Re: [exim] Brain Teaser - how would you do this?

Pàgina inicial
Delete this message
Reply to this message
Autor: W B Hacker
Data:  
A: exim users
Assumpte: Re: [exim] Brain Teaser - how would you do this?
Marc Perkel wrote:

> OK - here's what I want to do. I want to forward a copy of selected
> email to another email address, but only if the email address that I
> want to forward to is a valid working email address. So - I need to do a
> callout and check if the email address is valid before I forward a copy.
>
> How would I do that? Thanks in advance.
>
>


Presuming you mean to a dynamic or unpredictable set of
potential addressees (as a fixed list or one of your own servers
is easy), my answer to 'how' would have to be:

'poorly, and expensively if at all'.

- Suppose, for example, that you turn the 'verify_sender' tool,
sort of bass-ackwards (keeping in mind that it can work in
'steps', and may not reach a specific local_part stage anyway).

- or that you even go so far as to generate and send an advance
'test' message, hold real traffic until your logs show
successful acceptance at the far end (IF/AS/WHEN they can even
do so)

Either of which is a RPITA to configure and maintain, and gets
you into queue issues, as well as potentially fighting caching,
and perhaps even hints DB.

In both instances, you will still be at the mercy of MTA's that
accept, then either discard invalid addresses later, or dump
them into a 'catch-all'. Either being 'invisible' to you at
smtp-time or afterwards.

Qmail, for example, but loads of other MTA's also - depending on
how configured.

IOW - you are at the mercy of the consistent behaviour of MTA
designers and admins of peer MTA's in general.

An oxymoron, no? Even/especially with your *own* servers?

'Simple' mtp was not designed for a store, pre-validate,
conditionally-forward process.

X.400 - or even late-Holocene TWIX or Telex (WRU) is another
matter, as there are directories, and one can check address
validity separately from transporting actual traffic.

But TWX and Telex are no longer a factor, and X.400 not far
behind - outside of a few specialty organizations.

HTH,

Bill