In message <045a01c5a5f9$231b7950$6401a8c0@sativa>, Steve B
<exim@???> writes
>"Dave Lugo" <dlugo@???> writes:
>
>>> Well, I've got a problem with an SMTP server refusing a null DSN.
>>> Apparently this guy John Levine (iecc.com) refuses any connects with a
>>> MAIL FROM: <> and thus callout verification fails.
>>
>> Eh... I suspect your statement "refuses any connects with a MAIL FROM: <>"
>> is not entirely accurate.
>
>No, I should say he responds with "250 OK" if you have something between the
>"<>" and he responds with a "553 Not our message (#5.7.1)" if you simply
>send a null "<>"
you could always process this specific 553 message (rather than adding
the whole of rfc-ignorant to your exceptions)
>> If he never sent the original item that resulted in the bounce,
>> WHY would he want to accept the blowback that the bounce represents?
>
>There is no original item
exactly John Levine's point : and he suspects that you are about to send
him a DSN in which he has no interest whatsoever, and so he declines to
talk with you any further.
>So somehow sending a "553" reduces the load on his system versus sending a
>"250 OK"?
If it _was_ a bounce (and not one of Exim's callbacks) then there would
be further traffic ...
Your heuristic (generating extra traffic to strangers in the hope of
reducing spam) broke because it encountered John's heuristic (guessing
at the nature of a transaction to reduce bandwidth)...
Appealing to RFCs is not going to help you here (this is more to do with
faith than protocols) -- so get over it and tweak the nature of your
heuristic in some manner.
At least you got an informative error message !
- --
richard @ highwayman . com "Nothing seems the same
Still you never see the change from day to day
And no-one notices the customs slip away"
