
F. 56 kbps 

Answer: F 
50.In a managed CE scenario, the customer's network is supporting VoIP and bulk file transfers. According to the
best practices, which QoS mechanisms should be applied on the WAN edge CE-PE 56-kbps Frame Relay link on
the CE outbound direction? 
A. WRR, FRTS, FRF.12, and CB-RTP header compression 
B. WRR, CB-WRED, CB-Marking, FRF.12, and CB-RTP header compression 
C. CBWFQ, CB-WRED, CB-Marking, CB-Policing, and FRTS 
D. CBWFQ, FRTS, FRF.12, and CB-RTP header compression 
E. LLQ, CB-WRED, CB-Marking, FRTS, FRF.12, and CB-RTP header compression 
F. LLQ, CB-WRED, CB-Policing, and CB-TCP and CB-RTP header compressions 
Answer: E