Hi Bill,
> > If not, what's the closest to a specification?
>
> The documentation in the software that adds it. In this case
> specifically the man page for postconf(5)
I'd already read that, e.g. prepend_delivered_header, and it doesn't
describe Postfix's logic for producing "mail forwarding loop", e.g. does
it only check on final delivery so if it's a relay then it doesn't care?
I did find mutt has
3.22. bounce_delivered
Type: boolean
Default: yes
When this variable is set, mutt will include Delivered-To headers
when bouncing messages. Postfix users may wish to unset this
variable. — http://www.mutt.org/doc/manual/
But that's again imprecise, e.g. is Postfix involvement anywhere along
the route a problem?
If the MTA imposes new rules that affect the MUA's behaviour then I'd
expect it to offer guidance as to the MUA's policy. (I realise "new"
could mean several years old here, but it's still a change.) Every MTA
and MUA cooking up their own policy seems wrong. Deviations and
broken implementations are bad enough when an RFC does exist.
--
Cheers, Ralph.
https://plus.google.com/+RalphCorderoy
