Tony Finch wrote:
On Tue, 17 Jun 2008, Olaf Hopp wrote:
So I thought I must "somehow" modify my SA-ACL doing a "DENY" when
spamscore is above some treshold and final destination is not inhouse.

I thought I could accomplish this via "address_data" set by my router,
which is handling remote deliveries and then asking within the data-acl
for the value of address_data, but it is always unset :-(

Your idea is generally good. The problem is that address_data is not set
in the data acl because its value is specific to a single address, but a
message can have more than one recipient. What you need to do is decide
how to deal with a message that is sent to multiple recipients some of
which are delivered locally and some of which are forwarded. For example,
you might decide to reject spam in this case, just like messages to one
forwarding recipient. To implement this policy, you can set an acl_m
variable in the rcpt acl if the address_data indicates the recipient is
forwarded and leave it alone in other cases. Then check the acl_m variable
in the data acl to see if it is set and reject the message if the spam
score is too high. You can change the logic for setting the variable
according to the ploicy you want.

Thanks Tony,
I was just about reading about the acl-variables when you answer arrived.
My solution is as follows:

rcpt_acl:
  accept  domains = +local_domains
          endpass
          message    = Unknown user.
          verify     = recipient
          set acl_m0 = $address_data

And the router handling remote deliveries:

lookuphost:
  driver = dnslookup
  domains = ! +local_domains
  transport = remote_smtp_out
  ignore_target_hosts = 0.0.0.0 : 127.0.0.0/8
  address_data = remote

The router doing local deliveries does not set address_data

Further in the data_acl
  deny    message    = No relay of spam to foreign addresses
          condition  = ${if >{$spam_score_int}{50}{1}{0}}
          condition  = ${if def:acl_m0 {1}{0}}

I still have the (minor) problem as Tony pointed out to decide
what to to if there is at least one remote and one local delivery.
But this is a question of politics that technics :-)

Olaf


--

==============================================================================
     __0
   _-\<,_     Dipl.-Geophys. Olaf Hopp
  (_)/ (_)    ATIS - Abteilung Technische Infrastruktur

University of Karlsruhe          EMail: [EMAIL PROTECTED]
Faculty of Computer Science      WWW  : http://www.atis.uni-karlsruhe.de
Building 50.34 Room-No. 009
Am Fasanengarten 5               Fon  : +49 (721) 608-3973
D-76131 Karlsruhe / Germany      Fax  : +49 (721) 608-6699

==============================================================================

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

-- 
## List details at http://lists.exim.org/mailman/listinfo/exim-users 
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/

Reply via email to