RFC 93 UNKNOWN ARPANET · CONNECTIONS

ICP: Initial Connection Protocol

A clarification to the Initial Connection Protocol: the 32-bit number is preceded by a data type code of zero.

On rfc-editor.org ↗
PUBLISHED
January 1971
AUTHORS
A.M. McKenzie
STREAM
Legacy
BUILDS ON
RFC 66 — NIC - third level ideas and other noise, RFC 80 — Protocols and Data Formats
MESSAGE FROM SERVER SOCKET #1
An even 32 bit number
MESSAGE DATA TYPE CODE
Eight bits, value zero
DOI
10.17487/RFC0093

RFC text

Original text · RFC Editor View on rfc-editor.org ↗
Network Working Group                                        A. McKenzie
Request for Comments #93                                             BBN
Updates RFC's 66, 80                                        January 1971


                      Initial Connection Protocol


        A review of the Initial Connection Protocol (ICP) first
described in RFC #66 and restated in RFC #80 has revealed an area of
ambiguity, which in turn reflects an ambiguity in the Host-Host
Protocol Document No. 1.  This is the definition of the message sent
over the connection from "Server socket #1".  In both referenced
RFC's, the message is defined as "exactly an even 32 bit number".  It
is not clear, however, whether this 32 bit number is meant to follow
an eight-bit "message data type" code or not, stemming from the fact
that the Host-Host Protocol makes provision for such codes but does
not seem to absolutely demand them.

        Only one implementation of an ICP has been documented in the
NWG literature - that at UCSB (RFC #74).  The implementers of this ICP
have apparently interpreted the Host-Host Protocol as demanding a
message data type code, and therefore do transmit a code of zero.

        Steve Crocker indicates (private communication) that the Host-
Host Protocol was intended to require a message data type code.  We
therefore recommend that RFC numbers 66 and 80 be amended to show that
the "even 32 bit number" is preceded by a message data type code of
zero (zero is the only code currently defined).


       [ This RFC was put into machine readable form for entry ]
        [ into the online RFC archives by James Thompson 4/97 ]

AI ELI5

On this page
PROBLEMRFC 66 and RFC 80 defined the message from Server socket #1 as exactly an even 32 bit number, but it was unclear whether an eight-bit message data type code comes before it.
MECHANISMThe note observes that the only documented implementation, at UCSB, sends a code of zero, and that Steve Crocker indicated the Host-Host Protocol was intended to require the code.
THE FIXIt recommends amending RFC 66 and RFC 80 to show the 32 bit number is preceded by a message data type code of zero, the only code currently defined.

§1The Initial Connection Protocol

The Initial Connection Protocol (ICP) was first described in RFC 66 and restated in RFC 80. This note, by A. McKenzie of BBN, updates both after a review found an area of ambiguity.

The ambiguity also reflects one in the Host-Host Protocol Document No. 1.

There is a standard way to say hello to a service on another computer. This note fixes one unclear sentence in the instructions for it.

§2The Ambiguity

The unclear point is the message sent over the connection from Server socket #1. Both earlier RFCs define it as exactly an even 32 bit number. It was not clear whether this number follows an eight-bit message data type code, because the Host-Host Protocol makes provision for such codes but does not seem to absolutely demand them.

Only one implementation of an ICP had been documented in the NWG literature, the one at UCSB (RFC 74). Its implementers apparently read the Host-Host Protocol as demanding a message data type code, so they transmit a code of zero.

The instructions said to send one number, but did not say whether a small label should come before it. The one team that had built it had added the label.

§3The Recommended Amendment

Steve Crocker indicated in private communication that the Host-Host Protocol was intended to require a message data type code. On that basis the note recommends that RFC 66 and RFC 80 be amended so that the even 32 bit number is preceded by a message data type code of zero.

Zero is the only code currently defined, so the recommendation matches what the UCSB implementation already does.

The answer is yes: put the label first, and the label is a zero. That is also what the one working version already did.

§4References