spamass-milt-list
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Crosspost] Strange issues with user preferences and spamass-milter


From: Dan Nelson
Subject: Re: [Crosspost] Strange issues with user preferences and spamass-milter
Date: Wed, 4 Jun 2003 15:43:00 -0500
User-agent: Mutt/1.5.4i

In the last episode (Jun 04), Cassandra Lynette Brockett said:
> I don't know which header is actually the one being checked for the
> USER_IN_WHITELIST test, but the following headers contain the (in this case
> domain) information that the user has whitelisted :-
> >From  (yes, no : at the end there, this one is right at the top of the
>         message... added in by sendmail perhaps???)
> Return-Path:

^^ These two are added by sendmail, and both contain the envelope MAIL FROM
address.  From_ isn't really a header; it's used as a message separator
in the mbox mailbox format.

> To:
> Sender:
> Errors-To:
> X-BeenThere:
> <the "standard" well-formed list headers, eg List-Id, List-Unsunscribe, etc)
> 
> And the user is using "whitelist_from  *<listdomain>" as the method of
> checking....

Unfortunately, none of those other headers are checked by the
whitelist_from rule.  Here's the headers it does use:

From:
Envelope-Sender:
Resent-Sender:
X-Envelope-From:
Return-Path:
Resent-From:
 
> Just I know (as I've got sort of some free time, I have quite a bit
> of time to work on this though - I need the functionality working
> correctly), which headers are added by sendmail before it hands it
> off to procmail?  I already have one patch I need to write for
> spamass-milter (a patch to have it check the local alias database for
> the local user instead of just sending the To: user along to
> spamassassin - and a method of setting it so that it either does not
> check outbound messages, or if it does, only ones to "local"
> recipients).  Getting that sorted out is high in my list of
> priorities for this setup (in fact it's top of my list at the
> moment).

Current CVS has a -i option which takes a list of internal netblocks to
ignore, so that should fix your "don't scan outbound mail" issue at
least.  As for the alias problem, I think forking and running sendmail
-bv would be the most complete solution, except that its output isn't
designed to be machine-parsed.

-- 
        Dan Nelson
        address@hidden




reply via email to

[Prev in Thread] Current Thread [Next in Thread]