On 10/15/2014 12:50 PM, Miles Fidelman <mfidelman@meetinghouse.net> wrote:
I'll close by noting that this branch of discussion started with a focus
on silently dropping spam, and whether that's a violation of standards.
Actually, no, this branch started with a focus on whether or not it is a
good idea to break SMTP by accepting email from *invalid recipients*
then silently deleting them, as opposed to rejecting them at the RCPT-TO
stage.
It used to be a clear violation of the various MUST statements re.
sending non-delivery messages. It looks like more recent standards now
allow for dropping spam as a specific exception case.
My position is that:
1. email to invalid recipients should be rejected at the RCPT-TO stage,
2. under *no* circumstances should mail to invalid recipients be
accepted for delivery then silently deleted based solely on that one
criteria,
and
3. once an email has been accepted for final delivery, every effort
should be taken to deliver the message to the recipient, whether to
their Inbox clean or tagged as spam (if a spam threshhold is met), or to
a spam quarantine,
I allow for the very rare 'clear-and-present-danger' exceptional
circumstance that, if an after-queue content scanner determines with a
very high probability that something contains a malicious payload, an
admin might want to not deliver it to the recipient. But, I would also
argue that it should go into a quarantine that only the admin has access
to, and never just silently deleted.
But, as Jerry says, that is just my opinion...