Re: [exim] How many times does router run for multiple recip…

Page principale
Supprimer ce message
Répondre à ce message
Auteur: Andrew Johnson
Date:  
À: vitas1, exim-users
Sujet: Re: [exim] How many times does router run for multiple recipients?
Is there a specific reason why you don't just discard them in the acl's ?
Ie, if you don't whitelist then discard. Exim will then quietly drop the
message if all recipients happen to be discarded.

-Andy-

> -----Original Message-----
> From: vitas1@??? [mailto:vitas1@itera.ru]
> Sent: 30 September 2008 08:48
> To: exim-users@???
> Subject: [exim] How many times does router run for multiple recipients?
>
> Hi all.
>
> I want to blackhole some recipients of a message at router level. At
> ACL
> level a whitelist of recipients for this message is built.
>
> So I'm trying to do something like this:
>
> blackholes:
>     domains = +relay_to_domains
>     driver = redirect
>     data = :blackhole:
>    # blackhole ONLY those who aren't in whitelist
>     condition = ${if
> match_address{$local_part@$domain}{addr1@???:addr2@???:addr3@???}{no}{y
> es}}

>
>    # at verification time whitelist doesn't exist
>     no_verify

>
> Now suppose a message has multiple recipients and some of them are in
> whitelist.
>
> When processing any of the whitelisted recipients router gives me
> "...skipped: condition failure". And NO other addresses are then
> processed
> - i.e., message is delivered to all remaining recipients (including
> non-whitelisted). There are NO entries in debug log concerning whether
> THIS router REALLY processed remaining addresses.
>
> So seems like the first {no} result completely stops the work of the
> router.
>
> I tried to use simple conditions (condition = false) and has the same
> effect.
>
> So my question is: does router run ONCE for a message that has multiple
> recipients or for EVERY recipient separately? Are there some exceptions
> for REDIRECT router or for :BLACKHOLE: data particularly? Is my
> configuration completely stupid? :-)
>
> Thanks in advance,
>
> Vitas.
>