You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Kshatriya <ks...@gmail.com> on 2007/04/15 00:30:32 UTC

Question about trusted networks

Hey,

I've just read 127.* is now always trusted in the new release of 
SpamAssassin.

However, i have some mailinglists which are being handled with SmartList 
(which runs on top of procmail). So, when a spammail hits this list, it 
gets marked as spam (hopefully), but then it will be processed by 
smartlist and sent to multiple (local) addresses. As a result, it passes 
my local mailserver again, gets scanned again, but now it's in the trusted 
zone (since it's being sent from my own mailserver), so it doesn't get 
marked as spam anymore.

Is there a way to tell SpamAssassin not to "remark" this score (i.e., if 
it's spam, leave it being marked as spam, even if it gets forwarded by a 
local rule?)

Thanks!


Re: Question about trusted networks

Posted by John Rudd <jr...@ucsc.edu>.
Kshatriya wrote:
> Hey,
> 
> I've just read 127.* is now always trusted in the new release of 
> SpamAssassin.
> 
> However, i have some mailinglists which are being handled with SmartList 
> (which runs on top of procmail). So, when a spammail hits this list, it 
> gets marked as spam (hopefully), but then it will be processed by 
> smartlist and sent to multiple (local) addresses. As a result, it passes 
> my local mailserver again, gets scanned again, but now it's in the 
> trusted zone (since it's being sent from my own mailserver), so it 
> doesn't get marked as spam anymore.
> 
> Is there a way to tell SpamAssassin not to "remark" this score (i.e., if 
> it's spam, leave it being marked as spam, even if it gets forwarded by a 
> local rule?)


Instead of telling spamassassin not to re-mark it (which brings up 
issues of "when should spam assassin trust a header, and when should it 
not?"), why not just tell your mail server not to re-scan messages. 
Seems like if it came from a 127.* IP address then either:

1) it got to you via the above process, and has already been scanned 
once, so don't bother doing it again (that's just a waste of resources)

2) it's an internally generated message, and hopefully you don't 
generate spam internally, so that would again be wasted resources.


So ... why not just tell your mail server not to send messages from 
127.* to spamassassin?