manu@... writes:
Instead, I think it makes sense to have milter-greylist include the
functionality of spamass-milter (but not spamd or SA), and it seems with
the dacl for SA we are well on the way.
(One important thing is at least mild greylisting before running SA for
connections somewhat likely to be spam, but the rbl racl rules do that
very nicely.)
The next thing I would like most is variable greylisting delay depending
on spam score. Has anyone tried to do this? If so, could they share
their config?
> Yes, it would be nice to have a DATA-stage ACl clause that would lookupThere's a slippery slope here to reimplementing all of spamassassin.
> URL found in the message against domain black lists. This is a bit
> complicated to implement, though.
Instead, I think it makes sense to have milter-greylist include the
functionality of spamass-milter (but not spamd or SA), and it seems with
the dacl for SA we are well on the way.
(One important thing is at least mild greylisting before running SA for
connections somewhat likely to be spam, but the rbl racl rules do that
very nicely.)
The next thing I would like most is variable greylisting delay depending
on spam score. Has anyone tried to do this? If so, could they share
their config?