On Wed, Dec 04, 2024 at 08:01:34PM -0600, Grant Taylor via mailop wrote:
> On 12/4/24 09:45, Viktor Dukhovni via mailop wrote:
> > No, not a "421", since that would normally also be a connection abort,
> > and none of the recipients would get the mail.
>
> What do you think about a 450 4.2.1?
Well, the correct reply code is 452. The correct enhanced status code
is 4.5.3:
https://www.iana.org/assignments/smtp-enhanced-status-codes/smtp-enhanced-status-codes.xhtml
> > Rather, the correct handling is to softfail the excess recipients and
> > accept the initial batch that came in under the count.
>
> Please clarify what you mean by "softfail" in this context. I've seen
> different things and would hate to assume incorrectly.
A 4XX response, such as: 452 4.3.5 ...
Now an MTA will typically impose a smaller limit (than 100) on the count
of *invalid* recipients before the connection is closed with 421 or 521.
That is so long as the recipients are essentially all valid, it should
accept 100 + some tolerance for pipelining to have sent a bunch more,
before the first 452 was noted in response. Once enough of the
recipients are bad, abuse prevention kicks in and the RFC lower-bound
ceases to be the primary concern.
--
Viktor.
_______________________________________________
mailop mailing list
mailop@mailop.org
https://list.mailop.org/listinfo/mailop
