Network Working Group Jon Postel
Request for Comments: 318 UCLA-NMC
NIC: 9348 April 3, 1972
References: RFC 139, 158, and NIC 7104
Telnet Protocol
At the October 1971 Network Working Group Meeting, I promised to
promptly produce a document which clearly and succinctly specified
and explained the Official Telnet Protocol. This document fails to
meet any part of that promise. This document was not produced
promptly. This document is neither clear nor succinct. There is NO
Official Telnet Protocol.
The following pages present my understanding of the ad hoc Telnet
protocol. There are some who have serious questions about this
protocol. The proposed changes to the protocol are given in Section
IV.
Any comments should be promptly directed to me via the Network
Information Center (Ident = JBP) or by phone (213) 825-2368 or by
mail.
Jon Postel
SPADE Group
3804 Boelter Hall
UCLA
Los Angeles, California 90024
Postel [Page 1]
RFC 318 Telnet Protocol April 1972
I. DEFINITION OF THE NETWORK VIRTUAL TERMINAL
The Network Virtual Terminal (NVT) is a bi-directional character
device. The characters are represented by 8 bit codes. The NVT has
no timing characteristics. The character codes 0 through 127 are the
USASCII codes. (Note all code values are given in decimal.) The
codes 128 through 255 are used for special control signals. The NVT
is described as having a printer and a keyboard. The printer
responds to incoming data and the keyboard produces outgoing data.
The Printer
The NVT printer has an unspecified carriage width (common values are
40, 72, 80, 120, 128, 132). The printer can produce representations
of all 95 USASCII graphics (codes 32 through 126). Of the 33 USASCII
control codes (0 through 31 and 127) the following 8 have specific
meaning to the NVT printer.
NAME CODE MEANING
NULL (NUL) 0 A no operation.
BELL (BEL) 7 Produces an audible or visible signal.
Back Space (BS) 8 Backspaces the printer one character
position.
Horizontal Tab (HT) 9 Moves the printer to next horizontal
tab stop.
Line Feed (LF) 10 Moves the printer to next line (keeping
the same horizontal position).
Vertical Tab (VT) 11 Moves the printer to the next vertical
tab stop.
Form Feed (FF) 12 Moves the printer to the top of the
next page.
Carriage Return (CR) 13 Moves the printer to the left margin
of the current line.
The remaining USASCII codes (1 through 6, 14 through 31, and 127) do
not cause the NVT printer to take any action.
Postel [Page 2]
RFC 318 Telnet Protocol April 1972
The Keyboard
The NVT Keyboard has keys or key combinations or key sequences for
generating all of the 128 USASCII codes. Note that although there
are codes which have no effect on the NVT printer, the NVT Keyboard
is capable of generating these codes.
The End of the Line Convention
The end of a line of text shall be indicated by the character
sequence Carriage Return Line Feed (CR, LF). This convention applies
to both the sending (Keyboard) and receiving (Printer) (virtual)
mechanisms.
Break and Reverse Break
The Telnet control signals provide a BREAK signal which can be used
to simulate the use of the break or attention or interrupt button
found on most terminals. This signal has no effect on the NVT. When
the BREAK Telnet control signal is used from server to user it is
sometimes called "reverse break". Such a reverse break has no effect
on the NVT.
II. DEFINITION OF TELNET PROTOCOL
The purpose of Telnet Protocol is to provide a standard method of
interfacing terminals devices at one site to processes at another
site.
The Telnet Protocol is built up from three major substructures, first
the Initial Connection Protocol (ICP), second the Network Virtual
Terminal (NVT), and third the Telnet control signals described
herein.
Telnet user and server processes follow the ICP to establish
connections. The term "Logger" has been associated with the set of
processes in the serving system which respond to the ICP and perform
the initial interactions e.g. obtain a name and password. The ICP is
defined and the initial socket number and byte size parameters are
defined in "Current Network Protocols" (NIC #7104).
The data transmitted between the user and server programs (and vice
versa) is treated as a character stream with embedded control
signals.
Note that all code values are given in decimal.
Postel [Page 3]
RFC 318 Telnet Protocol April 1972
TELNET CONTROL SIGNALS
NAME CODE MEANING
DATA MARK 128 Used to mark a point in the data stream.
Used in conjunction with INS. See SYNCH.
BREAK 129 User-to-Server: Has the same meaning to
the server as the "Break," "Interrupt," or
"Attention" button found on many terminals.
Server-to-User: Has the same meaning to
to use as the "reverse break" used with
some terminals.
NOP 130 No Operation.
NO ECHO 131 User-to-Server: Asks the server not to
return Echos of the transmitted data.
Server-to-User: States that the server is
not sending echos of the transmitted data.
Sent only as a reply to ECHO or NO ECHO,
or to end the hide your input.
ECHO 132 User-to-Server: Asks the server to send
Echos of the transmitted data.
Server-to User: States that the server is
sending echos of the transmitted data.
Sent only as a reply to ECHO or NO ECHO.
Hide your input 133 The intention is that a server will send
this signal to a user system which is
echoing locally (to the user) when the user
is about to type something secret (e.g. a
password). In this case, the user system
is to suppress local echoing or overprint
the input (or something) until the server
sends a NOECHO signal. In situations where
the user system is not echoing locally,
this signal must not be sent by the server.
INS --- This is the "Interrupt on Send" signal,
defined by the Host-to-Host protocol and
implemented by the Network Control Program
(NCP). See SYNCH.
Postel [Page 4]
RFC 318 Telnet Protocol April 1972
SYNCH --- This is a condition indicated by the
combination of the DATA MARK and the INS.
User-to-Server: The Server is to examine
the input data stream looking for a DATA
MARK signal; if a DATA MARK is found, the
server must not process further until an
INS is received. If the server receives an
INS, it is required to examine the data
stream at once, taking any appropriate
action on "break type" characters (e.g.
etx, sub, BREAK), up to a DATA MARK signal
and thereupon continue its normal processing.
The passed over characters may be discarded.
Server-to-User: If the user finds a DATA
MARK in the data stream, it must wait for
an INS. If the user receives an INS, it
must examine and discard characters up to
and including a DATA MARK.
DATA TYPES
Telnet normally deals in ASCII characters, but there are provisions
for escaping to other code sets. If one of these escapes is used, it
is undefined (here) whether or not the Telnet signals still have
meaning or even how to return to the ASCII set: The Telnet signals
used to indicate a change of code set are:
CODE MEANING
160 ASCII - Standard Telnet
161 Transparent
162 EBCDIC
USER TELNET SIGNALS
The following signals are to be available to the human user to cause
the user Telnet to take the indicated action.
Transmit Now - Transmit all data entered and locally
buffered now. Intended to be used with line
mode.
Suppress end-of-line - Transmit all data entered and locally
buffered now, and do not transmit the
end-of-line immediately following this signal.
Postel [Page 5]
RFC 318 Telnet Protocol April 1972
STANDARD TELNET IMPLEMENTATION
Using Site
1) User is able to enter and transmit all ASCII codes
2) User is able to cause the Telnet signals BREAK, SYNCH, ECHO and
NOECHO to be transmitted.
3) Provides for the User Telnet signals, (e.g. Transmit Now).
4) Implements the CR LF end-of-line convention.
5) Provides local echo for local user terminals.
6) Correctly processes the Telnet signals BREAK, SYNCH, NOP, ECHO,
NOECHO, and Hide Your Input received from the server.
Serving Site:
1) Provides a mapping between ASCII and the local character set.
2) Correctly processes the Telnet signals BREAK, SYNCH, NOP, NOECHO,
and ECHO.
3) Implements the CR LF end-of-line convention.
4) Assumes the using site provides echoing. May provide a server
echo mode.
MINIMUM TELNET IMPLEMENTATION
Using Site:
1) User must be able to enter and transmit all ASCII codes.
2) Ignore and delete all Telnet signals from the serving site.
3) Provide local echo for local user terminals.
4) Implements CR LF end-of-line convention.
5) Provide for the User Telnet signals.
Postel [Page 6]
RFC 318 Telnet Protocol April 1972
Serving Site:
1) Provide a mapping between ASCII and the local character set.
2) Ignore and Delete all Telnet signals from the using site.
3) Assume the using site provides echoing.
4) Implements the CR LF end-of-line convention.
III. DISCUSSION OF TELNET PROTOCOL
The use of a standard, network-wide, intermediate representation of
terminal code between sites is intended to eliminate the need for
using and serving sites to keep information about the characteristics
of each other's terminals and terminal handling conventions. This
approach can be successful, but only if the user, the using site, and
the serving site assume certain responsibilities.
1. The serving site must specify how the intermediate code will be
mapped by it into the terminal codes that are expected at that
site.
2. The user must be familiar with that mapping.
3. The using site must provide some means for the user to enter all
of the intermediate codes, and as a convenience, special Telnet
signals, as well as specify for the user how the signals from the
serving site will be presented at the user terminal.
Other schemes were considered but rejected. For example, a proposal
that the using site be responsible for translating to and from the
code expected by the serving site was rejected since it required that
the using site keep tables of all serving site codes and provide a
mapping for each case. The information would require constant
maintenance as new hosts were added to the network.
Character Set
Since it is not known how the current or future sites will specify
the mapping between the network-wide standard code (7 bit ASCII in an
8 bit field) and the codes expected from their own terminals, it
seems necessary to permit the user to cause transmission of every one
of the 128 ASCII codes, plus (for full user power) selected signals
(either of a Telnet control nature, or of a special terminal nature
such as break or attention).
Postel [Page 7]
RFC 318 Telnet Protocol April 1972
There was strong feeling about the importance of the user/system
interface at the using site, but equally strong feeling that this
problem is one of local implementation and should reflect the using
site installation philosophy rather than be subject to network-wide
standards. Some topics of consideration in this area are:
1. How to represent special graphics, not available at the using
site, at the user's terminal.
2. Treatment of upper/lower case problem on upper case only
devices.
a. Representing lower-case output.
b. Providing users with shift and shiftlock signals.
3. Incorporating editing capability in Telnet.
4. Extending user options in Network mode not available to local
users, e.g., hold output or kill print.
5. Permit users to specify how keyboard input is to the
translated, e.g., let a character from the terminal cause a
specified string to be sent by the user's Telnet.
The proposed solution to the Telnet Protocol problem seems to provide
a mechanism for a minimum implementation while providing a basis for
developing richer sets or protocol for present and future use in
terminal applications, process-process communications, and use by
other conventions to pass data or control information.
The understanding that ASCII be used as a network-wide code has been
established for some time. Its use in Telnet provided a problem with
respect to the limitation of a maximum character set of 128. Some
systems provide for more than this number in their operation, and
therefore, as serving sites, cannot map on a one-for-one basis.
Each such serving site could probably provide a reasonably useful
character set, including all system control signals, by mapping 128
of its codes and just not provide a network user access to the other
codes. However, any character left out might later be used in a
major application at that site as a special control signal. This
could result in denying network users the facility offered by that
application. Serving sites are, therefore, encouraged to provide a
full mapping between the ASCII code and the code used on the serving
system. This may require that the server specify two character ASCII
sequences which map to single characters in the servers character
set.
Postel [Page 8]
RFC 318 Telnet Protocol April 1972
Notice that there are some ASCII codes which have no effect on the
NVT printer. These codes must be transmitted over the network when
output by the serving process or by entered by the user.
End of Line Convention
The representation of the end of a physical line at a terminal is
implemented differently on different network hosts. For example,
some use a return (or new line) key, the terminal hardware both
returns the carriage or printer to start of line and feeds the paper
to the next line. In other implementations, the user hits carriage
return and the hardware returns carriage while the software sends the
terminal a line feed. The network-wide representation is carriage
return followed by line feed. It represents the physical formatting
that is being attempted, and is to be interpreted and appropriately
translated by both using site and serving site.
EXAMPLE: A Multics user is working, through the network, on some
serving site host. In the course of the session the user has
numerous occasions to hit New Line on his Model 37 TTY. Each time
the Multics system is awakened by a New Line interrupt, the line
of buffered characters is passed to Telnet where it is scanned for
special characters. If none is found, carriage return followed by
line feed is inserted where New Line was entered, and the line is
turned over the NCP for transmission. Correspondingly, when the
Multics Telnet finds the carriage return line feed sequence in the
data stream coming from the serving site, the two characters are
replaced with the appropriate New Line code which is sent to the
terminal.
Telnet defines the end of a line to be indicated by the ASCII
character pair CR LR. Several of the real devices in the world have
only a single new line (NL) function. Several of the computer
systems have in some programs used the CR and LF functions to have
semantic meaning larger than the format effect they provide.
Further, several computer systems allow the CR and LF functions to be
used separately (e.g., such that a line may be overprinted). One
problem, for those Telnet (user) programs required to map the NVT
into a device which only has a NL function, is how is the CR LF to be
dealt with. One solution is to examine the character following the
CR. If an LF is found, then perform the NL function; if anything
else is found then back space to the beginning of the line. Another
problem is the case of a computer system which locally uses period,
".", to cause the new line function and which uses, in some programs,
CR and LF for semantically significant operations. Suppose the user
Telnet sends the sequence CR LF. Does this mean "new line" or the
"CR operation" followed by the "LF operation "? A solution to this
problem it to require that Telnet programs send a CR NOT intended to
Postel [Page 9]
RFC 318 Telnet Protocol April 1972
be part of a CR LF pair as a CR NUL pair. Then the receiving program
can always hold a CR and examine the next character to determine if a
new line function is intended. This solution is strongly
recommended.
One other question arises here, "Is it permitted to send the Telnet
signal NOP (code 130) between a CR and a LF when these are intended
to signify new line?" The answer is "yes, the NOP signal may occur
anywhere in the data stream."
Echoing
The decision to have the assumed condition for echo be that the using
site will provide any echo necessary for its terminals was taken
because of the difficulties faced by some installations that cannot
turn off their echo or that have terminals that print locally as a
result of key strokes. Serving sites could take the position "have
user turn echo off," but this seems an unnecessary burden on the
user. In addition, some serving sites may choose not to supply any
echo service, in which case the no echo assumption will supply a
network-wide condition, while other assumptions would give a mixed
starting condition.
The convention of using "ECHO," "NO ECHO" signals seems to fill both
the requirements for dynamic echo control and for a minimum
implementation of Telnet Protocol. Note that when the user request
ECHO or NO ECHO the server replies by switching to the desired mode
(and possibly returning the signal for the new mode), or by
continuing in the current mode and returning the signal for the
current mode. The server never spontaneously sends an ECHO or NO
ECHO signal. Except that a NOECHO may be used to cancel a HIDE YOUR
INPUT.
Hide Your Input
The HIDE YOUR INPUT signal presents some difficulty in that it is
unclear how much is to be hidden. The server site usually knows how
long the secret is but the user Telnet in general does not.
Furthermore, if the user site cannot suppress the local echoing,
there is a difficult implementation problem. One possibility is for
the using site to overprint a full line with a mask, then have the
user type his secret on the mask. If the secret were longer than one
line, the use of the mask should be repeated.
The use of HIDE YOUR INPUT can be avoided altogether by having the
serving site send a mask (which it knows to be just long enough) on
which the user is to type the secret information.
Postel [Page 10]
RFC 318 Telnet Protocol April 1972
EXAMPLE:
1. Default assumption is user site is echoing
2. Server-to User: Password Please CR LF
3. Server-to-User: XXXXCRIIIIICRMMMMCR NUL
4. User-to-Server: "password" CR LF
5. Server-to-User: Ready CR LF
Breaks and Attentions
There is a special control signal on some terminals that has no
corresponding bit pattern in ASCII, but is transmitted by a special
electrical signal. This control signal is Attn on a 2741 and Break
on a Teletype. This signal is represented by the Telnet control
signal BREAK. There is a corresponding control signal for use from
serving sites to using sites for reverse break. Notice, however,
that the NVT is a bi-directional character device, thus there is no
need to "turn the line around".
Some systems treat the Break as an extra code available for use in
conjunction with the data stream. For example, one system uses Break
as a special editing code meaning "delete the current line to this
point." In these cases, the code may simply be inserted in the data
stream with no special additional action by the user.
Other systems use Break or Attn in special interrupt fashion, to mean
stop processing the application and give me the supervisor, or cancel
the present job, etc. (Other systems which inspect input on a
character at a time basis use normal characters for this purpose,
such as <etx>.) In these cases, because of differences in the ways
both serving and using sites operate, it is necessary to take a route
in addition to the normal Telnet data stream to indicate that the
special control signal is embedded in the data stream.
Example -- Problem:
The PDP-10 normally will, when it fills its input buffer, continue
to accept characters from a terminal examining each to see if it
is a control character, then act on it if it is or throw it away
if it is not.
Since the Telnet server at the serving site is at the mercy of the
NCP with respect to controlling the bunching, and therefore,
arrival at the Telnet of bursts of characters, Telnet
Postel [Page 11]
RFC 318 Telnet Protocol April 1972
implementations might be expected to choke off flow to the buffers
until they are ready to accept characters without throwing them
away.
Under this condition, the serving process might be outputting to
the using terminal, the input buffers at the server fill up, (with
user generated characters) and <etx> get stuck (at the user's
site) in the data stream that has been choked off.
A similar problem could occur with Multics or some IBM system as a
line at a time server. The user at a using site gets his process at
the serving site into an output loop and wants to break the process
without having to release his Telnet connection. The buffers clog
the connection, transmission is choked off, and the <etx>, Break, or
other user control signal gets stuck in the pipeline.
Example -- Solution:
The user at the using site knows he is entering a special control
signal (Break, Attn, <etx>, etc.) and follows it with a SYNCH.
(The local instructions at using sites for accomplishing this may
differ from site to site.)
User to Using Site Telnet
Send SYNCH.
Using Site Telnet to Serving Site Telnet:
DATA MARK in Data Stream.
Using Site Telnet to Using Site NCP:
Send an INS.
Serving Site NCP to Telnet Server:
Interrupt "INS received".
Serving Site Telnet:
Examines the input data stream (looking for special control
signals) until it sees DATA MARK then resumes normal
handling.
Thus, depending on the server's local implementation to
provide adequate service, a special handling of the data
stream can be invoked whenever an INS is received in order
Postel [Page 12]
RFC 318 Telnet Protocol April 1972
to get to the special character. When it sees DATA MARK, it
recognizes it as a synchronization point and knowing that
the special character has been passed on, strips the DATA
MARK from the data stream and returns to normal mode.
If the DATA MARK arrives before the INS, the serving site
should not process the data stream further until an INS is
received.
This approach to handling selected special characters or signals
relieves the using Telnet processes from having to recognize the
special serving site characters, as well as from having to know how
the serving site wants to handle them. At the same time, the
procedure requires only a minimum level of user understanding of the
serving site. This seems appropriate, since the Telnet ASCII
conventions are providing a Network Virtual Terminal, not a Network
Virtual User.
Notice that the correct order is (1) special character or signal
(e.g. BREAK or <etx>), then (2) SYNCH.
User Telnet Signals
The ability of the user to cause the using site Telnet to send any
combination of ASCII characters in a string, and only that
combination, is viewed as important to the user utility of the Telnet
ASCII conventions. Because of this, some user sites may find it
necessary to provide special local Telnet signals from the human user
to the using site Telnet.
Example:
A user on a line at a time system (Multics, System 360, GCOS,
etx.), which require an end of line signal before processing the
user's input, is working through the Network on a serving site
that operates a character at a time. The application is a
debugging aid that permits the user to type in "location=" to
which it will respond with n where n represents the current
contents of that location. The serving site process does not
expect to see the "location=" followed by a carriage return line
feed sequence. The user at the using site should be able to type
in the location, follow it with a signal (to the user Telnet) to
suppress the end of line convention, followed by the end of line
signal, and expect the "location=" to be transmitted immediately
without an end of line sequence being transmitted to the server.
Postel [Page 13]
RFC 318 Telnet Protocol April 1972
Example:
In another case, a using site has decided that it is convenient to
accumulate four characters at a time and transmit them to the
serving site, unless an end of line signal is observed, in which
case the end of line sequence is sent preceded by whatever number
of characters have been accumulated (presumably three or less).
In the same debugging application, the address is such that the
end does not correspond with the four character buffer
demarcation. The user should have the ability to enter a code for
"transmit immediately" in place of the end of line signal in order
to preserve neat formatting, and expect the address to be sent to
the serving site.
Telnet Signals have been discussed and those introduced to date are
probably sufficient for an implementation of Telnet ASCII convention.
Terminology
ASCII - The USASCII character set as defined in NIC # 7104.
In Telnet Protocol, where eight bit codes are used
the lower half of the code set is defined to be
ASCII.
echoing - The display of a character entered is called echoing.
There are two modes in which this happens. If
the device used to enter characters displays the
character before (or as) it transmits the character
to the computer the echoing mode is called "local
echo." If, on the other hand, the device transmits
the entered character to the computer without
displaying it and the computer then transmits a
character to the device for the echo display, this
echoing mode is called "remote echo."
character mode - In this mode of operation Telnet transmits each
character as soon as possible. Generally speaking,
character mode is used when all of the using terminal,
using system, and serving system are operating
in a remote echo mode. The echos to the user
entered characters are transmitted from the serving
system (i.e., over the network).
line mode - In this mode of operation Telnet transmits groups
of characters which constitute lines. Generally
speaking, this mode is used when one or more of
the using terminal, using system, or serving
Postel [Page 14]
RFC 318 Telnet Protocol April 1972
system is operating in a local echo mode. The
echos to the user entered characters are not
transmitted over the network).
full duplex - This term indicates a transmission procedure using
a four wire connection, which permits simultaneous
transmission in both directions.
half duplex - This term indicates a transmission procedure using
a two wire connection, which requires that data be
transmitted in only one direction at at time.
Note that half duplex devices usually are also local echo but that
full duplex devices may be either local echo or remote echo.
IV. PROPOSED CHANGES TO TELNET PROTOCOL
The changes suggested here are not my ideas, thus the presentation may
be faulty. I welcome RFC or other communication suggesting other
changes or better arguments for and against these changes.
Echoing
It is proposed to delete from Telnet the control signals ECHO, NOECHO,
and HIDE YOUR INPUT. For Server systems which do not provide echoing,
these commands are useless. For server systems which do provide
echoing experience has shown that the control is most effectively
provided by server system commands.
Data Types
It is proposed to delete all mention of data types from Telnet.
Either the character stream is ASCII or its not a Telnet
communication. If it is really necessary to change the data type, a
command in ASCII could be sent in the data stream.
Minimum Implementation
It is proposed that the minimum implementation require the user Telnet
to allow the user to send and the server Telnet to correctly process
all the Telnet control signals.
Postel [Page 15]
RFC 318 Telnet Protocol April 1972
The work on Telnet Protocol has involved many people. This document
is taken from RFC's #139 and #158 by Tom O'Sullivan. Others who
have served on committees are:
Bob Bressler MIT-DMCG
Will Crowther BBN
Bob Long SDC
Alex McKenzie BBN
John Melvin SRI-ARC
Bob Metcalfe MIT-DMCG
Ed Meyer MIT-Multics
Tol O'Sullivan Raytheon
Mike Padlipsky Mit-Multics
Jon Postel UCLA-NMC
Bob Sundberg Harvard
Joel Winett LL
Steve Wolfe UCLA-CCN
[This RFC was put into machine readable form for entry]
[into the online RFC archives by Kelly Tardif, Viaginie 10/99]
Postel [Page 16]