|
|
This chapter describes how to make terminal-to-host connections using a communication server. To make protocol translation connections, refer to Chapter 4, "Connecting to a Host through a Protocol Translator." You can make the following types of connections from a terminal connected to the communication server:
These tasks are outlined in the following sections. See the end of this chapter for examples of XRemote connections on an X display terminal.
Digital Equipment Corporation's (Digital's) Local Area Transport (LAT) protocol is most often used to connect communication servers to Digital hosts. LAT is a Digital-proprietary protocol. Your communication server uses LAT technology licensed from Digital. You can perform the following LAT connection tasks from your communication server:
These tasks are described in the following sections.
To connect to a LAT host, enter the lat EXEC command at the system prompt. The lat command has the following syntax:
lat name [node nodename | port portname | /debug]| name | LAT learned service name. |
| node nodename | (Optional.) Specifies a connection to a particular LAT node that offers a service. If you do not include the node option, the node with the highest rating offering the service is used. Use the EXEC command show lat nodes to display information about all known LAT nodes. |
| port portname | (Optional.) Specifies a destination LAT port name. This keyword is ignored in most timesharing systems, but is used by communication servers offering reverse LAT services. Reverse LAT involves connecting to one communication server from another. In this case, the target communication server runs the host portion of the protocol. Enter the port name in the format of the remote system in place of the portname argument. |
| /debug | (Optional.) A switch that, when enabled, prints parameter changes and other special messages on the terminal. |
The LAT protocol must be explicitly specified when your preferred transport is set to none or to another protocol. (If your preferred transport is set to lat, you can use the connect command.) You can quit the connection by entering Ctrl-C, or complete the connection by entering the password for a given service.
The following example establishes a LAT connection from the communication server named cs to the host eng2:
cs> lat eng2
Trying ENG2...Open
ENG2 - VAX/VMS V5.2
Username: JSmith
Password:
Welcome to VAX/VMS version V5.2 on node ENG2
Last interactive login on Friday, 6-APR-1990 19:46
The system informs you of its progress by displaying the messages "Trying <system>..." and then "Open." If the connection were not successful, you would receive a failure message.
The following example establishes a LAT connection from the communication server named cs to the our-modems service and specifies port 24, which is a special modem:
cs> lat our-modems port 24
The following example establishes a LAT connection from the communication server named cs to the our-modems service and specifies a node named eng:
cs> lat our-modems node eng
The following example uses the LAT session debugging capability:
cs> lat Eng2 /debug
Trying ENG2...Open
ENG2 - VAX/VMS V5.2
Username: JSmith
Password:
Welcome to VAX/VMS version V5.2 on node ENG2
Last interactive login on Tuesday, 10-APR-1990 19:02
[Set Flow out off, Flow in on, Format 8:none, Speed 9600/9600]
[Set Flow out off, Flow in on, Format 8:none, Speed 9600/9600]
$ set ter/speed=2400
[Set Flow out off, Flow in on, Format 8:none, Speed 2400/2400]
A variety of LAT events are reported, including all requests by the remote system to set local line parameters. The messages within brackets ([ ]) above are the messages produced by the remote system setting line characteristics to operating system defaults.
You can temporarily define the list of services to which you or another user can connect. You do so by defining the group code lists used for connections from specific lines The group code range entered must be a subset of the line's configured group's code range. Use the following command:
terminal lat out-group {groupname | number | range}| groupname | Specifies the name of the group that has access to the system through the specified line. |
| number | Specifies the number of the group that has access to the system through the specified line. |
| range | Indicates the range of group numbers. Separate the range of numbers with a hyphen. |
cs> terminal lat out-group 4, 6-189
You can have several concurrent LAT sessions open and switch back and forth between them.
To open a subsequent session, first enter the escape sequence to quit out of the current session, which by default is Ctrl-^ X, then open a new session.
To list the available LAT services, issue the following command:
show lat servicesFor sample output to this command, see Chapter 5, "Managing and Monitoring Connections."
For more information about switching between LAT sessions, refer to Chapter 5, "Managing and Monitoring Connections."
Your communication server supports a subset of Digital commands, including the following:
| Task | Command |
|---|---|
| List EXEC commands. | help |
| Close the active session. | logout |
You can issue any of the following commands to terminate an active LAT session:
exit quit logoutTwo Transmission Control Protocol/Internet Protocols (TCP/IP), Telnet and rlogin, are available for making connections to a host.
Telnet, a virtual terminal protocol that is part of the TCP/IP protocol suite, allows for connections to hosts. Telnet is the more widely used protocol.
The rlogin protocol is a remote login service developed for the BSD UNIX system. It provides better control and output suppression than Telnet, but can only be used when the host (typically, a UNIX system) supports rlogin. Our implementation of rlogin does not subscribe to the trusted host model. That is, a user cannot automatically log onto a UNIX system from the communication server, but must provide a user ID and a password for each connection.
This implementation of Telnet and rlogin provides these connections services:
These tasks are described in the following sections.
To log into a host that supports Telnet, enter either of the following commands:
connect host [port] [keyword] telnet host [port] [keyword]| host | A host name or an Internet address. |
| port | (Optional.) A decimal TCP port number; the default is the Telnet server port (decimal 23) on the host. |
| keyword | (Optional.) One of the options listed in Table 3-1. |
| Option | Description |
|---|---|
| /route path | Specifies loose source routing. The argument path is a list of host names or Internet addresses that specify network nodes, ending with the final destination. |
| /line | Enables Telnet line mode. In this mode, the communication server does not send any data to the host until you press Return. You can edit the line using the standard communication server command editing characters (Backspace, Delete, Ctrl-U, Ctrl-W). The /line keyword is a local switch; the remote server is not notified of the mode change. |
| /debug | Enables Telnet debugging mode. |
| /stream | Turns on stream processing, which enables a raw TCP stream with no Telnet control sequences. A stream connection does not process Telnet options, and may be appropriate for connections to ports running UUCP and other non-Telnet protocols. |
With our implementation of TCP/IP, you are not required to enter the commands connect or telnet to establish a Telnet connection. You can just enter the learned host name, as long as the host name is different from a command word for the communication server.
To display a list of the available hosts, enter the following command:
show hostsTo display the status of all TCP connections, enter the following command:
show tcpThe communication server assigns a logical name to each connection; several commands use these names to identify connections. The logical name is the same as the host name, unless that name is already in use or you change the connection name with the EXEC command name-connection. If the name is already in use, the communication server assigns a null name to the connection.
The following example routes packets from the source system to kl.sri.com, then to 10.1.0.11, and finally to mathom:
cs> connect mathom /route:kl.sri.com 10.1.0.11 mathom
The following example connects to a host with logical name mathom:
cs> mathom
The Telnet software supports special Telnet commands in the form of Telnet sequences that map generic terminal control functions to operating system-specific functions.
To issue a special Telnet command, enter the escape sequence (usually Ctrl-^) and then a command character. You can enter the command character as you hold down Ctrl or with Ctrl released; you can type either uppercase or lowercase letters.
Table 3-2 lists the special Telnet commands.
| Task | Escape Sequence |
|---|---|
| Break | Ctrl-^ B |
| Interrupt Process (IP) | Ctrl-^ C |
| Erase Character (EC) | Ctrl-^ H |
| Abort Output (AO) | Ctrl-^ O |
| Are You There? (AYT) | Ctrl-^ T |
| Erase Line (EL) | Ctrl-^ U |
At any time during an active Telnet session, you can list the Telnet commands by entering the following command at the system prompt:
Ctrl-^ ?This is done by typing the escape sequence followed by a question mark. This command displays an online table of the special Telnet commands, for quick reference.
A sample of this list follows (the Ctrl key is represented by the first caret [^]).
CS> ^^?
[Special telnet escape help]
^^B sends telnet BREAK
^^C sends telnet IP
^^H sends telnet EC
^^O sends telnet AO
^^T sends telnet AYT
^^U sends telnet EL
You can have several concurrent rlogin connections open and switch back and forth between them.
To open a new connection, exit out of the current connection by typing the escape sequence, which by default is Ctrl-^ X, to return to the system command prompt, then open a new connection.
To log in to a UNIX host using rlogin, enter the following command:
rlogin host [debug]| host | Specifies the host name or Internet address. |
| debug | (Optional.) Enables debugging output from the rlogin protocol. |
The following example makes an rlogin connection to a host at address 108.33.21.2 and enables the message mode for debugging:
cs> rlogin 108.33.21.2 debug
You can have several concurrent sessions open and switch back and forth between them. The number of sessions that can be opened is defined by the session-limit command, which is described in the publication Communication Server Configuration Guide and Communication Server Command Reference.
You can switch between sessions by escaping out of one session and resuming a previously opened session, as follows:
Step 1 Escape out the current session and return to the EXEC prompt by pressing Ctrl-^ X.
Step 2 List the open sessions using the where command. You display information about all open sessions associated with the current terminal line.
Step 3 Type the resume command and the session number to make the connection.
You can also resume the previous session by pressing the Return key.
The where command has no additional syntax. The resume command has the following syntax when used on the communication server:
resume [connection] [keyword]| connection | (Optional.) The name or number of the connection; the default is the most recent connection. |
| keyword | (Optional.) One of the options listed in Table 3-3. |
| Option | Description |
|---|---|
| /debug | Prints parameter changes and messages. On a protocol translator, this option displays informational messages whenever the remote host changes an X.3 parameter or sends an X.29 control packet. |
| /echo | Performs local echo. |
| /line | Enables line-mode editing. |
| /nodebug | Cancels printing of parameter changes and messages. |
| /noecho | Disables local echo. |
| /noline | Disables line mode and enables character-at-a-time mode, which is the default. |
| /nostream | Disables stream processing. |
| /set parameter:value | Sets X.3 connection options. Refer to "Set X.3 PAD Parameters" in Chapter 5 for a list of these connection options. |
| /stream | Enables stream processing. |
The Ctrl-^-X, where, and resume commands are available with all supported connection protocols.
The following example shows how to resume connection 2:
CS> resume 2
You can omit the command name and simply type the connection number to resume that connection. The following example illustrates how to resume connection 3:
CS> 3
You can issue any of the following commands to terminate an active Telnet or rlogin session:
exit quit logoutYou connect to a host using TN3270 terminal emulation at the EXEC prompt. The system administrator must configure a default terminal emulation file that permits the terminal to correctly communicate with the host. Refer to the publications Communication Server Configuration Guide and Communication Server Command Reference to specify alternate terminal emulations. The administrator can also specify custom terminal emulations.
Unlike Telnet and LAT connections, you must enter the command tn3270 to make a connection to an IBM 3278 host. To begin a TN3270 session, enter the following command:
tn3270 hostname| hostname | Name of a specific host on a network that is reachable by the communication server. The default terminal emulation mode allows access using a VT100 emulation. |
The following example establishes a terminal session with an IBM host named finance:
cs> tn3270 finance
You can issue any of the following commands to terminate an active TN3270 session:
exit quit logoutYou make most connections to the remote host that supports the Serial Line Internet Protocol (SLIP) at the EXEC prompt. The system administrator can configure SLIP to expect a specific address or to provide one for you. It is also possible to set up SLIP in a mode that compresses packets for more efficient use of the line.
To make a SLIP connection, enter the following command:
slip {/default | remote-ip-address | remote-name} [@tacacs-server]} [/routing] [/compressed]| /default | (Optional.) Makes a SLIP connection when a default address has been configured. |
| remote-ip-address | Specifies the IP address of the client workstation or PC. |
| remote-name | Specifies the name of the client workstation or PC. |
| @tacacs-server | (Optional.) Specifies the IP address or IP host name of the TACACS server to which the user's TACACS authentication request is sent. |
| /routing | (Optional.) Indicates that the remote system is a router. Line must be configured for asynchronous routing using SLIP encapsulation. |
| /compressed | (Optional.) Indicates that IP header compression should be negotiated. The system administrator must have configured the system with the ip tcp header-compression passive command for this command to be valid in EXEC mode. The command ip tcp header-compression forces header compression on or off. |
If you specify an address for the TACACS server, either using default, or tacacs-server, the address must be the first parameter in the command after you type slip. If you do not specify an address or enter default, you will be prompted for an IP address or host name. You can enter default at this point to use the default address configured for the line.
If you do not use the tacacs-server argument to specify a TACACS server for SLIP address authentication, the TACACS server specified at login (if any) will be used for the SLIP address query.
To allow optimal use of bandwidth on a line, SLIP allows compressing of the SLIP packets using Van Jacobson TCP header compression as defined in RFC 1144. The default is to not compress the packets. The configuration file must have header compression on and the slip /compressed EXEC command must be explicitly entered for header compression to occur.
You can issue any of the following commands to terminate an active SLIP session:
exit quit logoutThe following example illustrates how to make a connection when a default IP address has been assigned. Once a correct password is entered, you are placed in SLIP mode, and the IP address is displayed.
cs> slip
Password:
Entering SLIP mode.
Your IP address is 192.31.7.28, MTU is 1524 bytes
The following example illustrates the prompts displayed and the response required when dynamic addressing is used to assign the SLIP address:
cs> slip
IP address or hostname? 192.31.6.15
Password:
Entering SLIP mode
Your IP address is 192.31.6.15, MTU is 1524 bytes
In the following example, the address 192.31.6.15 has been assigned as the default. Password verification is still required before SLIP mode can be enabled.
cs> slip default
Password:
Entering SLIP mode
Your IP address is 192.31.6.15, MTU is 1524 bytes
The following example illustrates the implementation of header compression on the interface with the IP address 128.66.2.1:
cs> slip 128.66.2.1 /compressed
Password:
Entering SLIP mode.
Interface IP address is 128.66.2.1, MTU is 1500 bytes.
Header compression will match your system.
In the previous example, the interface is configured for ip tcp header-compression passive, which permitted the user to enter the /compressed keyword at the EXEC mode prompt. The message "Header compression will match your system" indicates that the user indeed specified compression.
If the line was configured for ip tcp header-compression on, this line would read "Header compression is On."
The following example specifies a TACACS server parlance for address authentication:
cs> slip 1.0.0.1@parlance
Password:
Entering SLIP mode.
Interface IP address is 1.0.0.1, MTU is 1500 bytes
Header compression will match your system.
You can make asynchronous connections using the Point-to-Point Protocol (PPP). To start a PPP connection, enter the following command at the EXEC prompt:
ppp {/default | {remote-ip-address | remote-name} [@tacacs-server]} [/routing]| /default | (Optional.) Makes PPP connection when a default address has been configured. |
| remote-ip-address | Specifies an IP address of the client workstation or PC. This parameter can only be specified if the line is set for dynamic addresses using the line configuration command async address dynamic. |
| remote-name | Specifies the name of the client workstation or PC. This parameter can only be specified if the line is set for dynamic addresses using the line configuration command async address dynamic. |
| @tacacs-server | (Optional.) Specifies an IP address or IP host name of the TACACS server to which the user's TACACS authentication request is sent. |
| /routing | (Optional.) Indicates that the remote system is a router and that routing messages should be exchanged over the link. The line must be configured for asynchronous routing using PPP encapsulation. |
If you specify an address for the TACACS server, either using default or tacacs-server, the address must be the first parameter in the command after you type ppp. If you do not specify an address or enter default, you will be prompted for an IP address or host name. You can enter default at this point to use the default address configured for the line.
You can issue any of the following commands to terminate an active PPP session:
exit quit logoutThe following example shows a line that is in asynchronous mode using PPP encapsulation. The name of the PC is ntpc (assuming that the name ntpc is in the DNS so that it can be resolved to a real IP address). The person typing this command is on the PC running a terminal emulator program.
CS> ppp ntpc@server1
The XRemote protocol permits you to connect to remote hosts via TCP/IP and LAT using an X display station and modem. You make connections in any one of the following ways:
The following sections outline steps required for starting up XRemote in several typical environments. When possible, use the automated processes. Before any of these tasks can be performed, the system administrator must have configured a path for loading fonts.
If your host computer supports a server for XDMCP, such as the xdm program included in X11R4 or later, you can use automatic session startup to make an XRemote session connection. To do so, enter the following command:
xremote xdm [hostname]| hostname | (Optional.) Computer host name. |
This command causes an XDMCP session startup request to be made to the computer specified. If you do not specify a host name, a broadcast message is sent to all hosts. The first host to respond by starting up a session is used.
The communication server and X terminal stay in XRemote mode until either the display manager terminates the session, or a reset request is received from the X terminal.
The following example starts a session with a remote host named star.
cs> xremote xdm an
If your host computer supports DECwindows login sessions, you can use automatic session startup to make an XRemote session connection. Once the system administrator at the remote host has configured support for DECwindows over LAT, you can use the EXEC command xremote lat to initiate the connection. The command has the following syntax:
xremote lat service| service | Name of the desired LAT service. |
After you issue this command, expect the following to occur:
When the login box appears, log into the system. Upon completion of login, more fonts are loaded, then the remote session starts up.
The following example begins connection with a LAT service named WHIRL:
cs> xremote lat WHIRL
If you are not using a host computer that supports XDMCP or LAT, you must use manual session startup. Manual session startup involves several steps:
Step 1 Enable XRemote manually on the communication server port.
Step 2 Connect to the host computer.
Step 3 Set the location of the X display.
Step 4 Start up client applications.
Step 5 Return to the EXEC prompt.
Step 6 Enable XRemote manually again on the communication server port.
Step 7 What to do if a session terminates.
You can also establish XRemote sessions between communication servers and exit XRemote sessions. The following sections describe these tasks. See the "XRemote Examples" section at the end of this chapter for XRemote connection examples.
To prepare the communication server for manually starting an XRemote session, enter the xremote EXEC command at the system prompt.
xremoteThis command begins the instructions that prompt you through the connection.
The following example illustrates how a successful manual XRemote session begins:
dialup> xremote
XRemote enabled; your display is dialup:2006
Start your clients and type XRemote again
The system replies with a message informing you of your X display location This information will be used to inform the host of the location of your X display server.
If no clients are found, you see the following message:
No X clients waiting - check that your display is darkstar:2006
Check your hosts to determine if an error was made when starting the session. The most likely cause is an improperly specified display location. Another possible cause is the host computer not recognizing the name of the communication server through which you are attempting to make a connection.
You can connect to a host using one of the following connection commands, and log in as usual:
telnet lat rloginAt this point, you are logged in to the remote host.
Inform the host computer of the location of your X display, which was provided to you by the communication server when you enabled XRemote manually.
For most versions of the UNIX operating system, the X display location is set by using the setenv command to set the environment variable DISPLAY. Refer to your UNIX system's online X(1) manual page for more information.
On VAX/VMS, use the SET DISPLAY command. Refer to the VMS DCL Dictionary for more information.
Now you must start your client applications for your host operating system.
The communication server accepts the X connection attempt from the client application and places the client into a dormant state.
If it is possible to log off the host computer and keep your X clients running in the background, you can do so now. This conserves resources on the host and communication server that would otherwise be inaccessible until you exited from XRemote state.
If you cannot log off the host computer and keep your clients running, escape back to the communication server prompt using the escape sequence (Ctrl-^ X by default).
Begin a manual remote session again; see the step "Enable XRemote Manually." If the X clients connected successfully, the session will be put into XRemote mode, and the clients will be allowed to complete their startup.
If no clients were found, you will see the following message:
No X clients waiting - check that your display is darkstar:2018
Check your hosts to determine if an error was made when starting the session. The most likely cause is an improperly specified display location. Another possible cause is the host computer not recognizing the name of the communication server through which you are attempting to make a connection.
In manual operation, the communication server and X terminal remain in XRemote mode until either all clients disconnect or a reset request is received from the X terminal.
If a session terminates during startup, it might be because you invoked transient X clients which set some parameters and disconnected (such as xset or xmodmap). There must always be one session open or the connection will be reset.
A communication server allows a user of an X display server that does not support XRemote to run the XRemote protocols. An X display server (such as a PCX, MACX or UNIX workstation) connected to an Ethernet network can dial out through a communication server on a conventional modem to access an X client program on a host residing on another network. The communication server provides the server-side helper process.
To run XRemote, connect to one of the XRemote ports.
Find out from your administrator whether the connection from your X display terminal is configured as an individual line or a rotary connection. To connect to an individual line, telnet from the X display terminal to port 9000 plus the decimal value of the line number. To make a rotary connection, telnet from the X display server to port 10000 plus the decimal value of the line number. See the publications Communication Server Configuration Guide and the Communication Server Command Reference for information about how to configure individual lines and rotary connections.
Figure 3-2 illustrates a configuration in which a display server is not running XRemote. In this configuration, the server-side XRemote helper is running on Communication Server 1, and the client-side XRemote helper is running on Communication Server 2.
You can issue any of the following commands to terminate an active terminal session:
exit quit logoutUse the examples in this section to understand how to make XRemote connections.
The following example shows a connection from an X display terminal through a communication server to a host running client programs.
dialup> xremote
XRemote enabled; your display is dialup:2006
Start your clients and type XRemote again
dialup> telnet eureka
Trying EUREKA.NOWHERE.COM (252.122.1.55)... Open
SunOS UNIX (eureka)
login: deal
Password:
Last login: Fri Aug 7 17:17:46 from dialup.nowhere.com
SunOS Release (SERVER+FDDI+DBE.patched) #14: Fri Jul 31 10:37:29 PDT 1992
eureka% setenv DISPLAY dialup:2006
eureka% xterm &
[1] 15439
eureka% logout
[Connection to EUREKA closed by foreign host]
dialup> xremote
Entering XRemote
The communication server and X terminal stay in XRemote mode until either the display manager terminates the session or a reset request is received from the X terminal:
Connection closed by foreign host.
eureka%
This section provides two examples of XRemote connections between communication servers.
The following steps show how an XRemote connection is established for a configuration like the one shown in Figure 3-2. These steps assume that the administrator has set the user's display environment variable to identify the user's X display terminal.
The following example shows the steps to make an XRemote connection between communication servers. The number 9016 in the first line of the display indicates a connection to individual line 16. If the administrator had configured a rotary connection, the user would enter 10000 plus the number of the rotary instead of 9016.
Step 1 Enter the telnet command to make the connection:
telnet golden-road 9016
Step 2 Supply the password for TACACS verification:
Step 3 Dial in to the remote system using the modem, and then log in:
atdt 13125554141
deal
Step 4 Enter the xremote command at the EXEC prompt, and then read and follow the instructions from the host:
xremote
Step 5 Connect to the client:
telnet sparks
deal
Step 6 At the client prompt, enter the display name from step 4 and the xterm command:
setenv DISPLAY dialup:2006
xterm &
Step 7 Disconnect from the client:
logout
Step 8 Begin the XRemote session:
xremote
Once the connection is closed by the foreign host, the Xterm window appears on the local workstation screen:
Connection closed by foreign host.
sparks%
|
|