On 27 Feb 2019, at 11:16, Marco Franceschetti via mailop wrote:
Hello,
We at contactlab are considering a change in the deactivation of hard
bounces.
Currently, we suppress not existing mailboxes at the first hit.
We are aware of a small percentage of false positives.
Recent admissions criteria for Certified Senders states:
"The CSA sender must take email addresses from mailing lists, if,
after sending to this address,
the mailbox is identified as non-existent; at the latest, however,
this must occur after three hard
bounces".
We are evaluating to remove not existing mailboxes from the lists of
our clients after the second hit instead of the first one.
Do you have any considerations, suggestions about this?
There are subtle but important distinctions between types of "hard
bounce" which you should take into account. ANY 5xy reply in SMTP (or
asynchronous DSN message citing a 5xy reply) should be considered a
"hard bounce." However, there are specific basic and enhanced SMTP reply
codes which are direct explicit statements that an address is
non-existent which should be honored immediately rather than taken as
possibly mistaken and retested later with a different message. For
example, a 550 reply to RCPT (or either stage of DATA, if there's only
one accepted RCPT) without any enhanced status code should ALWAYS be
treated as an indication of a non-existent address, as should 550
followed by any 5.1.x enhanced status code. It is debatable how other
5xy + 5.x.y combinations at various stages should affect sending a
different message at a later time to the same address, but as every
modern SMTP RFC has made clear: ANY 5xy reply should be considered a
"hard bounce" for the specific transaction being tried. That means you
must not try to resend the same message to the same address in any way:
not 5 minutes later, not through a different outbound IP or to a
secondary MX, not with a different envelope sender address, NOT AT ALL.
It is good practice to remove an address from a list after just one hard
bounce of the subset that clearly indicate that an address does not
exist or after consecutive hard bounces (i.e. for different messages)
with less certain meanings. Whether you make 2 or 3 your limit for hard
bounce codes that might indicate problems other than address
non-existence is unlikely to make much difference.
--
Bill Cole
[email protected] or [email protected]
(AKA @grumpybozo and many *@billmail.scconsult.com addresses)
Available For Hire: https://linkedin.com/in/billcole
_______________________________________________
mailop mailing list
[email protected]
https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop