On 2009-03-10 at 10:46 -0700, Doug Jolley wrote: > When using the vacation facility of an Exim Filter, the envelope
> address of a vacation response is the null user. This is a documented
> feature intended to help prevent runaway message sequences. I am
> noticing that more and more SMTP hosts are testing incoming mail with
> an envelope address of the null user to determine if they are bounces
> and quarantining any such messages that are not bounces. Presumably
> this is an anti-spam effort. Obviously, this creates a problem for
> incoming vacation messages. I'm just wondering whose problem this is.
> Is this an Exim problem; or, are the receiving SMTP servers
> misbehaving by quarantining these incoming messages?
The whole point of a vacation message is that it's a friendly message
for humans and should not be in a structured report format.
The remote side is being overly restrictive in what they define as a
valid auto-reply; "Auto-Submitted: auto-replied" in the vacation
response (added automatically by Exim) *should* be enough.
But fighting this will be a losing battle.
If you're feeling pugnacious, then:
RFC 3834 Recommendations for Automatic Responses to Electronic Mail.
RFC 5230 Sieve Email Filtering: Vacation Extension
Even if you're not using Sieve, its recommendations on how
user-controlled vacation responses should be constructed still carry
weight.
