19.1 Objective 1: Fundamentals of TCP/IP The TCP/IP suite of protocols was adopted as a military standard in 1983 and has since become the world standard for network communications on the Internet and on many LANs, replacing proprietary protocols in many cases. Much has been written about TCP/IP and the history of the Internet. This section includes only material cited by LPI Objectives. 19.1.1 Addressing and Masks The early specification of the Internet Protocol (IP) recognized that it would be necessary to divide one's given allotment of IP addresses into manageable sub-networks. Such division allows for distributed management, added security (fewer hosts can potentially snoop network traffic), and the use of multiple networking technologies (Ethernet, Token Ring, ATM, etc.). IP also enables convenient partitioning of the physical portions of a network across physical and geographical boundaries. To provide the capability to locally define networks, IP addresses are considered as having two distinct parts: the part that specifies a subnet and the one that specifies a network interface. The boundary between the network and host portions of an IP address is delineated by a subnet mask, required by the TCP/IP configuration of any network interface. Like the IP address, the subnet mask is simply a 32-bit number specified in four 8-bit segments using dotted quad decimal notation. The familiar class A, B, and C networks have these subnet masks: - Class A: 255.0.0.0 (binary 11111111.00000000.00000000.00000000)
-
8-bit network address and 24-bit host address - Class B: 255.255.0.0 (binary 11111111.11111111.00000000.00000000)
-
16-bit network address and 16-bit host address - Class C: 255.255.255.0 (binary 11111111.11111111.11111111.00000000)
-
24-bit network address and 8-bit host address When logically AND'd with an IP address, the bits set to in the subnet mask obscure the host portion of the address. The remaining bits represent the network address. For example, a host on a class C network might have an IP address of 192.168.1.127. Applying the class C subnet mask 255.255.255.0, the network addressof the subnet would be 192.168.1.0, and the host address would be 127, as depicted in Figure 19-1. While it is typical to use the predefined classes (A, B, and C), the boundary can be moved left or right in the IP address, allowing for fewer or more subnets, respectively. For example, if a single additional bit were added to the class C subnet mask, its IP address would be: - 255.255.255.128 (binary 11111111.11111111.11111111.10000000)
-
25-bit network address and 7-bit host address With such a subnet defined on an existing class C network such as 192.168.1.0, the 256-bit range is split into two subnets, each with seven host bits. The first of the two subnets begins at 192.168.1.0 (the subnet address) and continues through 192.168.1.127 (the subnet broadcast address). The second subnet runs from 192.168.1.128 through 192.168.1.255. Each of the two subnets can accommodate 126 hosts. To extend this example, consider two additional bits: - 255.255.255.192 (binary 11111111.11111111.11111111.11000000)
-
26-bit network address and 6-bit host address When applied to a class C network, four subnets are created, each with six host bits. Just as before, the first subnet begins at 192.168.1.0 but continues only through 192.168.1.63. The next subnet runs from 192.168.1.64 through 192.168.1.127 and so on. Each of the four subnets can accommodate 62 hosts. Table 19-1 shows more detail on class C subnets, considering only the host portion of the address. Table 19-1. Class C IP Subnet Detail 128 | 2 | 0 | 127 | 1 | 126 | 126 | | | | 128 | 255 | 129 | 254 | 126 | 252 | 192 | 4 | 0 | 63 | 1 | 62 | 62 | | | | 64 | 127 | 65 | 126 | 62 | | | | 128 | 191 | 129 | 190 | 62 | | | | 192 | 255 | 193 | 254 | 62 | 248 | 224 | 8 | 0 | 31 | 1 | 30 | 30 | | | | 32 | 63 | 33 | 62 | 30 | | | | 64 | 95 | 65 | 94 | 30 | | | | 96 | 127 | 97 | 126 | 30 | | | | 128 | 159 | 129 | 158 | 30 | | | | 160 | 191 | 161 | 190 | 30 | | | | 192 | 223 | 193 | 222 | 30 | | | | 224 | 255 | 225 | 254 | 30 | 240 | Be prepared to define network and host addresses when provided an IP address and a subnet mask. Practice with a few subnet sizes within at least one classification (A, B, or C). Also, because the use of decimal notation can cloud human interpretation of IP addresses and masks, be ready to do binary-to-decimal conversion on address numbers. | As you can see, as the number of subnets increases, the total number of hosts that can be deployed within the original class C address range reduces. This is due to the loss of both broadcast addresses and network addresses to the additional subnets. 19.1.2 Protocols TCP/IP is a suite of protocols, including the Transmission Control Protocol (TCP), Internet Protocol (IP), User Datagram Protocol (UDP), and Internet Control Message Protocol (ICMP), among others. Some protocols use handshaking (the exchange of control information among communicating systems) to establish and maintain a connection. Such a protocol is said to be connection-oriented and reliable, because the protocol itself is responsible for handling transmission errors, lost packets, and packet arrival order. A protocol that does not exchange control information is said to be connectionless and unreliable. In this context, "unreliable" simply means that the protocol doesn't handle transmission problems itself; they must be corrected in the application or system libraries. Connectionless protocols are simpler and have less overhead than connection-oriented protocols. TCP/IP is often said to be a stack of protocols, because protocols are built in a hierarchy of layers. Low-level protocols are used by higher-level protocols on adjacent layers of the protocol stack: - TCP
-
TCP is a connection-oriented transport agent used by applications to establish a network connection. TCP transports information across networks by handshaking and retransmitting information as needed in response to errors on the network. TCP guarantees packet arrival and provides for the correct ordering of received packets. TCP is used by many network services, including FTP, Telnet, and SMTP. By using TCP, these applications don't need to establish their own error-checking mechanisms, thus making their design simpler and easier to manage. - IP
-
IP can be thought of as the fundamental building block of the Internet. IP, which is connectionless, defines datagrams (the basic unit of transmission), establishes the addressing scheme (the IP address), and provides for the routing of datagrams between networks. IP is said to provide a datagram delivery service. Other higher-level protocols use IP as an underlying carrier. - UDP
-
UDP is a connectionless transport agent. It provides application programs direct access to IP, allowing them to exchange information with a minimum of protocol overhead. On the other hand, because UDP offers no assurance that packets arrive at destinations as intended, software must manage transmission errors and other problems such as missing and incorrectly ordered packets. UDP is used by applications such as DNS and NFS. - ICMP
-
ICMP is a connectionless transport agent that is used to exchange control information among networked systems. It uses IP datagrams for the following control, error-reporting, and informational functions: - Flow control
-
Sometimes inbound traffic becomes too heavy for a receiving system to process. In such cases, the receiving system can send a message via ICMP to the source instructing it to temporarily stop sending datagrams. - Detecting unreachable destinations
-
Various parts of network infrastructure are capable of detecting that a network destination is unreachable. In this case, ICMP messages are sent to the requesting system. - Redirecting routes
-
ICMP is used among network components to instruct a sender to use a different gateway. - Checking remote hosts
-
Hosts can transmit echo messages via ICMP to verify that a remote system's Internet Protocol is functioning. If so, the original message is returned. This is implemented in the ping command. - PPP
-
PPP is used for TCP/IP dialup network access via modem. The configuration and use of PPP is described later in Objective 4. You will need a general understanding of the control messages sent via ICMP. In particular, note that ICMP does not transmit data and that it is used by ping. | 19.1.3 TCP/IP Services When an inbound network request is made, such as that from a web browser or FTP client, it is sent to the IP address of the server. In addition, the request carries inside it a port number (or just port), which is a 16-bit value placed near the beginning of a network packet. The port number defines the type of server software that should respond to the request. For example, by default, web browsers send requests encoded for port 80. Web servers "listen" to port 80 and respond to incoming requests. The encoded port can be considered part of the address of a request. While the IP address specifies a particular host, the port specifies a specific service available on that host. Many port numbers are predefined, and the list is expanded as needed to accommodate new technologies. The official list of port number assignments is managed by the Internet Assigned Numbers Authority (IANA). The ports known by your system are listed in /etc/services. Port numbers 1 through 1023 are often referred to as privileged ports because the services that use them often run with superuser authority. Many of these, such as ports used for FTP (21), Telnet (23), and HTTP (80), are often referred to as well-known ports because they are standards. Port numbers from 1024 through 65535 (the maximum) are unprivileged ports and can be used by applications run by ordinary system users. During the initial contact, the client includes a local, randomly selected, unprivileged port on the client machine for the server to use when responding to the request. Client-to-server communications use the well-known port and the server-to-client communications use the randomly selected port. This Objective requires you to be familiar with the privileged port numbers detailed in Table 19-2. Table 19-2. Common Privileged Port Numbers 20 | FTP data | When an FTP session is opened, the binary or ASCII data flows to the server using port 20, while control information flows on port 21. During use, both ports are managed by an ftp daemon, such as wu-ftpd or PROftpd. | 21 | FTP control | 23 | Telnet server | Inbound Telnet requests are sent to server port 23 and processed by telnetd. | 25 | SMTP server | This port is used by mail transfer agents (MTAs) such as sendmail. | 53 | DNS server | Used by the Domain Name System server, named. | 67 | BOOTP/DHCP server | A BOOTP, or the more commonly used, DHCP server. | 68 | BOOTP/DHCP client | The client side for BOOTP/DHCP. | 80 | HTTP server | Web servers, such as Apache (httpd), usually listen in on this port. | 110 | POP3 | The Post Office Protocol (POP) is used by mail client programs to transfer mail from a server. | 119 | NNTP Server | This port is used by news servers for Usenet news. | 139 | NetBIOS | Reserved for Microsoft's LAN network manager. | 143 | IMAP | An alternate to POP3, IMAP is another type of mail server. | 161 | SNMP | Agents running on monitored systems use this port for access to the Simple Network Management Protocol. | This list is a tiny fraction of the many well-known ports, but it may be necessary for you to know them both by name and by number. You should commit the list of ports in Table 19-2 to memory so you can recognize a type of network connection solely by its port number. Your exam is likely to have at least one question on how a specific port is used. | 19.1.4 TCP/IP Utilities The following popular applications, while not strictly a part of TCP/IP, are usually provided along with a TCP/IP implementation. Syntaxdig hostname Description dig obtains information from DNS servers. Note that additional command-line arguments and options are available for dig but are beyond the scope of Exam 102. Example$ dig redhat.com ; <<>> DiG 8.2 <<>> redhat.com any ;; res options: init recurs defnam dnsrch ;; got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 6 ;; flags: qr rd ra; QUERY: 1, ANSWER: 6, AUTHORITY: 4, ;; ADDITIONAL: 5 QUERY SECTION: ;; redhat.com, type = ANY, class = IN ;; ANSWER SECTION: redhat.com. 22h36m45s IN NS ns.redhat.com. redhat.com. 22h36m45s IN NS ns2.redhat.com. redhat.com. 22h36m45s IN NS ns3.redhat.com. redhat.com. 22h36m45s IN NS speedy.redhat.com. redhat.com. 23h48m10s IN MX 10 mail.redhat.com. redhat.com. 23h48m10s IN A 207.175.42.154 ;; AUTHORITY SECTION: redhat.com. 22h36m45s IN NS ns.redhat.com. redhat.com. 22h36m45s IN NS ns2.redhat.com. redhat.com. 22h36m45s IN NS ns3.redhat.com. redhat.com. 22h36m45s IN NS speedy.redhat.com. ;; ADDITIONAL SECTION: ns.redhat.com. 1d23h48m10s IN A 207.175.42.153 ns2.redhat.com. 1d23h48m10s IN A 208.178.165.229 ns3.redhat.com. 1d23h48m10s IN A 206.132.41.213 speedy.redhat.com. 23h48m10s IN A 199.183.24.251 mail.redhat.com. 23h48m10s IN A 199.183.24.239 ;; Total query time: 81 msec ;; FROM: smp to SERVER: default -- 209.195.201.3 ;; WHEN: Wed Apr 5 03:15:03 2000 ;; MSG SIZE sent: 28 rcvd: 275 Syntaxftp [options] host ...interactive commands... Description Establish an interactive File Transfer Protocol (FTP) connection with host in order to transfer binary or text files. FTP creates an interactive dialog and allows for two-way file transfer. The dialog includes username/password authentication, user commands, and server responses. Frequently used options - -i
-
Turns off interactive prompting during multiple file transfers (also see the prompt command). - -v
-
Sets verbose mode, displays server responses and transfer statistics. Frequently used commands - ascii, binary
-
Establish the transfer mode for files. ASCII mode is provided to correctly transfer text among computer architectures where character encoding differs. - get file
-
Receive a single file from the server. - mget files
-
Receive multiple files from the server. - ls [files]
-
Obtain a directory listing from the server, optionally listing files. - put file
-
Send a single file to the server - mput files
-
Send multiple files to the server. - prompt
-
Toggle on and off interactive prompting during mget and mput (also see the -i option). - pwd
-
Print the working remote directory. - quit, exit
-
Cleanly terminate the FTP session. Example 1 Get a file from machine smp: $ ftp -v smp Connected to smp. 220 smp FTP server (Version wu-2.4.2-VR17(1) Mon Apr 19 09:21:53 EDT 1999) ready. Name (smp:root): jdean 331 Password required for jdean. Password:<password here> 230 User jdean logged in. Remote system type is UNIX. Using binary mode to transfer files. ftp> ls myfile 200 PORT command successful. 150 Opening ASCII mode data connection for /bin/ls. -rw-r--r-- 1 jdean jdean 29 Jan 24 01:28 myfile 226 Transfer complete. ftp> binary 200 Type set to I. ftp> get myfile local: myfile remote: myfile 200 PORT command successful. 150 Opening BINARY mode data connection for myfile (29 bytes). 226 Transfer complete. 29 bytes received in 0.000176 secs (1.6e+02 Kbytes/sec) ftp> quit 221-You have transferred 29 bytes in 1 files. 221-Total traffic for this session was 773 bytes in 3 transfers. 221-Thank you for using the FTP service on smp. 221 Goodbye. Example 2 Many FTP servers are set up to receive requests from nonauthenticated users. Such public access is said to be anonymous. Anonymous FTP is established just like any FTP connection, except that anonymous is used as the username. An email address is commonly used as a password to let the system owner know who is transferring files: # ftp -v smp Connected to smp. 220 smp FTP server (Version wu-2.4.2-VR17(1) Mon Apr 19 09:21:53 EDT 1999) ready. Name (smp:root): anonymous 331 Guest login ok, send your complete e-mail address as password. Password: me@mydomain.com 230 Guest login ok, access restrictions apply. Remote system type is UNIX. Using binary mode to transfer files. ftp> <commands follow...> Syntaxping hostname Description The ping command is used to send an ICMP echo request to hostname and report on how long it takes to receive a corresponding ICMP echo reply. Much like sonar systems send a pulse (or "ping") to a target and measure transit time, ping sends a network packet to test the availability of a network node. This technique is often used as a basic debugging technique when network problems arise. 19.1.4.1 Example Ping a remote host and terminate using Ctrl-C after five packets are transmitted: $ ping lpi.org PING lpi.org (209.167.177.93) from 192.168.1.30 : 56(84) bytes of data. 64 bytes from new.lpi.org (209.167.177.93): icmp_seq=0 ttl=240 time=51.959 msec 64 bytes from new.lpi.org (209.167.177.93): icmp_seq=1 ttl=240 time=60.967 msec 64 bytes from new.lpi.org (209.167.177.93): icmp_seq=2 ttl=240 time=47.173 msec 64 bytes from new.lpi.org (209.167.177.93): icmp_seq=3 ttl=240 time=46.887 msec 64 bytes from new.lpi.org (209.167.177.93): icmp_seq=4 ttl=240 time=46.836 msec --- lpi.org ping statistics --- 5 packets transmitted, 5 packets received, 0% packet loss round-trip min/avg/max/mdev = 46.836/50.764/60.967/5.460 ms Syntaxtelnet [host] [port] Description Establish a connection to host (either a system name or IP address) using port. If a specific port is omitted, the default port of 23 is assumed. If host is omitted, telnet goes into an interactive mode similar to ftp. Syntaxtraceroute hostname Description Attempt to display the route over which packets must travel to reach a destination hostname. It is included here because it is mentioned in this Objective, but Objective 3 also requires traceroute. See the synopsis in Objective 3 for full information. Syntaxwhois target[@server] fwhois target[@server] Description Query the whois database for target. Such a database contains information on domain names, assigned IP addresses, and people associated with them. In the early days of the Internet, when domain registration was handled solely by the Internet Network Information Center (InterNIC), server was understood to be that of the InterNIC. Additional registrars now exist to process domain registrations, some of which have their own whois databases for public access. The version of whois provided with Linux is a link to fwhois. target is a domain name or user handle. server is a valid whois server, which defaults to rs.internic.net. The information returned includes contact information, domain names, IP addresses, and DNS servers. Note that many web sites are available for whois searches as well, particularly for checking on domain name availability. Example$ fwhois linuxdoc.org@whois.networksolutions.com Registrant: Linux Documentation Project (LINUXDOC-DOM) 4428 NE 74th Ave. Portland, OR 97218 US Domain Name: LINUXDOC.ORG Administrative Contact, Technical Contact, Zone Contact: Account, Hostmaster (AH243-ORG) hostmaster@LINUXPORTS.COM Command Prompt Software 4428 NE 74th Ave. Portland, OR 97218 US (503)493-1611 Billing Contact: Account, Hostmaster (AH243-ORG) hostmaster@LINUXPORTS.COM Command Prompt Software 4428 NE 74th Ave. Portland, OR 97218 US (503)493-1611 Record last updated on 15-Feb-2000 Record created on 20-Feb-1999 Database last updated on 5-Apr-2000 12:51:28 EDT Domain servers in listed order: NS1.OPENDOCS.ORG 209.102.107.110 NS1.INETARENA.COM 206.129.216.1 NS.UNC.EDU 152.2.21.1 You must have a working knowledge of when and how to use the dig, ftp, ping, telnet, traceroute, and whois commands. Practice using any that you are unfamiliar with by experimenting on a working networked system. | |