mouss wrote:
Daniel L. Miller a écrit :
I would like to have a policy server checked in
smtpd_recipient_restrictions - but only if a previous permit_* check is
true. Can I do this without coding the check within the policy server?
Right now I have my standard smtpd_recipient_restrictions:
smtpd_recipient_restrictions =
reject_unknown_recipient_domain,
reject_unverified_recipient,
permit_sasl_authenticated,
permit_mynetworks,
reject_unauth_destination,
permit
I would like to have the policy service executed if either
permit_sasl_authenticated or permit_mynetworks is true.
permit_mynetworks can be replaced by a check_client_access.
for permit_sasl_authenticated, you'll have to check that in your policy
server.
What problem are you trying to solve?
PS. It is better to use port 587 (submission) for mail submission.
I'm trying to implement an auto-whitelisting solution. I took the
greylist.pl and tweaked it - and it now adds recipient addresses to a
database very simply. But I only want that to happen with valid
senders/clients. I was hoping there was something simple I could do to
take advantage of Postfix's built-in checks without having to re-invent
them or add processing time. Using port 587 is a good thought to
pre-validate local senders - but I want to support verified remote
senders as well - so I need a more universal solution.
--
Daniel
