Part 1. Answer True or False (1 point each)
___ Distance Vector Routing assumes that each router knows the entire graph
of the network.
___ Link-State routing uses Dijkstra algorithm to compute the shortest routes.
___ Connection oriented networks keep state at both communication ends.
___ The usage of a network may affect the communication delay.
___ The Internet is formed by homogeneous networks.
___ The size of the IP address in IP version 4 is always 4 bytes.
___ Each computer in the Internet has at most one IP address.
___ There are 65536 class B addresses.
___ 192.5.48.3 is a class B address.
___ 10.2.0.32 is a class A address.
___ There are 256 host numbers in a class C address.
___ 255.255.255.255 is a valid IP destination address.
___ A "ping" to host 127.0.0.1 will never return "alive".
___ 128.10.255.255 is a valid IP host address.
___ The size of the IP ID number is 8 bits.
___ The maximum size of an IP packet is 1500 bytes.
___ The length of an Ethernet address is 6 bytes.
___ Backbone routers know the subnet masks of subnets.
___ Default routes are mostly used in the backbone routers of the Internet.
___ TCP is a "best delivery protocol".
___ The ARP request and response are multicasted.
___ ARP requests are cached for about 20 minutes.
___ Routers perform fragmentation and reassembly of packets larger than the
network MTU.
___ Hosts do not request missing IP fragments.
___ A transport protocol may be unreliable.
___ The size of the ethernet header is 14 bytes not including CRC.
___ The size of the IP header and the TCP header is the same.
___ Changing the mapping of an IP address to a name in DNS take effect
immediately in the Internet.
___ DNS caching is done by both servers and clients.
___ The IP destination address in the IP packet changes every time the packet
reaches a router.
Part 2. Short Questions: (4 points each)
1. Explain when the TTL field in the IP packet is changed and why this field is important.
2. Enumerate and describe the fields needed for IP reassembly.
3. Enumerate and describe the layers in the ISO-7 layer reference model.
4. Enumerate and describe the layers used in the TCP/IP protocols.
5. Explain the advantages/disadvantages of TCP vs. UDP and when each of these protocols should be used.
6. What does the Delay Throughput product mean?
7. Explain the advantages and disadvantages of using Remote Procedure Calls vs. sockets.
8. Enumerate and describe four ICMP message types.
9. Explain how traceroute and path MTU discovery work.
10. Explain how TCP achieves reliability.
11. Explain what is TCP adaptive retransmission.
12. Explain how TCP flow of control works.
13. Explain how TCP handles network congestion.
14. Write the diagrams of the three way handshake for opening and closing a connection.
15. What are the two kinds of DNS resolution and describe them.
Part 3. Programming. (10 points)
16. Write the procedure handle_cgibin( char * request, int ssocket ), where request contains the program requested by the client in the form "/cgi-bin/prog?a=b&c=d...". ssocket is the slave socket returned by accept(). You don't need to write the program in detail but don't forget to mention the important steps.