Document revision date: 19 July 1999

Protocols that interact with the Internet Protocol provide services that manage data movement problems. The following sections introduce these protocols.

2.6.1 Internet Control Message Protocol (ICMP)

The Internet Control Message Protocol (ICMP) is an integral part of IP that provides error-reporting mechanisms used for regulating network performance. ICMP performs the following:

• Flow control. When datagrams arrive too quickly, a gateway sends an ICMP message telling the source to suspend its data transmission.
• Unreachable destination detection.
• Route redirection.
• Remote host checking to see if a remote system's IP is up and running.

ICMP provides feedback about problems in the communications environment, but does not make IP reliable. That is, ICMP does not guarantee that an IP packet will be delivered.

ICMP messages are sent in varying situations, including the following:

• When a packet cannot reach its destination
• When a gateway host does not have the buffering capacity to forward a packet
• When a gateway can direct a host to send traffic on a better route

SLIP is a minimal protocol used to send datagrams for transmission across a serial line such as a telephone circuit. (SLIP is not an Internet standard.) SLIP is used only when both hosts know each other's address and only when IP datagrams are being transmitted. When a host makes a connection, the SLIP server behaves like a router for TCP/IP traffic. Once connected, the SLIP host sends all network traffic over the serial interface.

Because network traffic is exchanged over the telephone link, performance is an issue for applications that handle numerous graphics, file sharing, or hypermedia. Compressed SLIP (CSLIP) improves SLIP performance by compressing the TCP/IP headers.

2.6.3 Point-to-Point Protocol (PPP)

PPP was developed as an Internet standard to address the weaknesses of SLIP. Often used for dialup remote LAN access, PPP allows the remote host to connect to the network and use IP network protocols. PPP is defined as a three-layered protocol as follows:

• Data Link Layer, a modified version of the High-Level Data Link Control (HDLC) that guarantees reliable delivery over any type of serial line.
• Link Control, used to establish the connection, negotiate configuration parameters (including compression), check link quality, and close the connection.
• Network Control Protocols, individual protocols that provide configuration and control information for the Network Layer protocols.

ARP translates Internet addresses into hardware addresses. (ARP does not translate addresses for the Serial Line Interface or PPP because SLIP and PPP have no hardware addresses.) ARP dynamically traces Internet addresses to hardware addresses on local area networks and stores this information in a mapping table.

ARP is not generally available to users or applications. When an application sends an Internet packet to an interface driver, the driver requests the appropriate address mapping. If the mapping is not in the table, an ARP broadcast packet is sent through the requesting interface driver to the hosts on the local area network.

When a host that supports ARP receives an ARP request packet, the host notes the IP and hardware addresses of the requesting system and updates its mapping table, if necessary. If the receiving host's IP address does not match the requested address, the host ignores the request packet. If the IP address does match, the receiving host sends a reply packet to the requesting system. The requesting system stores the new mapping and uses it to transmit future Internet packets.

Unlike most protocols, ARP packets do not have fixed-format headers. Instead, the message is designed to be useful with a variety of network technologies.

This chapter summarizes the characteristics of common TCP/IP applications. Software products supplied by Digital and other vendors implement these applications for OpenVMS systems. (See Appendix A for descriptions of currently available TCP/IP products that run as layered products on OpenVMS systems.)

Users on OpenVMS systems running TCP/IP software can invoke certain TCP/IP applications by specifying related OpenVMS DCL commands (as indicated in Table 3-1). Chapter 5 specifies the DCL command formats.

This chapter also describes software tools for retrieving information on the Internet.

3.1 Commonly Used TCP/IP Applications

Table 3-1 identifies commonly used network applications that are supported by almost all TCP/IP systems. The standard command, which is a UNIX syntax, is traditionally used with TCP/IP. Because this syntax can vary slightly among vendors, and because the syntax can be unfamiliar to OpenVMS users, an alternative syntax was developed jointly with the TCP/IP vendors for OpenVMS. This OpenVMS syntax is consistent with all TCP/IP vendor products and is a layer on OpenVMS commands. Note that the traditional UNIX style commands can also be used.

Table 3-1 Commonly Used TCP/IP Applications

Operation	Standard Name	Standard Command	OpenVMS DCL Command
Remote terminal service	RFC 854 and others, 1 Telnet	telnet	SET HOST/TELNET
	RFC 1282, BSD Rlogin	rlogin	SET HOST/RLOGIN
Remote file access	RFC 959, File Transfer Protocol (FTP)	ftp>get ftp>put	COPY/FTP
	BSD RCP 2	rcp	COPY/RCP
Remote directory listings	RFC 959, File Transfer Protocol (FTP)	ftp>ls	DIR/FTP
Point-to-point connections	RFC 1661 3 and others, Point-to-Point Protocol (PPP)	pppd	PPPD

TCP/IP applications follow a client/server model. The client is a program on a host that uses the services of a server located on either the same host or on a different host than the client. Among the most frequently encountered client/server user applications for TCP/IP are TELNET and FTP.

3.2 Remote Terminal Service

TCP/IP supports two virtual terminal protocols:

• TELNET, which is the RFC standard virtual terminal interface
• RLOGIN, which is the Berkeley standard virtual terminal protocol

Both the TELNET service and RLOGIN utility perform remote login operations, but in different ways.

3.2.1 Connecting to a Remote Server Using the TELNET Service

TELNET, running on a local computer, enables the user to connect to a specified host through a network that supports TCP/IP connections. TELNET can connect a computer to a remote computer host located anywhere in the world.

The TELNET service supports the telnet command, which allows a user on a client system to connect with a server on a remote system. The default server is the TELNET server that provides an interactive terminal session to execute commands at the remote host. The telnet command uses TCP as the transport protocol to guarantee accurate data delivery.

If the user connects to a TELNET server, a login session begins on the remote system. Once the connection is made, TELNET allows interaction between the user and the remote host. The user's keystrokes pass to the host and the remote computer displays pass to the local terminal. The local keyboard and screen seem to be connected to the remote computer.

In most cases, the user needs an account and password on the remote host in order to use TELNET for logging in to the remote system. Some software products that implement the telnet command also use Kerberos authentication to validate the user's identity (see Section 5.1.1) before granting access to the remote system. Many computers on the Internet provide some type of public TELNET access, permitting users to log in to a special guest account.

The telnet command models the local terminal and the remote terminal into network virtual terminals (simplified ASCII devices with keyboards and printers). Services required by either end of the connection are negotiated between the client and server.

TELNET also permits the local user to connect to an IBM mainframe host over a TCP/IP connection. A special TELNET client supports 3270 mode, which provides for IBM 3270 terminal emulation. When the TN3270 mode is active, the local keyboard emulates the keyboard normally used on an IBM 3270 class terminal. The remote IBM mainframe host must support the 3270 TELNET server.

In addition to the standard telnet command, OpenVMS users on systems running TCP/IP software can use the alternative OpenVMS style commands to make TELNET connections. For example, the DCL command SET HOST/TELNET invokes the TELNET client program (see Section 5.1.4). The DCL command SET HOST/TN3270 invokes the TN3270 terminal emulator client program (see Section 5.1.5).

3.2.2 Logging In to a Remote Host Using the RLOGIN Utility

The RLOGIN utility permits the user on a local host to run commands interactively on a remote host. The RLOGIN client program on the local host is connected to a remote RLOGIN server through a TCP/IP connection.

A user at a local host invokes the rlogin command. This command logs the user in to a remote host and starts an interactive terminal session. After the session starts, all commands the user enters at the local host are executed by the remote host, while the local host displays all of the output. The login session at the remote host can also be started with the name of a different user. If necessary, an 8-bit path can be established between the local and remote hosts.

The RLOGIN utility allows different hosts to share resources. When a host receives a request for connection, the RLOGIN server validates the source port number and the client user name. If the user validation fails, the RLOGIN server prompts for the user's password and checks the password file on the remote host. Some application products that implement the rlogin command use Kerberos authentication to validate the user's identity (see Section 5.1.1). The Kerberos authentication standard is RFC 1411.

In addition to the standard rlogin command, OpenVMS users on systems running TCP/IP software can use the alternative OpenVMS style commands to invoke RLOGIN programs. If you are logged in to an OpenVMS system that is running TCP/IP software, you can use the OpenVMS DCL command SET HOST/RLOGIN to invoke the RLOGIN client program (see Section 5.1.3).

3.3 Remote File Access

TCP/IP provides the following file access protocols:

• FTP, which is the RFC standard file transfer protocol
• RCP, which is the Berkeley file transfer protocol

Both FTP and RCP can copy files to and from remote hosts. In addition, FTP can manage remote directories. FTP services download or upload files over the Internet (see Section 3.5.1).

3.3.1 Transferring Files Between Hosts Using FTP

FTP (File Transfer Protocol) is a simple way to move files across a TCP/IP network. The ftp command invokes a utility that permits the user to transfer files between hosts that do not support the same file systems (for example, between UNIX and OpenVMS hosts). Normally, the user executing the ftp command must have a password on the remote host. Some systems use the " anonymous FTP " service, which accepts a userid of anonymous and no password. (The standard convention is to provide a complete electronic mail address for the password.)

The FTP utility permits the user to transfer groups of files between local and remote hosts. First, the user establishes an interactive connection with the FTP server on the remote host by providing a user name and password. Once the connection is established, the user can invoke ftp commands to transfer text and binary files to or from the remote host. Downloading of files from FTP sites on the Internet is described in Section 3.5.1.

In addition to the standard ftp command, OpenVMS users on systems running TCP/IP software can use the alternative OpenVMS style commands to invoke FTP services and perform copy operations. On OpenVMS systems running TCP/IP software, the DCL command COPY/FTP invokes the FTP service and performs a copy operation (see Section 5.2.3).

3.3.2 Listing Remote Host Directories Using FTP

The ftp command supports several file manipulation commands for managing remote directories. After establishing the FTP interactive session (as described in Section 3.3.1), the user can enter the ftp command dir or ls. This causes the remote system to list its directory contents on the local terminal. Other ftp commands provide directory operations such as creating, changing, or removing the remote directory and for other file operations over the TCP/IP network.

In addition to the standard ftp, dir, or ls commands, OpenVMS users on systems running TCP/IP software can use the alternative OpenVMS style commands to invoke FTP services that will list directories. On OpenVMS systems running TCP/IP software, the DCL command DIR/FTP invokes the FTP service and performs the directory listing operation (see Section 5.3.1).

3.3.3 Copying Files from Host to Host Using RCP

The RCP application consists of an RCP client utility running on the local host and an RCP server on each remote host involved in the copy operation.

The rcp command copies one or more files from one host to another or copies whole directory trees. Some implementations of the rcp command permit the local user to copy files from one remote host to another.

The RCP server validates the rcp command from the local host by checking whether the source host name is in the destination host's database, or by resolving the IP address in the Domain Name System. Some products that implement the rcp command also use Kerberos authentication (see Section 5.1.1).

In addition to the standard rcp command, OpenVMS users on systems running TCP/IP software can use the alternative OpenVMS style commands to invoke RCP services that will perform copy operations. OpenVMS users on systems running TCP/IP software can use the DCL command COPY/RCP to invoke the RCP utility (see Section 5.2.4).

3.4 Point-to-Point Protocol (PPP) Connections

The Point-to-Point Protocol utility (PPPD) allows you to initiate and control a PPP connection from an OpenVMS Alpha host. With PPPD, you can:

• Establish temporary network connections between remote hosts. This includes both dial-in capability from a remote host to an OpenVMS Alpha system and dial-out capability from an OpenVMS host to a remote system or server that supports PPP.
• Establish permanent, low-speed network connections between local hosts, such as between a laptop computer and an Alpha workstation connected by a serial data line.
• Set and display communication characteristics, such as address compression, flow control, and line speed.

OpenVMS users on systems running TCP/IP software can use the DCL command PPPD to invoke the PPP utility. PPPD commands allow users to set and display device characteristics; connect and disconnect a PPP session; and establish a connection with a new system. Chapter 6 describes the PPPD commands.

3.5 Retrieving Information Through the Internet

The Internet offers many resources for obtaining information from systems on the thousands of networks that form the Internet. Internet software tools are designed to operate over TCP/IP connections. These tools use many of the same client/server techniques as the common FTP and TELNET applications. In many instances, the tools for retrieving information are FTP and TELNET.

3.5.1 Downloading Files from FTP Sites on the Internet

FTP provides an important service to Internet users by allowing them to move large quantities of data across the Internet. FTP sites on the Internet provide huge data storage facilities, storing files of all types. You can browse through the names and descriptions of files at FTP sites and download the files to your computer. Many files found on an FTP server are stored in a compressed format. When you receive a file, you can decompress (expand) the file to its original size.

Many sites permit you to log in as an anonymous user, which means you do not need a password. Anonymous FTP allows an organization to distribute certain files (for example, electronic magazines) for free to the general public.

3.5.2 Using Browsers with the World Wide Web (WWW)

The World Wide Web (WWW) is a client/server hypermedia system that runs over the Internet. Developed by CERN in Switzerland, + the WWW is an information management facility adopted by thousands of sites worldwide. Users access the WWW by pointing their client software at any of thousands of server connections. Hypertext multimedia documents stored in the WWW contain links to other data. The hypertext format allows you to retrieve and display data based on keyword searches.

Software tools called browsers are available to access data on the WWW. The Netscape Navigator browser is a hypertext interface to the WWW that has links to files, images, text, audio, and video. Netscape Navigator, which is a graphical, single-client application, offers point-and-click menus.

When you run the Netscape Navigator and connect to the WWW, the first display is usually a home page. The home page is specific to the site you selected and often contains a welcome message and hyperlink indicators. These indicators are highlighted phrases or graphics, which may be in color or underlined or both, that link to more specific information. By clicking on the hyperlinks, you can access data anywhere on the web.

For character-cell terminals, you can use the Lynx browser to interface with the WWW. Keyword links are highlighted on the screen. You can advance the cursor to a keyword and jump to the linked document residing anywhere on the web.

The Netscape Navigator is available on the Internet Product Suite CD. WWW tools and a WWW server are available on the Freeware CD.

3.5.3 Using the Gopher Service to Access Internet Resources

The Gopher service is a client/server system that provides a simple, consistent means of accessing the full resources of the Internet. The Gopher is a highly automated software directory.

The Gopher client displays a series of menus from which you can make selections that the client carries out (for example, getting a file from a remote computer). Gopher menu items can lead to other menus at different sites.

Gopher servers are set up at many companies, universities, and other organizations to provide information of interest to local users. Many of the Gopher servers are public and supply information of interest to the general public. Gopher servers in many countries throughout the world are interconnected to form the Internet Gopher.

A Gopher client and server are available on the Freeware CD.

3.5.4 Sending Electronic Mail over the Internet

Once you are connected to the Internet, you can send electronic mail (or e-mail) messages over TCP/IP connections to systems throughout the Internet. You can also send mail to accounts that do not use Internet addressing; gateways translate the messages and send them to the appropriate networks and systems.

To send mail to an Internet address, use the format user@node.org according to the guidelines in the following list.

• If the Internet transport on your machine is not SMTP, define the logical name MAIL$INTERNET_TRANSPORT to select an alternate transport.
• If you specify a DECnet phase IV node or a DECnet-Plus alias, you may use the user@node syntax. In this case, the syntax is treated as NODE::USER.

The format is also available in character-cell mail and DECwindows mail. For more information about the format for specifying Internet mailing addresses see the OpenVMS User's Manual.

Internet host computers may also maintain lists called "mailing lists," which are databases of people who have shared interests in a topic. Sending e-mail to the mailing list causes the mail to be sent to everyone on the list. Examples of messages include articles, comments, and other information about the topic of the mailing list. You can get on a mailing list by sending e-mail to the mailing list administrator. Some mailing lists have moderators who screen the messages for duplication and inappropriate content.

	privacy and legal statement
6436PRO_002.HTML