RE: -r option

[Jeff Powell]

I tried changing the debug level but I did not see anything that wasn't
there before.  It was almost like the option had no effect.  Here is the
log from  a spam message I sent from a Yahoo account to myself.  As you
can see, it does rewrite the header with the X-Spam-Status flag.  I have
a user called spamd for the milter.

Mar 16 23:38:45 titan milter-sender: 5912: create per-connection
resources
Mar 16 23:38:45 titan milter-sender: h2H7cjS4005911: checkSenderMx()
start
Mar 16 23:38:45 titan milter-sender: h2H7cjS4005911: sender
<address@hidden> cached, skipping
Mar 16 23:38:45 titan milter-sender: h2H7cjS4005911: destroyed
per-message resources
Mar 16 23:38:45 titan milter-sender: h2H7cjS4005911: checkSenderMx() end
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911:
from=<address@hidden>, size=1695, class=0, nrcpts=1,
msgid=<address@hidden>, proto=ESMTP,
daemon=MTA, relay=nixon.hostnoc.net [66.197.183.100]
Mar 16 23:38:46 titan milter-sender: h2H7cjS4005911: normal end of
message...
Mar 16 23:38:46 titan milter-sender: h2H7cjS4005911: destroyed
per-message resources
Mar 16 23:38:46 titan spamd[2545]: connection from localhost.localdomain
[127.0.0.1] at port 58736
Mar 16 23:38:46 titan spamd[5915]: processing message
<address@hidden> for spamd:65534.
Mar 16 23:38:46 titan spamd[5915]: identified spam (11.7/4.7) for
spamd:65534 in 0.5 seconds, 1742 bytes.
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911: Milter add:
header: X-Spam-Flag: YES
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911: Milter add:
header: X-Spam-Status: Yes, hits=11.7
required=4.7\n\ttests=BANG_GUARANTEE,CLICK_BELOW,EXCUSE_15,FROM_ENDS_IN_
NUMS,\n\t      HOT_NASTY,HTML_30_40,HTML_MESSAGE,PLING_PLING,\n\t
SUBJ_HAS_SPACES,SUBJ_HAS_UNIQ_ID,THIS_AINT_SPAM\n\tversion=2.50
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911: Milter add:
header: X-Spam-Level: ***********
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911: Milter add:
header: X-Spam-Checker-Version: SpamAssassin 2.50 1.173-2003-02-20-exp
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911: Milter change:
header  Subject: from Make money fast!!!!!                 sdfsdfdc to
*****SPAM***** Make money fast!!!!!                 sdfsdfdc
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911: Milter change:
header  Content-Type: from multipart/alternative;
boundary="0-1660296513-1047886720=:19788" to multipart/mixed;
boundary="----------=_3E757B86.7CFA11FD"
Mar 16 23:38:46 titan sendmail[5911]: h2H7cjS4005911: Milter message:
body replaced
Mar 16 23:38:46 titan milter-sender: 5912: destroyed per-connection
resources
Mar 16 23:38:46 titan sendmail[5917]: h2H7cjS4005911:
to=<address@hidden>, delay=00:00:01, xdelay=00:00:00,
mailer=esmtp, pri=30896, relay=olympus.jeffandclaire.com.
[64.164.192.98], dsn=2.0.0, stat=Sent (
<address@hidden> Queued mail for
delivery)

-----Original Message-----
From: Dan Nelson [mailto:address@hidden 
Sent: Sunday, March 16, 2003 11:03 PM
To: Jeff Powell
Cc: address@hidden
Subject: Re: -r option


In the last episode (Mar 16), Jeff Powell said:
> Interesting... I am also using SpamAssassin 2.5.  I did not try using 
> the -r option until after I upgraded to it.
> 
> Does the -r option merely pass a flag to SpamAssassin, or does it do 
> something internal to the milter?  Perhaps the option has changed in 
> SA. I'll have to review the documentation.  I know several options are

> different in 2.50, I had to make some changes when I upgraded.

-r tells the milter to tell sendmail to reject a message if SA has
scored it above the number given (whether SA actually tagged it as spam
or not).  If you run spamass-milter with a debug level of 1, you should
see lines saying "SA score: 10", and "Rejecting", showing the score it
extracted from the SA headers, and whether it decided to reject the
message or not.

I'm running SA 2.50 here and -r works fine.  It does rely on the
X-Spam-Status: header, though, so if you have SA configured not to add
that, then -r has nothing to get the score from.

-- 
        Dan Nelson
        address@hidden