Network Working Group                                          J. Winett
Request for Comments: 109                         MIT Lincoln Laboratory
NIC: 5805                                                  24 March 1971


         Level III Server Protocol for the Lincoln Laboratory
                             360/67 Host

Disclaimer

  This material has not been reviewed for public release and is
  intended only for use with the ARPA network.  It should not be quoted
  or cited in any publication not related to the ARPA network.

Introduction

  The Lincoln Laboratory IBM 360/67 is connected to the ARPA network
  and acts as a serving host providing access to the CP-67 virtual
  machine operating system.  Upon completion of the Login procedure,
  users have control of a 360 virtual machine through a virtual 1052
  online console.  Attached to the virtual machine is a virtual card
  reader, card punch and line printer, and a number of disk storage
  devices.  The 360 virtual machine can be either a virtual 360/67 with
  dynamic address translation hardware or a standard System/360.  Most
  users run a standard 360 with 256K bytes of virtual memory and
  operate the CMS conversational monitor system.  CMS provides
  facilities for file creation, maintenance and manipulation, program
  development, debugging and execution, and a number of other useful
  utility functions.  The section in the Network Notebook on the
  Lincoln Laboratory 360/67 more fully describes the facilities
  available.

Network Control Program

  All communication with the 360/67 through the IMP are processed by a
  Network Control Program (NCP).  The NCP operates with the Host-Host
  Protocol described in the Network Working Group Document No. 1 dated
  3 August 1970.

Initial Connection Protocol

  To create a virtual machine from the network, a pair of connections
  must be made with the LOGGER.  The sockets to be used are assigned
  following the Initial Connection Protocol (ICP).  The LOGGER is
  enabled and waiting for an RTS control command for socket X'0A 0000
  01'.  This ICP socket corresponds to home X'0A', user X'0000', and
  tag X'01' (send gender).  Requests for connection on the ICP socket
  are stacked until it becomes free.  If the LOGGER is willing to



Winett                                                          [Page 1]

RFC 109                 Level II Server Protocol           24 March 1971


  service another network user, a 32 bit socket ID of a receive socket
  will be sent over this initial connection and the ICP socket will
  then be closed.  If the LOGGER is not willing to service another
  network user, it will not complete the initial connection for the ICP
  socket and will refuse the request by closing the connection without
  completing it.

LOGGER Protocol

  Once a pair of user sockets have been assigned, the connection
  protocol should be completed on these sockets.  The LOGGER then
  expects to receive (on the receive socket) one 8-bit byte indicating
  the data type which characterizes the transmission code used to
  communicate with the network user over this pair of sockets.  A code
  of X'01' implies 7 bit ASCII code in 8-bit bytes with the leading bit
  zero.  A code of X'02' implies 8-bit EBCDIC code.  When the data type
  code is received, the LOGGER will echo back the data type code over
  the send socket followed by the message:

           LINCOLN LABORATORY CP-67 ONLINE NL

  in the appropriate code.  (In ASCII, NL is transmitted as CR LF).

  The procedure continues according to the normal CP-67 login protocol
  with the LOGGER performing an additional function of mapping network
  userids and passwords into valid CP-67 userids and passwords.  This
  mapping is specified by entries in a file (the LOGGER FILE) which the
  LOGGER accesses.  If a network userid does not match an entry in the
  file or if the password given does not match the corresponding
  network password, the usual CP responses will be sent to the users.
  Thus network access to the Lincoln Laboratory system is restricted to
  those accounts for which an appropriate entry has been made in the
  LOGGER FILE.

  It should be noted that CP transmits a BYP code (Bypass) to suspend
  the printing of characters keyed while a password is being entered.
  After the password has been entered, CP transmits a RES code
  (Restore) to resume the printing when characters are keyed.  When
  communicating in ASCII, these character codes are converted to X'FF'
  since no corresponding ASCII code is defined.  Refer to the Network
  Resource Notebook for more details on CP-67 and on CMS.

The NET Account

  Lincoln Laboratory is providing one account which can be used by
  network users to familiarize themselves with our time-sharing system.
  The userid of this account is NET and the password is ARPA.  This
  account has 900 records of storage, which can store approximately



Winett                                                          [Page 2]

RFC 109                 Level II Server Protocol           24 March 1971


  720,000 characters.  NET users are free to ERASE any file on this
  account since many different people may use this account.

The SERVER Protocol

  CP-67 operates on a line at a time, i.e., a group of characters are
  processed as a line and not as a sequence of individual characters.
  Also, the system does normally buffer input lines, that is, input is
  not normally entered until requested by a read from the system.  With
  IBM 2741 or 1052 terminals, the keyboard is locked until a read is
  requested.  The virtual terminals through which network users have
  access to the CP-67 system have been designed to support either a
  line oriented terminal or a character oriented terminal.  When CP
  requests a line of input, the SERVER transmits a prompting code
  X'80'.  This character can be used to signal a user process to change
  transmission modes and to transmit an input line.  Characters
  received by the SERVER are buffered until a NL character is received.
  Lines received can then be used to satisfy CP requests for an input
  line.

  CP may send out lines which may or may not end with a NL character.
  If a line does not end with a NL character, the prompting character
  will naturally be sent following the output line to request input to
  a CP process.

  When a user wishes to interrupt a CP process, i.e., to change modes,
  an interrupt code X'80' should be sent to the SERVER.  This code will
  result in an asynchronous interrupt being sent to the running
  process, stimulating the pressing of the 'attention' button on a 2741
  terminal.  Together with the transmission of the interrupt code, the
  user should cause an INS to be sent over the send link.  This signal
  will be synchronized with the interrupt code.  If the interrupt code
  has not yet been received and processed, all characters buffered and
  those received before the receipt of the interrupt code will be
  flushed, i.e., deleted.  When the interrupt code is received, it will
  be paired with the previously received INS.  If an INS is received
  after an interrupt code has been received and processed, the INS will
  be paired with this previously received interrupt code.

  If CP has a line to send to a user after it has requested an input
  line but before it has received any input, the SERVER will transmit
  an INS on the user's receive link to notify the user that previously
  sent prompting character should be retracted and that a line has been
  or will be sent to the user.  This message line is called a
  "warning".






Winett                                                          [Page 3]

RFC 109                 Level II Server Protocol           24 March 1971


Graphic and Control Codes

  Figure 1 gives the 8-bit codes for the EBCDIC graphics and controls.
  Figure 2 gives the 7-bit codes for the ASCII graphics and controls.
  The controls are tabulated and compared in Figure 3.  The standard
  interpretation of the ASCII controls are given in Figure 4.

  There are 4 ASCII codes which do not have a corresponding graphic or
  control in the EBCDIC code.  The EBCDIC codes given to these codes
  are as follows:

                  |     Hex Code
            ASCII |-------+--------
           Symbol | ASCII | EBCDIC
           -------+-------+--------
              DC3 |  13   |  3A
                  |       |
               `  |  60   |  70
                  |       |
               \  |  5C   |  71
                  |       |
               ^  |  5E   |  72

  There are 29 EBCDIC graphics codes and 19 EBCDIC control codes which
  do not have a corresponding graphic or control in the ASCII code.  In
  addition, there are 84 other EBCDIC codes whose interpretation is
  unspecified.  Four of these codes have been chosen to correspond to
  the ASCII control and ASCII graphics which do not have a
  corresponding EBCDIC code.  When converting EBCDIC codes to ASCII
  codes, the remaining 80 codes plus the 29 EBCDIC graphics and 18
  EBCDIC controls (not counting NL) are converted into the code X'FF'.
  The NL character is treated specially.  The NL character, EBCDIC code
  X'15', is converted into the two character sequence CR LF, i.e.,
  ASCII X'0D 0A'.  As stated above, the code X'80' is transmitted as a
  prompting character whenever CP requests an input line.

  On converting from ASCII to EBCDIC, if any code other than the 128
  ASCII codes, or the interrupt codem X'80', is received, it is
  converted to the code X'FF'.  In addition , whenever the two ASCII
  characters CR LF are found sequentially in the input stream, they are
  converted into the single EBCDIC character NL.

  [In Figure 1, positions shown as "[?]" cannot be printed in ASCII.]








Winett                                                          [Page 4]

RFC 109                 Level II Server Protocol           24 March 1971


       0 0  0   0   0   0   0   0   0   1   1   1   1   1   1   1   1
       1 0  0   0   0   1   1   1   1   0   0   0   0   1   1   1   1
       2 0  0   1   1   0   0   1   1   0   0   1   1   0   0   1   1
       3 0  1   0   1   0   1   0   1   0   1   0   1   0   1   0   1
  4567+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0000|NUL|DLE|DS |   |SP | & | - |   |   |   |[?]|[?]|   |   |   | 0 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0001|SOH|DC1|SOS|   |   |   | / |   | a | j |[?]|[?]| A | J |   | 1 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0010|STX|DC2|FS |SYN|   |   |   |   | b | k | s |[?]| B | K | S | 2 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0011|ETX|TM |   |   |   |   |   |   | c | l | t |[?]| C | L | T | 3 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0100|PF |RES|BYP|PN |   |   |   |   | d | m | u |[?]| D | M | U | 4 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0101|HT |NL |LF |RS |   |   |   |   | e | n | v |[?]| E | N | V | 5 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0110|LC |BS |ETB|UC |   |   |   |   | f | o | w |[?]| F | O | W | 6 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  0111|DEL|IL |ESC|EOT|   |   |   |   | g | p | x |[?]| G | P | X | 7 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1000|   |CAN|   |   |   |   |   |   | h | q | y |[?]| H | Q | Y | 8 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1001|   |EM |   |   |   |   |   |   | i | r | z |[?]| I | R | Z | 9 |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1010|SMM|CC |SM |   |[1]| ! |   | : |   |   |   |   |   |   |   |   |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1011|VT |CU1|CU2|CU3| . | $ | , | # | { | } |[?]|[?]|   |   |   |   |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1100|FF |IFS|   |DC4| < | * | % | @ |[?]|[?]|[?]|[?]|   |   |   |   |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1101|CR |IGS|ENQ|NAK| ( | ) | _ | ' |[?]|[?]| [ | ] |   |   |   |   |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1110|SO |IRS|ACK|   | + | ; | > | = |[?]|[?]|[?]|[?]|   |   |   |   |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
  1111|SI |IUS|BEL|SUB| | |[2]| ? | " |[?]|[?]|[?]|[?]|   |   |   |   |
      +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+

                        +---+---+---+---+---+---+---+---+
                        | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
                        +---+---+---+---+---+---+---+---+
                                  Code Structure

                                     Figure 1.

           Extended Binary-Coded Decimal Interchange Code (EBCDIC)





Winett                                                          [Page 5]

RFC 109                 Level II Server Protocol           24 March 1971


           8 0   0   0   0   0   0   0   0
           7 0   0   0   0   1   1   1   1
           6 0   0   1   1   0   0   1   1
           5 0   1   0   1   0   1   0   1
       4321+---+---+---+---+---+---+---+---+
       0000|NUL|DLE|SP | 0 | @ | P | ` | p |
           +---+---+---+---+---+---+---+---+
       0001|SOH|DC1| ! | 1 | A | Q | a | q |
           +---+---+---+---+---+---+---+---+
       0010|STX|DC2| " | 2 | B | R | b | r |
           +---+---+---+---+---+---+---+---+
       0011|ETX|DC3| # | 3 | C | S | c | s |
           +---+---+---+---+---+---+---+---+
       0100|EOT|DC4| $ | 4 | D | T | d | t |
           +---+---+---+---+---+---+---+---+
       0101|ENQ|NAK| % | 5 | E | U | e | u |
           +---+---+---+---+---+---+---+---+
       0110|ACK|SYN| & | 6 | F | V | f | v |
           +---+---+---+---+---+---+---+---+
       0111|BEL|ETB| ' | 7 | G | W | g | w |
           +---+---+---+---+---+---+---+---+
       1000|BS |CAN| ( | 8 | H | X | h | x |
           +---+---+---+---+---+---+---+---+
       1001|HT |EM | ) | 9 | I | Y | i | y |
           +---+---+---+---+---+---+---+---+
       1010|LF |SUB| * | : | J | Z | j | z |
           +---+---+---+---+---+---+---+---+
       1011|VT |ESC| + | ; | K | [ | k | { |
           +---+---+---+---+---+---+---+---+
       1100|FF |FS | , | < | L | \ | l | | |
           +---+---+---+---+---+---+---+---+
       1101|CR |GS | - | = | M | ] | m | } |
           +---+---+---+---+---+---+---+---+
       1110|SO |RS | . | > | N | ^ | n | ~ |
           +---+---+---+---+---+---+---+---+
       1111|SI |SU | / | ? | O | _ | o |DEL|
           +---+---+---+---+---+---+---+---+
   +---+---+---+---+---+---+---+---+---+---+---+---+
           | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 |
           +---+---+---+---+---+---+---+---+
                    Code Structure

                       Figure 2.

  USA Standard Code for Information Interchange (USASCII)






Winett                                                          [Page 6]

RFC 109                 Level II Server Protocol           24 March 1971


  CAT     EBCDIC  ASCII   TTY     PTTC    FUNCTION

          NUL     NUL     NULL            Null
  CC      SOH     SOH     SOM             Start of Heading
  CC      STX     STX     EOA     EOA (D) Start of Text
  CC      ETX     ETX     EOM             End of Text
  DC      PF                      PF      Punch Off
  FE      HT      HT      H.TAB   TAB     Horizontal Tab
  GR      LC                      LC      Lower Case
          DEL     DEL     RUBOUT  DEL     Delete
          SMM                             Start of Manual Message
  FE      VT      VT      V.TAB           Vertical Tab
  FE      FF      FF      FORM            Form Feed
  FE      CR      CR      RETURN          Carriage Return
  GR      SO      SO      SO              Shift Out
  GR      SI      SI      SI              Shift In
  CC      DLE     DLE     DC0             Data Link Escape
  DC      DC1     DC1     X-ON            Device Control 1
  DC      DC2     DC2     TAPE ON         Device Control 2
          TM                              Tape Mark
  DC      RES                     RES     Restore
  FE      NL                      NL      New Line
  FE      BS      BS              BS      Backspace
          IL                      IL      Idle
          CAN     CAN     FE0     CAN     Cancel
          EM      EM      S1              End of Medium
          CC                              Cursor Control
  CU      CU1                             Customer Use 1
  IS      IFS     FS      S4              Info. Field Separator
  IS      IGS     GS      S5              Info. Group Separator
  IS      IRS     RS      S6              Info. Record Separator
  IS      IUS     US      S7              Info Unit Separator
  ED      DS                              Digit Select
  ED      SOS                             Start of Significance
  ED      FS                              Field Separator
  DC      BYP                     BYP     Bypass
  FE      LF      LF      LF      LF      Line Feed
  CC      ETB     ETB     LEM     EOB (B) End of Text Block
          ESC     ESC     S3      PRE     Escape
          SM                              Set Mode
  CU      CU2                             Customer Use 2
  CC      ENQ     ENQ     WRU             Enquiry
  CC      ACK     ACK     RU      (Y)     Acknowledge
          BEL     BEL     BELL            Bell
  CC      SYN     SYN     SYNC            Synchronous Idle
  DC      PN                      PN      Punch On
  DC      RS                      RS      Reader Stop
  GR      UC                      UC      Upper Case



Winett                                                          [Page 7]

RFC 109                 Level II Server Protocol           24 March 1971


  CC      EOT     EOT     EOT     EOT (C) End of Transmission
  CU      CU3                             Customer Use 3
  DC      DC4     DC4     TAPE OFF        Device Control 4
  CC      NAK     NAK     ERROR   (N)     Negative Acknowledge
          SUB     SUB     S2              Substitute
  DC              DC3     X-OFF           Device Control 3

                      Figure 3

               Control Functions Compared

  CC   (Communication Control).  A functional character intended to
        control or facilitate transmission of information over
        communication networks.

  FE   (Format Effector).  A functional character which controls the
        layout or positioning of information in printing or display
        devices.

  IS   (Information Separator).  A character which is used to separate
        and qualify information in a logical sense.  There is a group
        of four such characters, which are to be used in a hierarchical
        order.

  DC   (Device Control).  A functional character used for the control
        of ancillary devices associated with data processing of
        telecommunication systems, more especially switching devices
        "on" and "off".

  ED   (Edit and Mark).  A control character used by the System/360
        Edit and Mark (EDMK) instruction for the formatting of
        alphanumeric fields.

  GB   (Graphic Control).  A control character indicating that the code
        combinations which follow are to be interpreted in a particular
        code table, depending upon the particular control character.

  CU   (Customer Use).  A character excluded from future assignment by
        IBM.  These "protected" codes are intended for use by customer
        systems so that their use will not conflict with a possible
        future IBM use.

                          Figure 3 (Continued)

                    Categories of Control Functions






Winett                                                          [Page 8]

RFC 109                 Level II Server Protocol           24 March 1971


  NUL  (Null).  The all-zeros character which may serve to accomplish
        time fill and media fill.

  SOH  (Start of Heading).  A communication control character used at
        the beginning of a sequence of characters which constitute a
        machine-sensible address or routing information.  Such a
        sequence is referred to as the _heading_.  An STX character has
        the effect of terminating a heading.

  STX  (Start of Text).  A communication control character which
        precedes a sequence of characters that is to be treated as an
        entity and transmitted through to the ultimate destination.
        Such a sequence is referred to as _text_.  SIX may be used to
        terminate a sequence of characters started by SOH.

  ETX  (End of Text).  A communication control character used to
        terminate a sequence of characters started with STX and
        transmitted as an entity.

  EOT  (End of Transmission).  A communication control character used
        to indicate the conclusion of a transmission, which may have
        contained one or more texts and any associated headings.

  ENQ  (Enquiry).  A communication control character used in data
        communication systems as a request for a response from a remote
        station.  It may be used as a "Who Are You" (WRU) to obtain
        identification, or may be used to obtain station status, or
        both.

  ACK  (Acknowledge).  A communication control character transmitted by
        a receiver as an affirmative response to a sender.

  BEL  (Bell).  A character for use when these is a need to call for
        human attention.  It may control alarm or attention devices.

  BS   (Backspace).  A format effector which controls the movement of
        the printing position one printing space backward on the same
        printing line (applicable also to display devices).

  HT   (Horizontal Tabulation).  A format effector which controls the
        movement of the printing position to the next in a series of
        predetermined positions along the printing line (applicable
        also to display devices and the skip function on punched
        cards.)

  LF   (Line Feed).  A format effector which controls the movement of
        the printing position to the next printing line (also
        applicable to display devices).



Winett                                                          [Page 9]

RFC 109                 Level II Server Protocol           24 March 1971


  VT   (Vertical Tabulation).  A format effector which controls the
        movement of the printing position to the next in a series of
        predetermined printing lines (also applicable to display
        devices).

  FF   (Form Feed).  A format effector which controls the movement of
        the printing position to the first predetermined printing line
        on the next form or page (also applicable to display devices).

  CR   (Carriage Return).  A format effector which controls the
        movement of the printing position to the first printing
        position on the same printing line (also applicable to display
        devices).

  SO    (Shift Out).  A control character indicating that the code
        combinations which follow shall be interpreted as outside of
        the character set of the standard code table until a Shift In
        Character is reached.

  SI    (Shift In).  A control character indicating that the code
        combinations which follow shall be interpreted according to the
        standard code table.

  DLE   (Data Link Escape).  A communication control character which
        will change the meaning of a limited number of contiguously
        following characters.  It is used exclusively to provide
        supplementary controls in data communication networks.

  DC1, DC2, DC3, DC4 (Device Controls).  Characters for the control of
        ancillary devices associated with data processing or
        telecommunication systems, more especially switching devices
        "on" and "off".  (If a single "stop" control is required to
        interrupt of turn off ancillary devices, DC4 is the preferred
        assignment.)

  NAK  (Negative Acknowledge).  A communication control character
        transmitted by a receiver as a negative response to a sender.

  SYN   (Synchronous Idle).  A communication control character used by
        a synchronous transmission system in the absence of any other
        character to provide a signal from which synchronism may be
        achieved or retained.









Winett                                                         [Page 10]

RFC 109                 Level II Server Protocol           24 March 1971


  ETB  (End of Transmission Block).  A communication control character
        used to indicate the end of a block of data for communication
        purposes.  ETB is used for blocking data where the block
        structure is not necessarily related to the processing format.

  CAN  (Cancel).  A control character used to indicate that the data
        with which it is sent is in error or is to be disregarded.

  EM   (End of Medium).  A control character associated with the sent
        data which may be used to identify the physical end of the
        medium, or the end of the used, or wanted, portion of
        information recorded on a medium.  (The position of this
        character does not necessarily correspond to the physical end
        of the medium.

  SS   (Start of Special Sequence).  A control character used to
        indicate the start of a variable length sequence of characters
        which have special significance or which are to have special
        handling.

  ESC  (Escape).  A control character intended to provide code
        extension (supplementary characters) in general information
        interchange.  The Escape character itself is a prefix affecting
        the interpretation of a limited number of contiguously
        following characters.

  FS   (File Separator), GS (Group Separator), RS (Record Separator)
        and US (Unit Separator).  These information separators may be
        used within data in optional fashion, except that the
        hierarchical relationship shall be : FS is the must inclusive,
        then GS, then RS, and US is least inclusive.  (The content and
        length of a File, Group, Record, or Unit are not specified.)

  DEL  (Delete).  This character is used primarily to "erase" or
        "obliterate" erroneous or unwanted characters in perforated
        tape.  (In the strict sense, DEL is not a control character.)

                              Figure 4

                       ASCII Control Functions











Winett                                                         [Page 11]

RFC 109                 Level II Server Protocol           24 March 1971


Endnotes

  [1] - Cent sign
  [2] - Logical not ("bent bar")
  [?] - Graphics not in ASCII.  See Postscript or PDF version of
        this document.


         [This RFC was put into machine readable form for entry]
         [into the online RFC archives by Lorrie Shiota, 10/02]









































Winett                                                         [Page 12]