Midterm Exam
The class has chosen to write the midterm exam.
Each student is required to submit four unique, suitable exam questions. They should be “thinking” questions that are not too long for an exam. One can be homework (state which one that is). One should be easy, two of medium difficulty, and one difficult. Submit questions and answers.
Questions should be submitted to exam@csci6433.org.
The questions need to be unique. If a question is submitted that has already been approved, the submitter will be told and given a chance to resubmit.
All questions are to be submitted and approved by October 18, to provide time to study before the midterm.
Although the answers will not be posted, you are encouraged to study together in any way you wish, including posting and assembling a list of questions and answers.
| Topic | Question |
|---|---|
| ARP | ARP is often cited as a security weakness. Explain why. |
| ARP | Explain the main difference between ARP and RARP. |
| ARP | Can ARP be used on a network that does not provide broadcast? Why or why not? |
| ARP | Not using checksums can be dangerous. Explain how a single corrupted ARP packet broadcast by machine P can make it impossible to reach another machine, Q. |
| ARP | ARP spoofing attack is a security issue in switching network environment, can you find a method to mitigate this? |
| BGP | Suppose BGP router A sends a BGP path vector to BGP peer router B. BGP peer B is connected to BGP peer C. Must B advertise that path to C? |
| DHCP | You want to implement a mechanism that automates the IP configuration, including IP address, subnet mask, default gateway, and DNS information. Which protocol will you use to accomplish this and is that done? |
| Ethernet | How does what an Ethernet switch does differ from what an Ethernet hub does? |
| Ethernet | What is the difference in terms of network traffic between two Ethernet segments connected by a bridge and two segments connected by an adaptive bridge? Why? |
| Ethernet | Is it necessary to use an IP checksum when sending packets over an Ethernet? Why or Why not? |
| Ethernet | If a switch receives a frame and the source MAC address is not in the MAC address table but the destination address is, what will the switch do with the frame? |
| Ethernet | Using network hubs may put your online credentials at risk of getting stolen, do you know why? Can you explain? |
| General | What are checksum field used for? |
| General | Could we use a single checksum for complete IP message without using UDP checksum? |
| General | What is the Good Citizen Principle? |
| General | From the standpoint of protocol design, what is the most critical resource on the Internet? |
| General | Suppose each person on the planet had a smart phone, laptop computer, and ten other devices that each had an IPv6 address. What percentage of the IPv6 address space would be required? |
| General | What are four general principles of designing Internet Protocols? Briefly explain each. |
| General | Which is better for data networks: circuit switching or packet switching? Why? |
| General | Why is it that voice and video traffic is often sent over TCP rather than UDP in today’s Internet? |
| General | Which layers in the Internet protocol stack does a host process? Which layers does a router process? |
| General | What is the difference between direct and indirect delivery? |
| General | Layered protocols hide all underlying details from applications. Could application software be optimized if an application knew about the underlying networks being used? |
| General | Which is better, UDP or TCP? Why? |
| General | What is one advantage and one disadvantage of layered network protocol? |
| General | Internet protocols sometimes violate the layering principle. Why? Give two examples. |
| General | Why is the Internet so successful? Give two reasons. |
| General | Which condition is better, LAN MTU higher than WAN MTU, or WAN MTU higher than LAN MTU? Why? |
| General | If you had network adapters with changeable MAC addresses, and you set the MAC address to be the same as the IP address for each adapter, would the Internet still work? What would change? |
| ICMP | How does traceroute use ICMP message (suppose traceroute is implemented by ICMP)? How many messages involve in this process totally (without considering duplicated sent messages)? |
| ICMP | How does ping use ICMP messages, and which ones does it use? |
| ICMP | How will you ping the loopback address of your local host with IPv4 and IPv6? |
| ICMP | Can an IP datagram carrying an ICMP message, if it causes an error, cause another IP datagram to be sent with an ICMP message about the first ICMP message? Why? |
| ICMP | Who sends ICMP messages and what is the destination? |
| IP | How does the IP addressing scheme differ from the U.S. telephone numbering scheme? |
| IP | What are the advantages of doing reassembly of fragments at the ultimate destination, rather than after the datagram traverses a single network? |
| IP | Each IP address has a subnet mask that can help determine which subnet is it in. We can do and operation (&) on each IP address with its subnet mask, and if the result is the same, then we know that this two IP address is in the same subnet. |
| IP | What is the role of IP in the TCP/IP protocol suite? |
| IP | How many IP address can one host have? |
| IP | Why should datagrams be fragmented? |
| IP | How can a host find the MAC address of another computer on the same network? |
| IP | How many computers can be setup with the network address 192.168.1.0/24 network. |
| IP | Suppose the internet design was changed to allow routers along a path to reassemble datagrams. How would the change affect security? |
| IP | You want to implement a mechanism that automates the IP configuration, including IP address, subnet mask, default gateway, and DNS information. Which protocol will you use to accomplish this and is that done? |
| IP | Fragmentation and reassembly are different for IPv4 and IPv6. Which is better? Why? |
| IP | How is the TTL field used to prevent indefinite looping of IP datagrams? |
| IPv6 | What are some of the modifications of IPv6, from the IPv4, that allow a router to process a packet quicker? |
| IPv6 | Explain three important reasons for the new approach to fragmentation taken by IPv6. |
| MPLS | What is the significance of MPLS? |
| MPLS | How do LSRs change MPLS headers? |
| MPLS | In what ways does CISCO's Multi- Layer switching, MLS differ from MPLS? |
| Multicast | What happens when I ping a multicast address? Why? |
| Multicast | What are three applications that can benefit from IP multicast. |
| NAT | What kinds of applications cannot run through a NAT box? |
| NAT | Can a local address in a network behind a NAT box be the same as a globally unique address? Explain. |
| NAT | NAT is a very important feature of IPv4. Does IPv6 need NAT? Why or why not? |
| NAT | Give two important advantages of NAT and two important disadvantages of NAT |
| Routing | Should the routing table contain IP address or MAC address of next machine? Why? |
| Routing | Routers send HELLO datagram->Building neighbor relation->Generating link state->Getting the best path by SPF algorithm->Getting a routing table |
| Routing | Why do link state protocols have better convergence rate than distance vector protocols? |
| Routing | When does a router modify the hop limit (or time-to-live) field in a datagram header? |
| Routing | How does a router know whether an incoming datagram carries a routing update message? |
| TCP | A single lost TCP acknowledgement does not necessarily force a retransmission. Explain why. |
| TCP | What is the minimum value for the timeout of a reliable transmission protocol? Why is there a minimum value? |
| TCP | Can two programs both use the same TCP port on the same machine at once? |
| UDP | Why is UDP checksum separate from IP checksum? |
| UDP | Suppose that I have decided to use UDP to transfer data to my application. My application needs to receive the data in the order it was transmitted. Will this be a problem with UDP? If it is, how can I solve the problem? |
| UDP | UDP is unreliable and connectionless protocol. Why do we use it? |
Final Exam
The class has chosen to write the final exam.
Each student is required to submit four unique, suitable exam questions. They should be “thinking” questions that are not too long for an exam. One can be homework (state which one that is). One should be easy, two of medium difficulty, and one difficult. Two are to be from period before the midterm, two after the midterm. Submit questions and answers.
Questions should be emailed to exam@csci6433.org.
The questions need to be unique. If a question is submitted that has already been approved, the submitter will be told and given a chance to resubmit.
All questions are to be submitted and approved by November 29, to provide time to study before the midterm.
Although the answers will not be posted, you are encouraged to study together in any way you wish, including posting and assembling a list of questions and answers.
| Topic | Question |
|---|---|
| Addressing | Suppose a computer is moved from one department to another. Does its physical address need to change? Does the IP address need to change? Why or why not? |
| Addressing | How can we determine if two IP addresses are in the same subnet? |
| ARP | What is the purpose of an ARP request? |
| ARP | ARP spoofing attack is a security issue in switching network environment, can you find a method to mitigate this? |
| DHCP | What is the purpose of DHCP? |
| DNS | Is it possible for an organization’s Web server and mail server to have exactly the same alias for a hostname? What would be the type for the RR that contains the hostname of the mail server? |
| DNS | What are the benefits of a caching DNS server? |
| FTP | How much data can be sent in a Single FTP session? |
| FTP | Why does FTP use two TCP connections between client and server? |
| General | Consider the process of mailing a physical letter from GWU to Grandma in Moscow, and compare it to the process of sending Grandma an IP datagram. What are the physical analogies of the following: destination MAC address, source MAC address, destination IP address, source IP address, datagram? |
| General | Give five important principles of the design of Internet protocols, and one example of the use of each principle in a protocol. |
| General | Give three examples of the Good Citizen Principle in internet protocols |
| General | Why is the internet successful? |
| General | Why do we call the Internet a stupid network? |
| General | What are the principal tradeoffs between a stupid network and a smart network? |
| General | Explain how do applications coexist over TCP and UDP? |
| General | Question 2 from midterm |
| General | Question 3 from midterm |
| General | Explain the extent of uniqueness of port numbers on a single machine. |
| General | Moses needed 10 commandments, but that covered all of life. To simply design Internet protocols, we can get by with five. Write Five Commandments of Internet Protocol Design. Each one starts with "Thou shalt" or "Thou shalt not". |
| General | From the viewpoint of protocol design, what is the most scarce resource on the Internet? |
| General | Which one is better, UDP or TCP? |
| HTTP | Why we say “HTTP is stateless”? |
| HTTP | What is a non-persistent TCP connection and what is a persistent TCP connection? What is the default mode of HTTP? |
| HTTP | How does a web proxy server handle HTTP requests for web objects? What general principle of Internet protocol design does this illustrate? |
| HTTP | What are the benefits of a persistent connection in HTTP? |
| HTTP | What's the difference between the POST and PUT commands? |
| ICMP | Which ICMP messages does ping use, and just how does it use them? |
| IGP | What are the two types of IGP, and how do they compare? |
| IMAP | How does IMAP4 guarantee consistency when multiple concurrent clients access a given mailbox at the same time? |
| IP | What do you mean when we say that IP is an unreliable protocol? |
| IP | Which part of the IP datagram is used for calculating the checksum? Why? What are the advantages and disadvantages of this design choice? |
| IP | Give four reasons why a datagram might be undeliverable. |
| IP | What is the role of IP in the TCP/UDP protocol suite? |
| IP | What is the role of IP in a TCP/IP protocol suite? What is the role of IP in a TCP/IP protocol suite? |
| IP | Fragmentation and reassembly are different for IPv4 and IPv6. Which is better? Why? |
| IPSec | What are the methods for IPsec implementation? |
| Layering | Sate one advantage and one disadvantage of layered network protocols. |
| Layering | Which layer does the network management protocol operated on? Why? |
| Multicast | Explain why Multicast is used. |
| Multicast | Does a multicasting router need to know how many subscribers are on a network that it connects to for a multicast group? Explain. |
| Multicast | IP software must always check the destination addresses on incoming multicast datagrams and discard datagrams if the host is not in the specified multicast group. Explain how the host might receive a multicast destined for a group to which that host is not a member. |
| NAT | Software implementations of NAT that are used to provide remote employees access to an organization’s intranet often reduce the network MTU that is reported to local applications. Explain why. |
| Routiing | What is the function of UDP on a router? |
| Routing | What does a router do if a packet is received with a time to live value of 1? |
| Routing | In RIP protocols, how does poison reverse work? |
| Routing | Why do link state protocols have better convergence rate than distance vector protocols? |
| Routing | If there is a router loop, how long can an IP datagram circulate in the loop? |
| SMTP | Is email through a gateway better or worse than a direct SMTP transfer? |
| SMTP | Simply describe what is IMAP. |
| SMTP | If you use a smartphone with a limit on the number of data bytes received each month, would you prefer POP3 or IMAP4? Explain. |
| SMTP | Is email through a gateway better or worse than a direct SMTP transfer? |
| SNMP | What is the purpose of the SNMP trap message? |
| SNMP | What does SNMP do? |
| SNMP | What are the basic components of SNMP? |
| TCP | What is the TCP windowing concept? |
| UDP | How much data can be sent in a single UDP session? |
| VPN | Under what circumstances will a VPN transfer substantially more packets than conventional IP when sending the same data across the Internet? (Hint: think about encapsulation.) |
| WebDAVWhich layer should network management protocol be operated on? why? | What are the three most important features of WebDAV? |