On 5/25/22 13:36, Wietse Venema wrote:
James Feeney:
Is that what actually happens, between Postfix and a milter,
literally "sending" and "receiving" mail messages, using SMTP? If
anyone actually knows the form of that conversation between Postfix
and a milter, and what form these "mystery macro commands" might
take there, please share.
Here is a Milter prototol description. I implemented the Postfix
Milter client implementation based on a version of this text.
https://github.com/avar/sendmail-pmilter/blob/master/doc/milter-protocol.txt
Wietse
Yes, that is the article which I first referenced in my original post, and the source of my complaint about
the misleading use of the phrase "rejects a command", ascribed to a "Milter application"
in the Postfix "Milter Readme", as currently written.
The milter protocol article uses the terminology "Accept/reject action" and "Modification
action" in reference to the *milter*, and also states: "The milter program sends responses when
(and only when) specified by the particular command code sent by the MTA."
It can be seen clearly there that "the milter program sends RESPONSES", and
that the MTA sends COMMANDS.
For the milter to "reject a command" implies that the milter is rejecting a
command from the MTA for some reason, which makes no sense, since such behavior would
appear to contradict the milter protocol described in the milter protocol article.
The milter protocol article also fails to clearly articulate whether its
capitalized text strings are literal command strings from the MTA to the
milter's socket, or whether these text strings are simply variable or function
names. So this article is useless as an illustration of a typical milter
session, with MTA Commands and milter Responses.
James
