Chapter 13, TCP/IP Administration
|1| Chapter Overview
A. TCP/IP Overview
B. Configuring TCP/IP Clients
C. Using TCP/IP Utilities
Chapter 13, Lesson 1
|2| TCP/IP Overview
1. Introduction
A. TCP/IP is an industry-standard suite of protocols used on local area networks (LANs) and wide area networks (WANs).
B. Windows 2000 supports TCP/IP, both as a protocol suite and as a set of services for connectivity and management of IP networks.
|3| 2. Architectural Overview of the TCP/IP Protocol Suite
A. The TCP/IP protocols
1. Provide networking connectivity support for computers (called hosts) on LANs and WANs
2. Follow a set of standards for how computers communicate and how networks are interconnected
B. The TCP/IP protocols follow a four-layer conceptual model known as the Department of Defense (DOD) model.
|4| 1. The four layers, from top to bottom, are
a. Application
b. Transport
c. Internet
d. Link
|5| C. Application layer
1. Software programs gain access to the network through the application layer.
2. This layer maps roughly to the session, presentation, and application layers of the Open Systems Interconnection (OSI) reference model.
3. TCP/IP utilities and services that run at the application layer
a. Hypertext Transfer Protocol (HTTP): used for Web-based communications
b. File Transfer Protocol (FTP): an Internet service that transfers files from one computer to another
c. Simple Mail Transfer Protocol (SMTP): used by mail servers to transfer e-mail
|6| d. Telnet: a terminal emulation protocol used to log on to remote network hosts
e. Domain Name System (DNS): a set of protocols and services on a TCP/IP network that lets network users use hierarchical, user-friendly names when locating hosts instead of remembering and using their IP addresses
f. Simple Network Management Protocol (SNMP): lets you manage network nodes such as servers, workstations, routers, bridges, and hubs by gathering information using SNMP agents and displaying it on a central host computer
4. Microsoft TCP/IP also provides two interfaces for network applications to use the services of the TCP/IP protocol stack:
a. WinSock: the Windows 2000 implementation of the Sockets application programming interface (API)
b. NetBIOS: a standard API used as an interprocess communication (IPC) mechanism in the Microsoft Windows environment
|7| D. Transport layer
1. The transport protocols
a. Provide communication sessions between computers
b. Define the type of transport service as either connection-oriented or connectionless datagram-oriented
2. This transport layer maps roughly to the transport layer in the OSI model.
3. Transport layer protocols
a. TCP: provides connection-oriented, reliable communications for applications that typically transfer large amounts of data at one time
(1) Also used for applications that require an acknowledgment for data received by the destination computer
b. UDP: provides connectionless communications and does not guarantee packet delivery
(1) Applications that use UDP typically transfer small amounts of data at one time in a request/reply format, leaving the reliable delivery of data up to the application.
|8| E. Internet layer
1. The protocols at the Internet layer encapsulate transport layer data into units called datagrams, address them, and route them to their destinations.
a. In essence, datagrams are the envelopes that carry the messages generated by the protocols at the upper layers.
b. These routing functions are necessary to enable hosts to interoperate with other networks.
2. This Internet layer maps roughly to the network layer in the OSI model.
3. Windows 2000 implements three main protocols at this layer:
a. IP: the primary Internet layer protocol, responsible for addressing and routing packets
(1) All user data generated by applications is transmitted inside IP datagrams.
b. Address Resolution Protocol (ARP): responsible for converting the IP addresses of outgoing packets into the hardware addresses needed to transmit them over a network using a LAN protocol, such as Ethernet or Token Ring
c. Internet Control Message Protocol (ICMP): a protocol that IP uses to transmit error messages generated by Internet layer processes and to perform diagnostic functions on the network
|9| F. Link layer
1. Specifies the physical requirements for data transmission
2. Each of the LAN, metropolitan area network (MAN), and WAN types (such as Ethernet, Token Ring, and Fiber Distributed Data Interface [FDDI]) have different requirements for cables, signaling, and data encoding.
3. The link layer sends and receives frames (packets of information transmitted on a network as a single unit).
a. Puts frames on the network and pulls frames off the network
4. Equivalent to the data-link and physical layers of the OSI model
5. On LANs, the protocols responsible for link layer functionality are generally not part of the TCP/IP suite.
6. Two protocols often used for WAN connections are part of the TCP/IP suite:
a. Point-to-Point Protocol (PPP)
b. Serial Line Internet Protocol (SLIP)
7. PPP and SLIP are relatively simple protocols because they are designed for communications on a dedicated link between only two computers.
8. Windows 2000 supports both PPP and SLIP, but PPP is more common because it provides greater security, configuration handling, and error detection than SLIP does.
9. Windows 2000 supports only SLIP client functionality, not SLIP server functionality.
a. The Windows 2000 Routing and Remote Access Service does not accept SLIP client connections.
|10| 3. Transmission Control Protocol
A. A reliable, connection-oriented delivery service
1. In a reliable service, messages transmitted back to the sender acknowledge the data received by the destination system.
2. In a connection-oriented service, two computers must establish a connection before they can exchange data.
a. Ensures that the computers are ready to transmit and receive data before any packets are sent
b. TCP uses byte-stream communications, which means that the data is treated as a sequence of bytes.
B. Achieves reliability by assigning a sequence number to each transmitted segment
1. If a segment is broken into smaller pieces in transit, the receiving host knows whether it has received all of the pieces.
2. An acknowledgment message verifies that the receiving host received the data.
a. The receiving host must return an acknowledgment (ACK) for the segments transmitted to it within a specified period.
b. If the sender does not receive an ACK for particular segments, it retransmits the data.
(1) This system is called positive acknowledgment with retransmission.
c. If the segment is damaged, the receiving host discards it without sending an ACK.
(1) Because an ACK is not sent, the sender retransmits the segment.
C. Many Internet client applications, such as Web browsers and FTP clients, rely on TCP to transmit files.
D. Virtually every application that transmits large amounts of data over a network uses TCP.
|11| 4. User Datagram Protocol
A. Provides a connectionless datagram service that does not guarantee delivery or the correct sequencing of delivered packets
B. Can use checksums for error correction
1. This service is optional.
C. Used by applications that
1. Do not require an acknowledgment of data receipt
2. Usually transmit small amounts of data at one time
D. Used by
1. Broadcast transmissions
2. Services and applications such as DNS, Dynamic Host Configuration Protocol (DHCP), and SNMP
E. Also used in some high-bandwidth situations, such as streaming media, which uses application layer protocols to ensure data delivery
|12| 5. Internet Protocol
A. TCP separates data into discrete segments and is responsible for guaranteeing their delivery, but IP actually delivers the packets.
B. Each incoming or outgoing packet is referred to as a datagram.
C. IP adds the following header fields to each packet when it is passed down from the transport layer:
1. Source IP Address: identifies the sender of the datagram by the IP address
2. Destination IP Address: identifies the ultimate destination of the datagram by the IP address
3. Protocol: informs IP at the destination host whether to pass the packet up to the TCP or UDP protocol at the transport layer
4. Checksum: a simple mathematical computation that verifies the packet’s IP header information has not been damaged in transit
5. Time to Live (TTL): specifies the number of routers that a datagram can pass through on the way to its destination before it is discarded.
a. The default TTL value in Windows 2000 is 128.
|13| D. IP addressing
1. Every host on a TCP/IP network must have a unique IP address.
a. An IP address is a 32-bit number that identifies both the host and the network the host is located on.
2. The IP address is strictly a network layer address.
a. It does not depend on data-link layer addresses, such as the hardware address coded into network interface adapters.
3. IP addresses are expressed in dotted-decimal format, such as 192.168.123.132.
a. Each of the four numbers (called bytes, octets, or quads) represents eight bits of the binary address, ranging from 00000000 to 11111111, or, in decimal notation, from 0 to 255.
4. The division of the IP address into a network identifier and a host identifier makes it possible for TCP/IP to route data packets to other networks.
|14| 5. An IP address in Windows 2000 (and on most operating systems) is accompanied by a second 32-bit number called a subnet mask.
a. The subnet mask looks similar to an IP address because it uses the same dotted-decimal notation, but it has a different function.
6. The subnet mask, when compared to the IP address, identifies the part of the IP address that is the network identifier and the part that is the host identifier.
a. In a subnet mask, 1s identify network bits and 0s identify host bits.
b. Thus, in the subnet mask 255.255.0.0, the first 16 bits (or two octets) are the equivalent of all 1s in binary form.
(1) They indicate that the first two octets of the IP address are the network identifier.
(2) The last 16 bits (or two octets) represent the host identifier because they are both 0s.
|15| E. Understanding IP address classes
1. To communicate on the Internet, a computer must use an IP address that is registered with the Internet Assigned Numbers Authority (IANA).
a. In practice, you obtain a valid, registered network address from your Internet Service Provider (ISP), not directly from the IANA.
2. The IANA assigns network identifiers only.
a. The network administrator must assign a unique host identifier to each computer on the network.
|16| 3. Three primary classes of network addresses: A, B, and C
a. The actual class used is based on the size of the network.
b. Each address class has a different default subnet mask because it uses a different number of bits as the network identifier.
|17| 4. Characteristics of the three IP address classes
a. You can identify the class of an IP address by looking at the value of its first octet.
b. Because a Class A address uses only eight bits for its network identifier, only 126 possible Class A addresses exist.
|18| 5. General guidelines for planning IP address assignments
a. The first octet of the network identifier (ID) cannot be 127.
(1) This value is reserved for loopback and diagnostic functions.
b. The network and host ID bits cannot be all 1s.
c. The network and host ID bits cannot be all 0s.
d. The host ID for each computer must be unique to the local network identifier.
e. A unique network ID is required for each network and wide area connection.
(1) If you are connecting to the Internet, a registered network ID is required.
f. All TCP/IP hosts, including interfaces to routers, require unique host IDs.
(1) The host ID of the router is the IP address configured as a workstation’s default gateway.
g. Each host on a TCP/IP network requires a subnet mask.
|19| F. Using private addresses
1. Workstations that are protected by firewalls and computers that do not connect to the Internet can use unregistered, private IP addresses.
a. An administrator can assign these IP addresses without obtaining them from an ISP or the IANA.
2. When building your own private network, you should use one of the special ranges of private IP addresses, rather than assigning IP addresses randomly.
|20|
a. Class A can use 10.0.0.0 through 10.255.255.255.
b. Class B can use 172.16.0.0 through 172.31.255.255.
c. Class C can use 192.168.0.0 through 192.168.255.255.
|21| 6. IP Routing
A. Routing is the process of choosing a path to send packets over.
1. Routing is the primary function of IP.
B. A router (also called a gateway) is a device that forwards packets from one physical network to another.
1. Helps LANs and WANs achieve interoperability and connectivity
2. Can link LANs that have different network topologies, such as Ethernet and Token Ring
C. Each packet sent over a LAN has a packet header that contains source and destination address fields.
D. Routers match packet headers to a LAN segment and choose the best path for the packet, which optimizes network performance.
1. The best path is the path with the fewest number of hops.
|22|
|23| E. To make routing decisions, the IP layer consults a routing table, which consists of a series of entries, called routes.
1. The routing table is stored in memory on the computer.
2. The routes show where network identifiers are located.
3. On a computer running Windows 2000, a routing table is built automatically based on its TCP/IP configuration.
4. To view a routing table on a computer running Windows 2000, type route print at the command prompt, and then press Enter.
|24| F. Static and dynamic IP routing
1. The process that routers use to obtain routing information differs based on whether the router performs static or dynamic IP routing.
2. Static routing
a. Is a function of IP that uses only fixed routing tables
b. Static routers require manual building and updating of routing tables.
c. You can use the Route.exe command to add static entries to the routing table on a computer running Windows 2000.
3. Dynamic routing
a. Automatically updates routing tables
(1) Uses routing protocols, such as the Routing Information Protocol (RIP) and Open Shortest Path First (OSPF), to periodically transmit the contents of their routing tables to the other routers on the network
4. To function as a dynamic router, a computer running Windows 2000