Team LiB
Previous Section Next Section

Chapter 7

"Do I Know This Already" Quiz

Congestion-Avoidance Concepts and RED Questions

1.

TCP Slow Start controls the rate a TCP sender sends data by controlling:

  1. Growth of the Advertised Window

  2. Growth of the Congestion Window

  3. Calculation of the Average Queue Depth

  4. The wait-for-acknowledgement timer

Answer:

B

2.

For which of the following WRED categories will WRED discard all packets?

  1. Tail Drop

  2. Full Drop

  3. Random Drop

  4. Partial Drop

  5. No Drop

Answer:

B

Answer: While "Tail Drop" may be considered to be the same thing as "Full Drop" by many people, the important distinction is that WRED acts based on the average queue depth, whereas Tail Drop acts on the actual queue depth.

3.

For which of the following WRED categories will WRED discard a subset of the packets?

  1. Tail Drop

  2. Full Drop

  3. Random Drop

  4. Partial Drop

  5. No Drop

Answer:

C

WRED Questions

4.

On which of the following types of queues can you enable WRED on routers that are not part of the 7500-series router line?

  1. On the physical interface

  2. On a CBWFQ class that has been configured with the bandwidth command

  3. On an LLQ class that has been configured with the priority command

  4. On a single FIFO queue created by CB Shaping

The next three questions refer to the following configuration snippet:

     ip cef
     !
     ! The following classes are used in the LLQ configuration applied to S0/0
     !
     class-map match-all class1
       match protocol http url "*important*"
     class-map match-all class2
       match protocol http url "*not-so*"
     class-map match-all class3
       match protocol http
     !
     policy-map wred-q
       class class1
        bandwidth percent 25
        random-detect dscp-based
        random-detect dscp af22 25 35 50
       class class2
        bandwidth percent 20
        random-detect

        random-detect precedence 2 25 35 50
       class class3
        bandwidth percent 15
        random-detect dscp-based
        random-detect dscp af22 50 25 35
       class class-default
        random-detect dscp-based
        random-detect dscp af22 2 25 50
     !
     interface s0/0
      ip address 1.1.1.1 255.255.255.0
      random-detect dscp-based
      random-detect dscp af22 50 25 35
     !
     interface s0/1
     ip address 2.2.2.2 255.255.255.0
     service-policy output wred-q

Answer:

A, B

WRED can be enabled on a physical interface, but it creates a single FIFO queue, and can't be used with another queuing tool (like WFQ) at the same time. It can also be enabled in a CBWFQ, but it cannot be enabled in an LLQ.

5.

For which of the following will WRED discard 2% of packets of some precedence or DSCP value, when the average queue depth approaches the maximum threshold?

  1. On physical interface S0/0

  2. On serial 0/1, class class1

  3. On serial 0/1, class class2

  4. On serial 0/1, class class3

  5. On serial 0/1, class class-default

  6. None of the above

Answer:

B, C, E

Please refer to the answer to the next question for some background information.

6.

Imagine a packet marked as AF22. Out which interface or class must the packet be forwarded in order to have a 35% chance of being discarded, assuming that WRED's average queue depth calculation was approaching the maximum threshold?

  1. On physical interface S0/0

  2. On serial 0/1, class class1

  3. On serial 0/1, class class2

  4. On serial 0/1, class class3

  5. On serial 0/1, class class-default

  6. None of the above

Answer:

F

The last of the parameters on the random-detect dscp and random-detect precedence commands set the mark probability denominator (MPD). MPD is not set as a percentage, but the percentage is 1/MPD. None of the settings give a percentage of 35. Note that on the previous question, with a setting of 50 in some cases for MPD, that 2 percent of the packets would be dropped (1/50 = 2 percent).

7.

Assuming the commands in the configuration snippet were typed into configuration mode in a router, one of the random-detect commands would be rejected. Under which configuration mode can that erroneous command be found?

  1. On physical interface S0/0

  2. On serial 0/1, class class1

  3. On serial 0/1, class class2

  4. On serial 0/1, class class3

  5. On serial 0/1, class class-default

  6. None of the above

Answer:

D

The random-detect dscp AF22 50 25 35 command attempts to set the minimum threshold to 50, and the maximum to 25, which is not allowed.

ECN Questions

8.

Imagine that WRED with ECN has been configured for a CBWFQ class. Under which of the following cases could WRED randomly choose to discard a packet, but instead, mark the ECN bits inside the packet header and allowing the packet to pass?

  1. Average queue depth between the min and max thresholds, plus the incoming packet ECN field set to 00

  2. Average queue depth between the min and max thresholds, plus the incoming packet ECN field set to 01 or 10

  3. Average queue depth above max threshold, plus the incoming packet ECN field set to 01 or 10

  4. Average queue depth between the min and max thresholds, plus the incoming packet TCP ECE flag must be set to "1", and ECN field must be set to 01

Answer:

B

9.

Referring to the configuration snippet before question 5, what command would be required to enable ECN for class2 in policy-map wred-q?

  1. ecn enable

  2. no ecn disable

  3. random-detect ecn

  4. None required WRED automatically does it if the TCP sender sets the ECN bits correctly

Answer:

C

Q&A

Congestion-Avoidance Concepts and Random Early Detection (RED)

1.

Describe the function of the congestion window in TCP, and how it is changed as a result of packet loss.

Answer:

The TCP congestion window, or CWND, is one of two windowing mechanisms that limit TCP senders. CWND can be split in half as a result of packet loss, slowing the sending rate. CWND can also be slammed shut to the size of a single segment in some cases.

2.

Identify the two TCP windowing mechanisms, and describe when each is used.

Answer:

The TCP congestion window, or CWND, and the TCP receiver window, are the two windowing mechanisms. The lower of the two values is used at all times.

3.

Describe the process of TCP slow start, and when it occurs.

A3:

Answer: TCP slow start governs the growth of the TCP congestion window after the window has been lowered in reaction to a packet drop. Slow start increases the window by one segment size for each positively acknowledged packet received.

4.

Describe the meaning of the term "global synchronization," and discuss what causes it.

A4:

Answer: Global synchronization describes a condition in which many TCP connections have their congestion windows lowered due to unacknowledged or lost segments at around the same instant in time. The connections all grow CWND at about the same rate, re-creating the same congestion levels again, causing more drops, which in turn reduces again the TCP congestion windows. Global synchronization is caused by a large number of packet drops in a very short period, typically the result of tail drops.

5.

Define the meaning of the term "tail drop."

A5:

Answer: When a queue fills, and a new packet must be placed into the queue, the packet is dropped. Because the packet would be placed into the end, or tail, of the queue, it is called tail drop.

6.

Define the meaning of the term "TCP starvation."

A6:

Answer: When packets are dropped, TCP connections slow down, but UDP flows do not slow down. UDP packets can consume a disproportionate amount of queue space as a result, which could get to the point that the TCP connections simply get little or no queue space; this is called TCP starvation.

7.

Does RED compare the actual queue depth or the average queue depth to queue thresholds when deciding whether it should discard a packet? Why this one, and not the other?

A7:

Answer: RED uses average queue depth. By using the average, rather than the actual queue depth, RED behaves more consistently, rather than more erratically, which helps prevent synchronization of TCP flows.

8.

Describe how RED uses actual queue depth to calculate average queue depth. Do not list the formula, but just describe the general idea.

A8:

Answer: RED calculates the average by adjusting the previously calculated average a small amount based on the current actual queue depth. By default, the current queue depth is weighted at about .2 percent in the formula.

9.

Assume the RED minimum threshold is 20, the maximum threshold is 40, and the mark probability denominator is 10. What must be true for RED to discard all new packets?

A9:

Answer: The average queue depth must be above 40.

10.

Assume the RED minimum threshold is 20, the maximum threshold is 40, and the mark probability denominator is 10. What must be true for RED to discard 5 percent of all new packets?

A10:

Answer: The average queue depth must be at 30. Because the discard percentage grows linearly from 0 percent to 10 percent (in this case), between average queue depth of 20 through 40, average queue depth of 30 would mean that the discard percentage had grown to 5 percent.

11.

Define how RED uses the mark probability denominator. Give one example.

A11:

Answer: RED calculates the discard percentage based on the formula 1/MPD. For instance, with an MPD of 20, the discard percentage is 1/20, or 5 percent.

12.

Define the term "exponential weighting constant." If the value is lowered compared to the default setting of 9, how does RED behave differently?

A12:

Answer: The exponential weighting constant defines how quickly the average queue depth changes, by determining how much the actual queue depth affects the rolling average queue depth. If EWC is lowered, the average changes more quickly, because the formula weights the current actual queue depth more than before. Therefore, a larger constant provides more handling of bursty traffic, but too large and congestion avoidance will be ineffective.

13.

Define the term "WRED Profile."

A13:

Answer: A WRED profile is a collection of WRED parameters applied to a single IP Precedence or DSCP value. The parameters include the minimum threshold, the maximum threshold, and the Mark Probability Denominator (MPD).

14.

Explain how you can tune how fast or slow that WRED changes the calculated average queue depth over time.

A14:

Answer: WRED calculates a new average based on the old average and the current queue depth. You can tell WRED to count the current queue depth as a larger or smaller part of the calculation by tuning the exponential weighting constant. The formula is:

New average = (Old_average * (1 2n)) + (Current_Q_depth * 2n)

Where "n" is the exponential weighting constant.

Weighted RED (WRED)

15.

Spell out the words represented by the initials RED, WRED, and FRED.

A1:

Answer: Random Early Detection (RED), Weighted Random Early Detection (WRED), Flow-Based Weighted Random Early Detection (FRED).

16.

List the three WRED terms that name the separate states in which WRED discards no packets, a percentage of packets, and all packets.

A2:

Answer: No Discard, Random Discard, and Full Discard, respectively.

17.

List the queuing tools that can be concurrently supported on an interface when WRED has been enabled directly on a serial interface, assuming no retrictions on the particular model of router.

A3:

Answer: FIFO Queuing only.

18.

Identify the most important difference between RED operation and WRED operation.

A4:

Answer: WRED weights its discard decisions based on precedence or DSCP, whereas RED ignores precedence and DSCP.

19.

Describe how WRED "weights" packets.

A5:

Answer: WRED weights packets based on precedence or DSCP by assigning different minimum threshold, maximum threshold, and mark probability denominator values for each precedence or DSCP.

20.

List the queuing tools that can enable WRED for use with some or all of their queues, effectively enabling WRED concurrently with the queuing tool, assuming no retrictions on the particular model of router.

A6:

Answer: CBWFQ and LLQ.

21.

What command enables you to look at WRED drop statistics when WRED is configured inside an MQC class?

A7:

Answer:

show policy-map interface

22.

Taking as many defaults as possible, list the configuration commands needed to configure precedence-based WRED on interface S1/1.

A8:

Answer:

interface serial 1/1
         random-detect

23.

Taking as many defaults as possible, list the configuration commands needed to configure DSCP-based WRED on interface S1/1.

A9:

Answer:

interface serial 1/1
         random-detect dscp-based

24.

Taking as many defaults as possible, list the configuration commands needed to configure DSCP-based WRED inside class class1, inside policy map my-policy. (You can assume that the CBWFQ configuration has already been completed, and you just entered global configuration mode. Assume that you need just to enable WRED in class class1.)

A10:

Answer:

policy-map my-policy
         class class1
           random-detect dscp-based

25.

List the command needed to set the minimum threshold to 25, the maximum threshold to 50, and the mark probability denominator to 4, for precedence 2.

A11:

Answer:

random-detect precedence 2 25 50 4

26.

What show command lists detailed statistics about random drops on interface S1/1?

A12:

Answer:

show queueing interface s1/1

Explicit Congestion Notification

27.

For a single WRED profile, WRED can be either dropping no packets, randomly choosing packets to discard, or dropping all packets. For which of these three states does ECN impact WRED's discard actions? How does it change what WRED does to the packets?

A1:

Answer: For Random Discard only. WRED forwards the packets instead of discarding them, but only after setting the ECN bits to "11".

28.

Identify the bits in the IP header used with ECN, by name and location.

A2:

Answer: The low-order 2 bits of the DSCP byte are called the ECN field. The first bit is called the ECN Capable Transport (ECT) bit, and the second one is the Congestion Experienced (CE) bit.

29.

Imagine a router on which WRED and ECN are enabled, and WRED decides to randomly discard a packet. What must be true in order for WRED to discard the packet, instead of using ECN logic to mark and forward the packet? Explain the role of any other devices besides the router.

A3:

Answer: With ECN enabled, it would set the ECN bits to "11", unless the ECN field was set to 00. An ECN field of 00 means that the sender did not support ECN for that TCP connection.

30.

Imagine a router on which WRED and ECN are enabled, and WRED decides to randomly discard a packet. What must be true in order for WRED to use ECN logic to mark and forward the packet, instead of discarding the packet? Explain the role of any other devices besides the router.

A4:

Answer: The ECN field must be set to something besides 00. The sender of the packet would choose to set ECN to one of those two values if it did support ECN for that TCP connection.

31.

Imagine a policy map with WRED already configured for class class-web. What additional command is required to also enable ECN for the packets in that class?

A5:

Answer: The random-detect ecn command.

    Team LiB
    Previous Section Next Section