Network Working Group                                         D. Crocker
Request For Comments: 1845                        Brandenburg Consulting
Category: Experimental                                          N. Freed
                                           Innosoft International, Inc.
                                                  A. Cargille, WG Chair
                                                         September 1995


                        SMTP Service Extension
                        for Checkpoint/Restart

Status of this Memo

  This memo defines an Experimental Protocol for the Internet
  community.  This memo does not specify an Internet standard of any
  kind.  Discussion and suggestions for improvement are requested.
  Distribution of this memo is unlimited.

Abstract

  This memo defines an extension to the SMTP service whereby an
  interrupted SMTP transaction can be restarted at a later time without
  having to repeat all of the commands and message content sent prior
  to the interruption.

1.  Introduction

  Although SMTP is widely and robustly deployed, various extensions
  have been requested by parts of the Internet community. In
  particular, when dealing with very large messages over less reliable
  connections it is possible for substantial resources to be consumed
  by repeated unsuccessful attempts to transmit the message in its
  entirety. The original SMTP specification [1] does not provide any
  means to pick up a partially completed transaction after the
  underlying TCP connection has been broken and reestablished.

  This memo provides a facility by which a client can uniquely identify
  a particular SMTP transaction. The server then stores this
  identifying information along with all the information it receives as
  the transaction proceeds. If the transaction is interrupted during
  the data transfer phase the SMTP client may establish a new SMTP
  session at a later time and ask the server to continue the
  transaction from the point where the server lost its connection with
  the client. The server then reestablishes the transaction context and
  tells the client where to resume operations. If this is acceptable
  the client resumes operations at this point.





Crocker, Freed & Cargille     Experimental                      [Page 1]

RFC 1845                SMTP Checkpoint/Restart           September 1995


  This extension may also be used to work around the common timeout
  problem where a client times out waiting for a response from the
  server acknowledging that the message has been accepted. However, use
  of this extension is not an acceptable substitute for proper setting
  of timeout parameters.

2.  Framework for the Checkpointing Extension

  The checkpointing extension is laid out as follows:

(1)   the name of the SMTP service extension defined here is
      checkpointing;

(2)   the EHLO keyword value associated with the extension is
      CHECKPOINT;

(3)   no parameter is used with the CHECKPOINT EHLO keyword;

(4)   one optional parameter using the keyword TRANSID is
      added to the MAIL FROM command.  The value associated
      with this parameter, coupled with the name of the
      client taken from EHLO command, forms a globally unique
      value that identifies this particular transaction and
      serves to distinguish it from all others. This value is
      case-sensitive. The syntax of the value is as follows,
      using the ABNF notation of [2]:

           transid-value  ::= "<" transid-spec ">"
                              ; transid-value may not be longer than
                              ; 80 characters
           transid-spec   ::= transid-local "@" transid-domain
           transid-domain ::= transid-token
           transid-local  ::= transid-token
           transid-token  ::= transid-atom *("." transid-atom)
           transid-atom   ::= 1*<any (ASCII) CHAR except SPACE,
                                 CTLs, tspecials, or ".">

      NOTE: tspecials is defined in [3]. The TRANSID is
      likely to be different from the RFC822 message id,
      since it must uniquely identify the particular copy of
      the message being sent over this SMTP link. However,
      the syntax of transid-value is designed so that any
      TRANSID is both a legal RFC822 msg-id as well as being
      a legal esmtp-value [4].

(5)   The maximum length of a MAIL FROM command line is
      increased by 88 characters by the possible addition of
      the TRANSID keyword and value;



Crocker, Freed & Cargille     Experimental                      [Page 2]

RFC 1845                SMTP Checkpoint/Restart           September 1995


(6)   no additional SMTP verbs are defined by this extension;
      and,

(7)   the next section specifies how support for the
      extension affects the behavior of a server and client
      SMTP.

3.  The checkpointing service extension

  When a client SMTP wishes to use checkpointing to eliminate the need
  to retransmit all message data in its entirety in the event of a
  session interruption, it first issues the EHLO command to the server
  SMTP. If the server SMTP responds with code 250 to the EHLO command,
  and the response includes the EHLO keyword value CHECKPOINT, then the
  server SMTP is indicating that it supports SMTP checkpointing and
  will honor requests to restart interrupted SMTP transactions.

  The extended MAIL command is issued by a client SMTP when it wishes
  to enable server checkpointing. The syntax for this command is
  identical to the MAIL command in [1], except that a TRANSID parameter
  must appear after the address.

  The complete syntax of this extended command is defined in [4], with
  the esmtp-keyword TRANSID and transid-value parameter as previously
  defined.

  The value associated with the TRANSID parameter must be an identifier
  that serves to uniquely identify this particular SMTP transaction.
  Only one TRANSID parameter may be used in a single MAIL command. Care
  must be used in constructing TRANSID values to simultaneously insure
  both uniqueness and the ability to reidentify interrupted
  transactions.

  The TRANSID is structured to ensure globally uniqueness without any
  additional registry. The transid-domain part should be a valid domain
  name that uniquely identifies the SMTP client. Note that this is
  usually the same as the domain name given in conjunction with the
  EHLO command, but not always. The EHLO domain name identifies the
  specific host the SMTP connection originated from, whereas the
  transid-domain may refer to a group of hosts that collectively host a
  multi-homed SMTP client. The transid-local part should be an
  identifier that distinguishes this SMTP transaction from any other
  originating from this SMTP client.

  Despite the structured nature of the TRANSID the server should treat
  the value as an opaque, case-sensitive string.





Crocker, Freed & Cargille     Experimental                      [Page 3]

RFC 1845                SMTP Checkpoint/Restart           September 1995


  Note that the contents of the RFC822 message-id header typically are
  NOT appropriate for use as the TRANSID parameter value, since such
  identifiers may be associated with multiple copies of the same
  message -- e.g., as it is split during transmission down different
  network paths -- and hence with multiple distinct SMTP transactions.

  A server which supports the checkpointing extension will then retain
  the transaction identifer as well as the most recent state of the
  transaction in non-volatile storage. This information should deleted
  only when the transaction is known to be complete from the client's
  perspective. Completion is assured only when the client either
  explicitly aborts the transaction, starts a new transaction, or
  requests that the connection be closed with a QUIT command.

  In the event of an interruption prior to completing a transaction
  this preserved state will remain for some period of time defined by
  the operational policies of the server administrator. It is
  recommended that transaction state information be preserved for at
  least 48 hours, although no specific time is required.

  When a client detects that a transaction has been interrupted, it
  then must wait for some period before reconnecting. This period must
  be long enough for server connections to time out and for the
  transaction state associated with such connections to be released for
  use by a new connection. The Internet Host Requirements [5] also
  impose restriction on how quickly reconnection attempts can be made
  (section 5.3.1.1).

  Once the necessary period has elapsed the client first checks the DNS
  as described in [6] and determine the set of acceptable IP addresses
  the message can be transferred to. If the IP address used to connect
  to the original server is still on this list it should be tried
  first, since this server is most likely to be capable of restarting
  the transaction. If this connection attempt fails the client must
  then proceed as described in [6] to try all the remaining IP
  addresses and restart the transaction there. If the attempt to
  restart fails on one of the other servers the client is required to
  retransmit the transaction in its entirety at that point.  Waiting
  for a server with an interrupted transaction state to come back
  online is not acceptable.

  Note: Multi-homed SMTP servers do exist, which means that it is
  entirely possible for a transaction to restart on a different server
  host.

  Once the connection is made the client issues the same MAIL command
  with exactly the same transaction identifier. If the transaction was
  interrupted during or at the end of the transfer of actual message



Crocker, Freed & Cargille     Experimental                      [Page 4]

RFC 1845                SMTP Checkpoint/Restart           September 1995


  data, the server first reestablishes its context to a point close as
  possible to the point of interruption and then responds with the
  status message:

    355 octet-offset is the transaction offset

  The actual status text can vary. However the octet-offset field is
  required to be the first thing on the first line of the reply, it
  must be separated from any following text by linear whitespace, and
  it is structured as follows:

    octet-offset ::= 1*DIGIT

  The octet-offset represents an offset, counting from zero, to the
  particular octet in the actual message data the server expects to see
  next. (This is also a count of how many octets the server has
  received and stored successfully.) This offset does NOT account for
  envelope data, i.e., MAIL FROM and RCPT TO commands. A value of 0
  would indicate that the client needs to start sending the message
  from the beginning, a value of 1 would indicate that the client
  should skip one octet, and so on.

  The SMTP canonical format for messages is used when this offset is
  computed.  Any octets added by any SMTP data-stuffing algorithm do
  not count as part of this offset. In the case of data transferred
  with the DATA command the offset must also correspond to the
  beginning of a line.

  Once this context is reestablished the client issues another data
  transfer command (e.g., DATA) and sends the remaining message data.
  Once this data is terminated the transaction completes in the normal
  fashion and the server deletes the transaction context from non-
  volatile storage.

  Note that the semantics of the octet-offset immediately suggest a
  particularly simple implementation strategy, where the client
  retransmits the message data as it normally would but suppresses
  output of the first octet-offset octets of material. The semantics
  used here are intentionally designed to make such implementation
  possible, but care must be taken to insure that such an
  implementation strategy does not impose a significant performance
  penalty on the client.









Crocker, Freed & Cargille     Experimental                      [Page 5]

RFC 1845                SMTP Checkpoint/Restart           September 1995


5.  Usage Example

  The following dialogue illustrates the use of the checkpointing
  service extension:

S: <wait for connection on TCP port 25>
C: <open connection to server>
S: 220 dbc.mtview.ca.us SMTP service ready
C: EHLO ymir.claremont.edu
S: 250-dbc.mtview.ca.us says hello
S: 250 CHECKPOINT
C: MAIL FROM:<[email protected]> TRANSID=<[email protected]>
S: 250 <[email protected]>... Sender and TRANSID ok
C: RCPT TO:<[email protected]>
S: 250 <[email protected]>... Recipient ok
C: DATA
S: 354 Send checkpointed message, ending in CRLF.CRLF
<some amount of message data transmitted>
<session is interrupted and TCP connection is broken>

Some time later a new connection is established:
S: <wait for connection on TCP port 25>
C: <open connection to server>
S: 220 dbc.mtview.ca.us SMTP service ready
C: EHLO ymir.claremont.edu
S: 250-dbc.mtview.ca.us says hello
S: 250 CHECKPOINT
C: MAIL FROM:<[email protected]> TRANSID=<[email protected]>
S: 355 6135 is the transaction offset
C: DATA
S: 354 Send previously checkpointed message starting at octet 6135
C: <message data minus first 6135 octets sent>
C: .
S: 250 OK
C: QUIT
S: 221 Goodbye

6.  Security Considerations

  This RFC does not discuss security issues and is not believed to
  raise any security issues not already endemic in electronic mail and
  present in fully conforming implementations of [1].









Crocker, Freed & Cargille     Experimental                      [Page 6]

RFC 1845                SMTP Checkpoint/Restart           September 1995


7.  References

  [1] Postel, J., "Simple Mail Transfer Protocol", STD 10, RFC 821,
      USC/Information Sciences Institute, August 1982.

  [2] Crocker, D., "Standard for the Format of ARPA Internet Text
      Messages", STD 11, RFC 822, UDEL, August 1982.

  [3] Borenstein, N., and N. Freed, "Multipurpose Internet Mail
      Extensions", RFC 1521, Bellcore, Innosoft, September 1993.

  [4] Rose, M., Stefferud, E., Crocker, D., Klensin, J., and N. Freed,
      "SMTP Service Extensions", RFC 1651, Dover Beach Consulting,
      Inc., Network Management Associates, Inc., Silicon Graphics,
      Inc., MCI, Innosoft, July 1994.

  [5] Braden, R., Editor, "Requirements for Internet Hosts -
      Application and Support", STD 3, RFC 1123, USC/Information
      Sciences Institute, October 1989.

  [6] Partridge, C., "Mail Routing and the Domain System", STD 14, RFC
      974, BBN, January 1986.

8.  Authors' Addresses

      Dave Crocker
      Brandenburg Consulting
      675 Spruce Dr.
      Sunnyvale, CA 94086 USA
      USA

      Phone: +1 408 246 8253
      Fax: +1 408 249 6205
      EMail: [email protected]


      Ned Freed
      Innosoft International, Inc.
      1050 East Garvey Avenue South
      West Covina, CA 91790
      USA

      Phone: +1 818 919 3600
      Fax: +1 818 919 3614
      EMail: [email protected]






Crocker, Freed & Cargille     Experimental                      [Page 7]

You are viewing proxied material from gopher.fnord.one. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.