Author: Tom Marazita Date: To: exim-users Subject: Re: [Exim] Wish list (I think) regarding sender verify callout.
Philip Hazel wrote: > I personally take the hard line on this issue. We should make life
> difficult for the users of MTAs that are unable to accept bounce
> messages. That way, those users might put enough pressure on their
> sysadmins to fix their MTAs.
I completely concur, and this is the path I have taken so far.
> Nevertheless, I realise that some people come under political pressure
> to break the rules. I am prepared to contemplate some kind of scheme
> (details as yet unknown because I haven't thought about it) by which you
> can say "if callout fails because MAIL was rejected, still accept the
> message". What I do not think is a good idea is using something other
> than <> in the MAIL command.
Unfortunately, the reaction of both our faculty and of the faculty/researchers they
are in communication with is "we must have our mail communications function."
Oh course, they also direct "we must not receive spam, and mail must
be secure." Oh well... (spamassasin goes a long way to help with the spam :-)
I don't like the solution of accepting messages this way either, but
it seems like my only choice if I wish to meet both the conditions
of using callback (as best as possible) and allowing all (most) legitimate
messages to arrive.
> Did you contact the senders to tell them they are using a system that
> doesn't accept bounces?
Yes. As well as their postmasters. In all but one case (where
I actually went and pointed out to them the option they needed
to fix in their rather odd mail transport software) the postmasters were
silent; a couple of legitimate sites didn't even have valid postmaster
addresses. Invariably the response from the user-side was "I don't understand
what you're talking about or what a 'bounce' is", etc.
Oh well, such is the job I guess. In any case, I appreciate your
willingness and effort to consider this. I really don't like
breaking the protocol either.
