Index

1. TCP/IP Support Protocols

• are another reason TCP/IPs popularity

• 

2. IP datagram structure

• an IP datagram consists of a header parts and a text part
  • header has a 20 byte fixed part and a variable length optional part
• type of service field allows different combinations of reliability and speed to be chosen
  • for digital speech IP can be told to emphasize fast delivery
  • for file transfer is taking IP can be told that accuracy is paramount at the expense of speed

3. IP fields

4. IP fields

• total length
  • length of both header and data
• identification
  • all fragments of a datagram contain the same id value host can determine which datagram an incoming fragment belongs
• DF
  • do not fragment
• more fragments
  • All fragments except last one must have this bit set to true
• Fragment offset
  • must be a multiple of 8. Tells receiver where this fragment belongs in the datagram.
• Time to live
  • in seconds. Decrements each second or each hop, when it reaches 0 it is thrown away
• Protocol field
  • tells which of the various transport processes the datagram belongs. Ie TCP or UDP
• Header checksum
  • verifies header only
• Source and destination address
  • indicate the network number

5. IP addressing

• internet is a virtual structure
  • implemented entirely in software
  • packet frames and addresses were designed on merit
• addresses contained with 4 bytes
  • conceptually the 32 bit number has two parts
  • hostid
  • netid
• three primary classes of IP addresses

6. IP address classes

7. Address Classes

• class A
  • a handful of network which have more than 65536 hosts
• class B
  • addresses for intermediate size networks. 256..65535 hosts. 14 bits for netid 16 bits for hostid
• class C
  • networks which have less than 256 hosts
• class D
  • multicast, hosts may dynamically join/leave multicast group
  • hosts may be in many different multicast groups

8. Network structure

• consider the following IP network

  

• machine R is a router
  • has at least two IP numbers - one for each network card
• because Internet addresses encode both a network and a host on that network; they do not specify a host, but a connection to a network
  

9. Network and Broadcast addresses

• two reserved hostid’s
• Internet addresses can be used to refer to networks as well as network cards. By convention the network address has hostid all bits 0
• a broadcast address conversely has hostid bits all 1
• one of the weakness of IP addressing is that if a machine changes network - its IP address must change

10. Fragmentation and reassembly

• IP datagrams may be fragmented en route
  • if intermediate nodes cannot cope with a large datagram (MTU (maximum transmission unit) is smaller than datagram size)
• IP datagrams may be reassembled en route
  • although not a good idea as routing is dynamic. (So datagrams may not always travel the same route)
• to fragment a datagram into two a node creates two new datagrams with same fragment ids
  • the first offset is 0, MF = 1
  • the second offset is n, MF = 0

11. Fragmentation and reassembly

• consider trying to send a 1420 byte datagram when the MTU is 620
  • 1420 = 1400 data + 20 IP header
• split into 3 packets
  • first packet length = 620 = 20 new IP header + 600 old data, offset 0
  • second packet length = 620 = 20 new IP header + 600 old data, offset 600
  • third packet length = 220 = 20 new IP header + 200 old data, offset 1200
  • the new fragments have the same unique frag id as the original why?
• reassembly reverses this process

12. IP Support Protocols

• ICMP (Internet Control Message Protocol)
  • sends control information between the hosts
  • routers generate most of this information
• routers use ICMP to
  • inform hosts that a packet could not be delivered because of an error
  • or a better route exists to a particular destination
• ICMP messages are send using IP frames
  • ICMP messages use the IP protocol field and set it to 1

13. Typical ICMP messages are

• destination unreachable - when a router cannot find a routing table entry for the destination of an IP packet
• routing redirect - a router sends a routing redirect message to inform a host that a better route exists via another router
• time expired - message indicates a packets ttl field has reached 0
  • usually because of a configuration error
  • malfunctioning router
• echo request and echo reply - echo request messages request that the destination return the data in an echo reply message (ping)

14. Address Resolution Protocol

• IP address space is virtual and has no addressing relationship with the underlying datalink protocols
  • every network interface has an IP address
  • every network interface has a datalink address
  • datalink addresses vary in format and size
• suppose IP is sending a packet to a remote host on the same Ethernet
  • IP needs destination Ethernet address
  • could manually keep track of hosts and their interface card datalink addresses

15. Address Resolution Protocol

• clearly on a large network this becomes unmanageable
  • ARP (Address Resolution Protocol) is an automatic method which maps any network level address (IP address) to datalink address
  • ARP does this by exploiting the broadcast capability commonly found in most LAN datalink protocols

16. RARP (Reverse ARP)

• ARP maps from network addresses to datalink addresses
  • sometimes you require the opposite mapping
• many machines can read their datalink hardware to find out the datalink address
  • but then needs to find out its IP address
  • for example, disk less workstation, X terminal, printer

17. RARP (Reverse ARP)

• at least one host on the IP network must contain a list of IP addresses with corresponding datalink addresses
  • whereas ARP does not require that this list is present
  • a RARP is a broadcast request - any host may reply
• the sender fills in its datalink address
  • its network address is filled with zeros
  • specifies the target datalink address (normally the same as sender)
  • the RARP server fills in the requested IP (network) address

18. RARP (cont)

• RARP is normally the first step taken when a diskless workstation it powered up
  • once it knows its own IP address it can then proceed to load its operating system from a network server by using a simple file transfer protocol (TFTP)

19. TCP and UDP

• primarily there are two transport protocols used with IP: TCP and UDP
  • remember that IP may provide an unreliable service
• Transmission Control Protocol (TCP)
  • provides a flexible two-way byte stream protocol (byte stream allows addressing within a host - to user, process or service)
  • provides a bidirectional pipe
  • the source and destination address are called a Port
  • TCP is the most popular transport protocol on top of IP
  • it uses sliding window technique to provide a reliable service
  • it uses a three way handshake to establish a connection
  • and a two way handshake to disconnect

20. TCP

• typically it is used for:
  • mail
  • file transfer
  • remote login protocols
• technically it provides
  • a reliable duplex byte stream
  • eliminating duplicate packets
  • handling retransmission of lost packets
  • and ensuring data is delivered in order
• User Datagram Protocol (UDP)
  • is an unreliable datagram protocol and is deliberately simple
  • it does not ensure that packets arrive in order, un duplicated, or even at all!
  • it sends discrete datagrams, and delivers messages that arrive to the appropriate Port (same addressing schema as TCP)
  • a port may belong to a user, process or service
  • the standard Internet name service, DNS, uses UDP
  • it can be regarded as multiplexing many users, processes and services through one IP address
  • UDP has no standard connection procedure and no disconnect procedure

21. References

• refer the interested reader to explore UNIX BSD socket interface

  Stevens UNIX Network Programming, Prentice-Hall 1990

  Bach The design of the UNIX operating system, Prentice-Hall 1986
  

22. TCP/IP services

• Internet services are usually designed and implemented using the client/server model of computing
• divides the service into three distinct parts.

  

  Client/Server model

• each performs a distinct function in implementing the client/server model:
  • client process. The user who is using this service
  • server process. Maybe on another machine - providing the service
  • the protocol which the client and server are using to communicate
• note that this client server model is different to that discussed by business schools and management information systems as the client server model used by TCP/IP involves processes running on machines which are not dedicated as clients or servers
  • using TCP/IP users may dictate which machines are client or servers
  • one machine maybe a client for one application
  • and a server for another
• TCP/IP clients and servers are software processes
• whereas MIS treats clients and servers as machines (Ie mainframes and PCs)

23. TCP/UDP model

24. TCP Header format

• 

• HLEN 4 bits
  • header length = n × 32

• 

25. Services offered by TCP/IP

26. File Transfer Protocol

• uses TCP
• allows users to transfer files from one machine to another machine providing that the user has an account on both machines
• requires little configuration
• a special FTP service called anonymous FTP does not require users to have accounts on both machines but needs more configuration

27. Telnet

• uses TCP
• Virtual Terminal Protocol
  • was the Internet standard for remote login (now replaced by ssh)
• users can remotely login to another machine on the same LAN or any machine on the TCP/IP offering this service
  • eg. login to a machine in Japan
  • passwords are sent in plain text

28. ssh

• uses TCP
• a secure replacement for ftp and telnet
• both ftp and telnet should be retired if at all possible as they pass login name and password across the network in plain text..
• ssh also comes with scp
  • secure shell copy
• sftp, secure shell ftp
  • all username passwords and data contents are heavily encrypted

29. Services offered by TCP/IP

• inetd: the servers server
  • a host wanting to support many services could require a process for a server for each service
  • even on UNIX systems, the number of idle processes can quickly grow beyond a reasonable size, thus absorbing system resources unnecessarily (servers typically spend most of their time waiting for connections from clients)
  • inetd server accepts connections for a range of protocols and invokes the appropriate servers when needed

30. Services (continued)

• Network File System (NFS)
  • provides transparent file access over the network so that remote file appear as if they are local
  • NFS was designed for use over the LANs and is widely used in such environments
• NFS is widely available for different operating systems: UNIX, Macintosh, DOS clients exist.
  • (Ie at University. of Glamorgan. we use NFS between GNU/Linux, Sun OS)

31. Services (continued)

• Printer Spooling
  • many PC users still think erroneously that print spooling is a network’s only use
  • the most common UNIX print spooling protocol is Line Printing Protocol (LPP)" this was introduced in 4.2 BSD (released in 1981)
• allows a central server to be connected to a print resource
  • users can access this resource for other clients

32. Services (continued)

• System logging (syslog)
  • important for system administrators - it permits logging information to be directed to any host on the network.
• Remote Shell Protocol (RSH)
  • permits the user to run a shell command on the remote machine. (A form of remote job execution). It uses the same protocol as rcp (remote copy) and rlogin (remote login)
  • commonly known as the R commands

33. A few R commands

• rup
  • list all machines which are up and running and report their load average
• rusers
  • list all users logged in over the (local!) network
• rwho
  • list all (local) users and where they are logged in

34. More R commands

• netfind
  • try to find a particular user and give this persons email address.
• rcp
  • allows a user to copy files between machines ie:

    rcp merlin:/user/fred/foo lancelot:/user/mary/bar
    copying from source to destination
    

35. Naming and infrastructure

• there is a need for a mapping of textual domain names to numeric IP addresses
  • difficult to remember 193.63.130.52 is the class C address for floppsie!
• also require services such as a consistent time between different machines

36. Domain Name Service (DNS)

• the Internet standard DNS maps host names, such as floppsie.comp.glam.ac.uk to IP addresses such as 193.63.130.52
• DNS namespace is partitioned hierarchically into a tree
  • glam.ac.uk - may map onto several class C networks
  • floppsie.comp - indicates a machine within the computer studies network
  • an interface card on class C network 193.63.130.xx

37. Naming and infrastructure (continued)

• we could have a simple lookup table that is manually updated
  • soon becomes unmanageable
  • use a dynamic mechanism, domain name service
• have a machine which will keep track of IP addresses and ASCII names
  • if it cannot resolve a name it requests help from another machine higher up the tree
  • the DNS protocol specifies how DNS clients ask DNS servers for mappings
  • and how DNS servers communicate with each other.

38. Network Time Protocol

• used so that different hosts can keep the same time of day
• required by many applications - email to ensure that you don’t receive a message before it was sent!

39. Communication services

• many different services provided above TCP/IP
  • one of the most widespread use of computer networks is to communicate with other people
• one of the oldest utility is Electronic mail
  • users compose a message with a user agent (UA)
  • a user agent is an editor which passes the composed message to a MTA
  • one of the oldest is the UNIX program /bin/mail
  • one of the newest is outlook

40. Communication services

• these programs send the message to a message transfer agent (MTA)
  • typically this information is sent using SMTP (Simple Mail Transfer Protocol)
  • SMTP is the Internet standard protocol and all MTAs on the Internet use SMTP.
  • one of the dangers of SMTP is that it is relatively insecure (it is possible to forge email messages!)
• Network News Transfer Protocol (NNTP) protocol for delivering and accessing USENET news over Internet

41. Example IP configuration

• Case study - adding a machine onto the Computer Studies network
• the IP network in Computer Studies connects:
  • Apple computers
  • NT
  • SPARC machines
  • GNU/Linux
  • routers and various print services, etc
• each machine may run a different protocol above the IP layer if they wish
  • but most machines will run the IP protocol
• the Computer Studies IP network is connected (via a gateway) to the University of Glamorgan IP network
  • in turn is connected to the world IP network via another gateway (through the University of Glamorgan 1M bit line)

42. IP Configuration

• every IP address actually refers to the interface card and NOT the machine!
• thus a gateway machine will have at least two interface cards
• to add a new machine floppsie onto the Computer Studies network
  • floppsie’s interface card has to be assigned a unique IP address
  • first three numbers the same as the Computer Studies network (193.63.130)
  • class C network - means first 3 bytes are always the same
  • last number is the interface card number (hostid) 52
  • 193.63.130.52

43. IP Configuration

• software on the new machine needs to know:
  • the gateway on the Computer Studies network to other networks
  • its own interface card IP address!
  • the nameserver IP address. The nameserver translates all ASCII names to IP numerical addresses.
  • hop metric given with each gateway

44. IP Configuration

Index

This document was produced using groff-1.19.