On Mar 16, 2021, at 5:32 PM, Dave Crocker <dhc(_at_)dcrocker(_dot_)net> wrote:
That's nice, but you won't be surprised to hear that there are other
implementations that /do/ pay attention to more than the first digit.
Sure, but I worry that these are somewhat brittle, because with all
the manual anti-spam measures that users tweak into the rulesets of
production MTAs via lookup tables and milters, where the user gets to
specify the 4XX or 5XX response code, it is much to optimistic to
expect them to get these right.
Were the (XX) in 4XX or 5XX in response to the usual MAIL transaction
commands (MAIL, RCPT, DATA, BDAT, ".") to actually matter, it would be
liable to get confused by all manner of misleading signals.
The non-transaction commands (AUTH, STARTTLS, ...) tend to see less
random meddling from milters.
ietf-smtp mailing list