From: Ed Schouten (edfxq.nl)
Date: Tue Mar 06 2007 - 03:15:23 CST

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

* Wietse Venema <wietseporcupine.org> wrote:
> Ignoring "defer" errors with only one feature and not all the other
> ones makes Postfix less consistent. Obviously, adding a _defer_code
> to every _reject_code is not a sustainable model.
>
> Apart from that, ignoring "defer" errors makes the result of address
> verification less predictable. The same address will or won't pass
> depending on the phase of the moon and the state of the network.

The question is if a _defer_code is applicable to all features. The
problem is that in the case of sender verification this is really
useful, because there are a lot of reasons why a mail server should send
back a 450 (greylisting, maximum delivery limits, too busy, etc).

In the reject_unknown_sender_domain test, it is less useful to configure
the defer code, because DNS will tend to `always work', even when a
couple of DNS servers have been down for a couple of days.

Sender verification depends on a resource that isn't as reliable as the
one used for simple hostname checks. It would be nice to put less weight
on the outcome.

--
 Ed Schouten <edfxq.nl>
 WWW: http://g-rave.nl/

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (FreeBSD)

iD8DBQFF7TEr52SDGA2eCwURAhECAJ4o1Tq7GEgwLAU1lYeTm2r08bpHXACffvD1
L/6huLsY1FpmGqsXNhIq1hk=
=mqbm
-----END PGP SIGNATURE-----

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]