On 13 Oct 2006, at 11:17, W B Hacker wrote:
> Stuart Gall wrote:
>> Hello,
>> The documentation on the spam = acl condition is a bit sparse. It
>> says alot about what you can do, but not much on how you do it.
>> Nothing much in the FAQ
>> It claims you can implement a per domain or per user spam setting.
>>
>> spam = X
>>
>> What is X is it envelope to local part ? How then do I do per domain
>> filtering?
>>
>> I tried
>> accept spam = postmaster
>>
>> in a data ACL I would expect this to allow all mail to the
>> postmaster, however it appeared to allow all mail to everyone.
>> i.e. it functions as
>> accept spam=nobody would
>>
>> exim 4.54 by the way.
>>
>>
>> There are basically two things I want to do
>> 1) Set specific reject levels per virtual domain
>> 2) set a higher level for the postmaster
>>
>> TIA
>> Stuart
>>
>
> # DATA_SCAN_10: Call SA IF not flagged otherwise
> #
> warn
> !condition = ${if eq{$acl_m0}{1}}
> spam = spamd:true
> add_header = X-Spam-Class: Message was classified as SPAM at
> $spam_score
> log_message = DS10 Spam test/score $spam_score
>
So can I infer that spam = X
where X is the user of the process under which spamd is running - hmm
that wont fly on a virtual host.
>
> Before reaching this stage, a 'warn' verb was used to set an acl
> 'flag' for
> certain authenticated clients, etc. This also controlled skipping
> other checks.
>
> And/or you can add the line something like this one:
>
> !local_parts = postmaster : abuse : hostmaster
I would love to, but I think that I cant use local_parts in the
data_acl right ?
>
> - or whatever fits your need...
>
> --
> ## List details at http://www.exim.org/mailman/listinfo/exim-users
> ## Exim details at http://www.exim.org/
> ## Please use the Wiki with this list - http://www.exim.org/eximwiki/
>
