4.4BSD/usr/src/contrib/mh-6.8/support/pop/pop.txt
Request For Comments: draft
Post Office Protocol (revised)
Wed Jan 14 18:47:59 1987
Marshall T. Rose
Computer Science Laboratory
Northrop Research and Technical Center
Palos Verdes Peninsula
MRose@NRTC.NORTHROP.COM
This memo suggests a simple method for workstations to dynamically
access mail from a mailbox server. This RFC specifies a proposed
protocol for the ARPA Internet community, and requests discussion
and suggestions for improvements.
Acknowledgements
This memo is based on RFC918. Although similar in form to the
original POP proposed for the ARPA Internet community, the protocol
discussed in this memo is similar in spirit to the ideas
investigated by the MZnet project at the University of California,
Irvine.
Further, substantial work was done on examining POP in a PC-based
environment. This work, which resulted in additional functionality
in this protocol, was performed by the ACIS Networking Systems Group
at Stanford University. The author gratefully acknowledges their
interest.
Request For Comments: draft M. Rose
Post Office Protocol (revised) UDel
Introduction
On certain types of smaller nodes in the ARPA Internet it is often
impractical to maintain a message transport system(MTS). For
example, a workstation may not have sufficient resources (cycles,
disk space) in order to permit a SMTP server and associated local
mail delivery system to be kept resident and continuously running.
Similarly, it may be expensive (or impossible) to keep a personal
computer interconnected to an IP-style network for long amounts of
time (the node is lacking the resource known as "connectivity").
Despite this, it is often very useful to be able to manage mail on
these smaller nodes, and they often support a user agent(UA) to aid
the tasks of mail handling. To solve this problem, a node which
can support an MTS entity offers a maildrop service to these less
endowned nodes. The Post Office Protocol (POP) is intended to
permit a workstation to dynamically access a maildrop on a server
host in a useful fashion. Usually, this means that the POP is used
to allow a workstation to retrieve mail that the server is holding
for it.
For the remainder of this memo, the term "client host" refers to a
host making use of the POP service, while the term "server host"
refers to a host which offers the POP service.
A Short Digression
This memo does not specify how a client host enters mail into the
transport system, although a method consistent with the philosophy
of this memo is presented here:
When the user agent on a client host wishes to enter a message
into the transport system, it establishes an SMTP connection
to its relay host (this relay host could be, but need not be,
the POP server host for the client host).
If this method is followed, then the client host appears to the MTS
as a user agent, and should NOT be regarded as a "trusted" MTS
entity in any sense whatsoever. This concept, along with the role
of the POP as a part of a split-UA model is discussed later in this
memo.
The Protocol
Initially the server host starts the POP service by listening on
TCP port 109. When a client host wishes to make use of the service
it establishes a TCP connection with the server host. When the
connection is established, the POP server sends a greeting. The
client and POP server then exchange commands and responses
(respectively) until the connection is closed or aborted.
Commands in the POP consist of a keyword possibly followed by an
argument. All commands are terminated by a CRLF pair.
Responses in the POP consist of a success indicator and a keyword
possibly followed by additional information. All responses are
terminated by a CRLF pair. There are currently two success
indicators: positive ("+OK") and negative ("-ERR").
Responses to certain commands are multi-line. In these cases,
which are clearly indicated below, after sending the first line of
the response and a CRLF, any additional lines are sent, each
terminated by a CRLF pair. When all lines of the response have
been sent, a final line is sent, consisting of a termination octet
(octal code 056, ".") and a CRLF pair. If any line of the
multi-line response begins with the termination octet, the line is
"bit-stuffed" by pre-pending the termination octet to that line of
the response. Hence a multi-line response is terminated with the
five octets "CRLF.CRLF". When examining a multi-line response, the
client checks to see if the line begins with the termination
octet. If so and if octets other than CRLF follow, the the first
octet of the line (the termination octet) is stripped away. If so
and if CRLF immediately follows the termination character, then the
response from the POP server is ended and the line containing
".CRLF" is not considered part of the multi-line response.
A POP session progresses through a number of states during its
lifetime. Once the TCP connection has been opened and the POP
server has sent the greeting, the session enters the AUTHORIZATION
state. In this state, the client must identify itself to the POP
server. Once the client has successfully done this, the server
acquires resources associated with the client's maildrop, and the
session enters the TRANSACTION state. In this state, the client
requests actions on the part of the POP server. When the client
has finished its transactions, the session enters the UPDATE state.
In this state, the POP server releases any resources acquired
during the TRANSACTION state and says goodbye. The TCP connection
is then closed.
The AUTHORIZATION State
Once the TCP connection has been opened by a POP client, the POP
server issues a one line greeting. This can be any string
terminated by CRLF. An example might be:
S: +OK dewey POP server ready (Comments to: PostMaster@UDel)
Note that this greeting is a POP reply. The POP server should always
give a positive response as the greeting.
The POP session is now in the AUTHORIZATION state. The client must
now issue the USER command. If the POP server responds with a
positive success indicator ("+OK"), then the client may issue
either the PASS command to complete the authorization, or the QUIT
command to terminate the POP session. If the POP server responds
with a negative success indicator ("-ERR") to the USER command,
then the client may either issue a new USER command or may issue
the QUIT command.
When the client issues the PASS command, the POP server uses the
argument pair from the USER and PASS commands to determine if the
client should be given access to the appropriate maildrop. If so,
the POP server then acquires an exclusive-access lock on the
maildrop. If the lock is successfully acquired, the POP server
parses the maildrop into individual messages (read note below),
determines the last message (if any) present in the maildrop that
was referenced by the RETR command, and responds with a positive
success indicator. The POP session now enters the TRANSACTION
state. If the lock can not be acquired or the client should is
denied access to the appropriate maildrop or the maildrop can't be
parsed for some reason, the POP server responds with a negative
success indicator. (If a lock was acquired but the POP server
intends to respond with a negative success indicator, the POP server
must release the lock prior to rejecting the command.) At this
point, the client may either issue a new USER command and start
again, or the client may issue the QUIT command.
NOTE: Minimal implementations of the POP need only be
able to break a maildrop into its component messages;
they need NOT be able to parse individual messages. More
advanced implementations may wish to have this
capability, for reasons discussed later.
After the POP server has parsed the maildrop into individual
messages, it assigns a message-id to each message, and notes the
size of the message in octets. The first message in the maildrop
is assigned a message-id of "1", the second is assigned "2", and so
on, so that the n'th message in a maildrop is assigned a message-id
of "n". In POP commands and responses, all message-id's and
message sizes are expressed in base-10.
It sets the "highest number accessed" to be that of the last
message referenced by the RETR command.
Here are summaries for the three POP command discussed thus far:
USER name
Arguments: a server specific user-id (required)
Restrictions: may only be given in the AUTHORIZATION state
after the POP greeting or after an unsuccessful USER
or PASS command
Possible Responses:
+OK name is welcome here
-ERR never heard of name
Examples:
C: USER mrose
S: +OK mrose is a real hoopy frood
...
C: USER frated
S: -ERR sorry, frated doesn't get his mail here
PASS string
Arguments: a server/user-id specific password (required)
Restrictions: may only be given in the AUTHORIZATION state
after a successful USER command
Possible Responses:
+OK maildrop locked and ready
-ERR invalid password
-ERR unable to lock maildrop
Examples:
C: USER mrose
S: +OK mrose is a real hoopy frood
C: PASS secret
S: +OK mrose's maildrop has 2 messages (320 octets)
...
C: USER mrose
S: +OK mrose is a real hoopy frood
C: PASS secret
S: -ERR unable to lock mrose's maildrop, file already locked
QUIT
Arguments: none
Restrictions: none
Possible Responses:
+OK
Examples:
C: QUIT
S: +OK dewey POP server signing off
The TRANSACTION State
Once the client has successfully identified itself to the POP
server and the POP server has locked and burst the appropriate
maildrop, the POP session is now in the TRANSACTION state. The
client may now issue any of the following POP commands repeatedly.
After each command, the POP server issues a response. Eventually,
the client issues the QUIT command and the POP session enters the
UPDATE state.
Here are the POP commands valid in the TRANSACTION state:
STAT
Arguments: none
Restrictions: may only be given in the TRANSACTION state.
Discussion:
The POP server issues a positive response with a line
containing information for the maildrop. This line is
called a "drop listing" for that maildrop.
In order to simplify parsing, all POP servers are
required to use a certain format for drop listings. The
first octets present must indicate the number of messages
in the maildrop. Following this is the size of the
maildrop in octets. This memo makes no requirement on
what follows the maildrop size. Minimal implementations
should just end that line of the response with a CRLF
pair. More advanced implementations may include other
information.
NOTE: This memo STRONGLY discourages implementations
from supplying additional information in the drop
listing. Other, optional, facilities are discussed
later on which permit the client to parse the
messages in the maildrop.
Note that messages marked as deleted are not counted in
either total.
Possible Responses:
+OK nn mm
Examples:
C: STAT
S: +OK 2 320
LIST [msg]
Arguments: a message-id (optionally) If a message-id is
given, it may NOT refer to a message marked as deleted.
Restrictions: may only be given in the TRANSACTION state.
Discussion:
If an argument was given and the POP server issues a
positive response with a line containing information for
that message. This line is called a "scan listing"
for that message.
If no argument was given and the POP server issues a
positive response, then the response given is multi-line.
After the initial +OK, for each message in the maildrop,
the POP server responds with a line containing information
for that message. This line is called a "scan listing"
for that message.
In order to simplify parsing, all POP servers are required
to use a certain format for scan listings. The first
octets present must be the message-id of the message.
Following the message-id is the size of the message in
octets. This memo makes no requirement on what follows
the message size in the scan listing. Minimal
implementations should just end that line of the response
with a CRLF pair. More advanced implementations may
include other information, as parsed from the message.
NOTE: This memo STRONGLY discourages implementations
from supplying additional information in the scan
listing. Other, optional, facilities are discussed
later on which permit the client to parse the
messages in the maildrop.
Note that messages marked as deleted are not listed.
Possible Responses:
+OK scan listing follows
-ERR no such message
Examples:
C: LIST
S: +OK 2 messages (320 octets)
S: 1 120
S: 2 200
S: .
...
C: LIST 2
S: +OK 2 200
...
C: LIST 3
S: -ERR no such message, only 2 messages in maildrop
RETR msg
Arguments: a message-id (required) This message-id may NOT
refer to a message marked as deleted.
Restrictions: may only be given in the TRANSACTION state.
Discussion:
If the POP server issues a positive response, then the
response given is multi-line. After the initial +OK, the
POP server sends the message corresponding to the given
message-id, being careful to bit-stuff the termination
character (as with all multi-line responses).
If the number associated with this message is higher than
the "highest number accessed" in the maildrop, the POP
server updates the "highest number accessed" to the number
associated with this message.
Possible Responses:
+OK message follows
-ERR no such message
Examples:
C: RETR 1
S: +OK 120 octets
S: <the POP server sends the entire message here>
S: .
DELE msg
Arguments: a message-id (required) This message-id may NOT
refer to a message marked as deleted.
Restrictions: may only be given in the TRANSACTION state.
Discussion:
The POP server marks the message as deleted. Any future
reference to the message-id associated with the message
in a POP command generates an error. The POP server does
not actually delete the message until the POP session
enters the UPDATE state.
If the number associated with this message is higher than
the "highest number accessed" in the maildrop, the POP
server updates the "highest number accessed" to the number
associated with this message.
Possible Responses:
+OK message deleted
-ERR no such message
Examples:
C: DELE 1
S: +OK message 1 deleted
...
C: DELE 2
S: -ERR message 2 already deleted
NOOP
Arguments: none
Restrictions: may only be given in the TRANSACTION state.
Discussion:
The POP server does nothing, it merely replies with a
positive response.
Possible Responses:
+OK
Examples:
C: NOOP
S: +OK
LAST
Arguments: none
Restrictions: may only be issued in the TRANSACTION state.
Discussion:
The POP server issues a positive response with a line
containing the highest message number which accessed.
Zero is returned in case no message in the maildrop has
been accessed during previous transactions. A client
may thereafter infer that messages, if any, numbered greater
than the response to the LAST command are messages not yet
accessed by the client.
Possible Response:
+OK nn
Examples:
C: STAT
S: +OK 4 320
C: LAST
S: +OK 1
C: RETR 3
S: +OK 120 octets
S: <the POP server sends the entire message here>
S: .
C: LAST
S: +OK 3
C: DELE 2
S: +OK message 2 deleted
C: LAST
S: +OK 3
C: RSET
S: +OK
C: LAST
S: +OK 1
RSET
Arguments: none
Restrictions: may only be given in the TRANSACTION state.
Discussion:
If any messages have been marked as deleted by the POP
server, they are unmarked. The POP server then replies
with a positive response. In addition, the "highest
number accessed" is also reset to the value determined
at the beginning of the POP session.
Possible Responses:
+OK
Examples:
C: RSET
S: +OK maildrop has 2 messages (320 octets)
The UPDATE State
When the client issues the QUIT command from the TRANSACTION state
the POP session enters the UPDATE state. (Note that if the client
issues the QUIT command from the AUTHORIZATION state, the POP
session terminates but does NOT enter the UPDATE state).
QUIT
Arguments: none
Restrictions: none
Discussion:
The POP server removes all messages marked as deleted
from the maildrop. It then releases the exclusive-access
lock on the maildrop and replies as to the success of
these operations. The TCP connection is then closed.
Possible Responses:
+OK
Examples:
C: QUIT
S: +OK dewey POP server signing off (maildrop empty)
...
C: QUIT
S: +OK dewey POP server signing off (2 messages left)
...
Optional POP Commands
The POP commands discussed above must be supported by all minimal
implementations of POP servers.
The optional POP commands described below permit a POP client
greater freedom in message handling, while preserving a simple POP
server implementation.
NOTE: This memo STRONGLY encourages implementations to
support these commands in lieu of developing augmented
drop and scan listings. In short, the philosophy of this
memo is to put intelligence in the part of the POP client
and not the POP server.
TOP msg n
Arguments: a message-id (required) and a number. This
message-id may NOT refer to a message marked as deleted.
Restrictions: may only be given in the TRANSACTION state.
Discussion:
If the POP server issues a positive response, then the
response given is multi-line. After the initial +OK, the
POP server sends the headers of the message, the blank
line separating the headers from the body, and then the
number of lines indicated message's body, being careful to
bit-stuff the termination character (as with all
multi-line responses).
Note that if the number of lines requested by the POP
client is greater than than the number of lines in the
body, then the POP server sends the entire message.
Possible Responses:
+OK top of message follows
-ERR no such message
Examples:
C: TOP 10
S: +OK
S: <the POP server sends the headers of the message,
a blank line, and the first 10 lines of the
body of the message>
S: .
...
C: TOP 100
S: -ERR no such message
RPOP user
Arguments: a client specific user-id (required)
Restrictions: may only be given in the AUTHORIZATION state
after a successful USER command; in addition, may
only be given if the client used a reserved (privileged)
TCP port to connect to the server.
Discussion:
The RPOP command may be used instead of the PASS command
to authenticate access to the maildrop. In order for this
command to be successful, the POP client must use a
reserved TCP port (port < 1024) to connect to the server.
The POP server uses the argument pair from the USER and
RPOP commands to determine if the client should be given
access to the appropriate maildrop. Unlike the PASS
command however, the POP server considers if the remote
user specified by the RPOP command who resides on the POP
client host is allowed to access the maildrop for the user
specified by the USER command (e.g., on Berkeley UNIX, the
.rhosts mechanism is used). With the exception of this
differing in authentication, this command is identical to
the PASS command.
Possible Responses:
+OK maildrop locked and ready
-ERR permission denied
Examples:
C: USER mrose
S: +OK mrose is a real hoopy frood
C: RPOP mrose
S: +OK mrose's maildrop has 2 messages (320 octets)
POP Command/Reply Summary
Minimal POP Commands:
USER name valid in the AUTHORIZATION state
PASS string
QUIT
STAT valid in the TRANSACTION state
LIST [msg]
RETR msg
DELE msg
NOOP
LAST
RSET
QUIT valid in the UPDATE state
Optional POP Commands:
RPOP user valid in the AUTHORIZATION state
TOP msg n valid in the TRANSACTION state
POP Replies:
+OK
-ERR
Note that with the exception of the STAT command, the reply given
by the POP server to any command is significant only to "+OK" and
"-ERR". Any text occurring after this reply may be ignored by the
client.
Example POP Session
S: <wait for connection on TCP port 109>
...
C: <open connection>
S: +OK dewey POP server ready (Comments to: PostMaster@UDel)
C: USER mrose
S: +OK mrose is a real hoopy frood
C: PASS secret
S: +OK mrose's maildrop has 2 messages (320 octets)
C: STAT
S: +OK 2 320
C: LIST
S: +OK 2 messages (320 octets)
S: 1 120
S: 2 200
S: .
C: RETR 1
S: +OK 120 octets
S: <the POP server sends message 1>
S: .
C: DELE 1
S: +OK message 1 deleted
C: RETR 2
S: +OK 200 octets
S: <the POP server sends message 2>
S: .
C: DELE 2
S: +OK message 2 deleted
C: QUIT
S: +OK dewey POP server signing off (maildrop empty)
C: <close connection>
S: <wait for next connection>
Message Format
All messages transmitted during a POP session are assumed to
conform to the standard for the format of ARPA Internet text
messages [RFC822].
It is important to note that the byte count for a message on the
server host may differ from the octet count assigned to that
message due to local conventions for designating end-of-line.
Usually, during the AUTHORIZATION state of the POP session, the POP
client can calculate the size of each message in octets when it
parses the maildrop into messages. For example, if the POP server
host internally represents end-of-line as a single character, then
the POP server simply counts each occurrence of this character in a
message as two octets. Note that lines in the message which start
with the termination octet need not be counted twice, since the POP
client will remove all bit-stuffed termination characters when it
receives a multi-line response.
The POP and the Split-UA model
The underlying paradigm in which the POP functions is that of a
split-UA model. The POP client host, being a remote PC based
workstation, acts solely as a client to the message transport
system. It does not provide delivery/authentication services to
others. Hence, it is acting as a UA, on behalf of the person using
the workstation. Furthermore, the workstation uses SMTP to enter
mail into the MTS.
In this sense we have two UA functions which interface to the
message transport system: Posting (SMTP) and Retrieval (POP). The
entity which supports this type of environment is called a split-UA
(since the user agent is split between two hosts which must
interoperate to provide these functions).
ASIDE: Others might term this a remote-UA instead. There
are arguments supporting the use of both terms.
This memo has explicitly referenced TCP as the underlying transport
agent for the POP. This need not be the case. In the MZnet
split-UA, for example, personal micro-computer systems are used
which do not have IP-style networking capability. To connect to
the POP server host, a PC establishes a terminal connection using
some simple protocol (PhoneNet). A program on the PC drives the
connection, first establishing a login session as a normal user.
The login shell for this pseudo-user is a program which drives the
other half of the terminal protocol and communicates with one of
two servers. Although MZnet can support several PCs, a single
pseudo-user login is present on the server host. The user-id and
password for this pseudo-user login is known to all members of
MZnet. Hence, the first action of the login shell, after starting
the terminal protocol, is to demand a USER/PASS authorization pair
from the PC. This second level of authorization is used to
ascertain who is interacting with the MTS. Although the server host
is deemed to support a "trusted" MTS entity, PCs in MZnet are not.
Naturally, the USER/PASS authorization pair for a PC is known only
to the owner of the PC (in theory, at least).
After successfully verifying the identity of the client, a modified
SMTP server is started, and the PC posts mail with the server host.
After the QUIT command is given to the SMTP server and it
terminates, a modified POP server is started, and the PC retrieves
mail from the server host. After the QUIT command is given to the
POP server and it terminates, the login shell for the pseudo-user
terminates the terminal protocol and logs the job out. The PC then
closes the terminal connection to the server host.
The SMTP server used by MZnet is modified in the sense that it
knows that it's talking to a user agent and not a "trusted" entity
in the message transport system. Hence, it does performs the
validation activities normally performed by an entity in the MTS
when it accepts a message from a UA.
The POP server used by MZnet is modified in the sense that it does
not require a USER/PASS combination before entering the TRANSACTION
state. The reason for this (of course) is that the PC has already
identified itself during the second-level authorization step
described above.
NOTE: Truth in advertising laws require that the author
of this memo state that MZnet has not actually been fully
implemented. The concepts presented and proven by the
project led to the notion of the MZnet split-slot model.
This notion has inspired the split-UA concept described
in this memo, led to the author's interest in the POP,
and heavily influenced the the description of the POP
herein.
In fact, some UAs present in the ARPA Internet already support the
notion of posting directly to an SMTP server and retreiving mail
directly from a POP server, even if the POP server and client
resided on the same host!
ASIDE: this discussion raises an issue which this memo
purposedly avoids: how does SMTP know that it's talking
to a "trusted" MTS entity?
References
[MZnet] E.A. Stefferud, J.N. Sweet, T.P. Domae.
"MZnet: Mail Service for Personal Micro-Computer
Systems", Proceedings, IFIP 6.5 International
Conference on Computer Message Systems, Nottingham, U.K.
(May, 1984)
[RFC821] J.B. Postel.
"Simple Mail Transfer Protocol", USC/Information Sciences
Institute. (August, 1982)
[RFC822] D.H. Crocker.
"Standard for the Format of ARPA Internet Text
Messages", University of Delaware. (August, 1982)
[RFC918] J.K. Reynolds.
"Post Office Protocol", USC/Information Sciences Institute.
(October, 1984)
[RFC923] J.K. Reynolds, J.B. Postel.
"Assigned Numbers", USC/Information Sciences Institute.
(October, 1984)