You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Dietmar Lippold <di...@web.de> on 2004/10/19 00:19:18 UTC

network tests

I have some questions about network tests in spamassassin version 3.0.0
(on Fedora 2):

 * Which tests (SYMBOLIC_TEST_NAMEs) are not used when I run
   spamassassin or spamd which option "-L"?

 * How can I disable network tests by entries in user_prefs (e.g. when
   I use spamc but spamd runs without option "-L")? I put these lines in
   user_prefs:
>>>
use_dcc     0
use_pyzor   0
use_razor2  0
skip_rbl_checks  1
dns_available    no
<<<

   Then spamassassin without option "-L" do not use network tests (don't
   generate network traffic) but spamc generate network traffic (do use
   some network tests). Why there is a difference?

 * When I disable network tests by entries in user_prefs, which values
   of the score entries are used (thoses with or without network tests)?

Dietmar

Re: network tests

Posted by Dietmar Lippold <di...@web.de>.
Am Mittwoch 20 Oktober 2004 04:46 schrieb Robert Menschel:
> Monday, October 18, 2004, 3:27:58 PM, Theo wrote:
>
> TVD> On Tue, Oct 19, 2004 at 12:19:18AM +0200, Dietmar Lippold wrote:
> >>  * Which tests (SYMBOLIC_TEST_NAMEs) are not used when I run
> >> spamassassin or spamd which option "-L"?
>
> TVD> any that have a "tflags net" set.
>
> >>  * When I disable network tests by entries in user_prefs, which
> >> values of the score entries are used (thoses with or without
> >> network tests)?
>
> TVD> -L is the only thing that impacts scoreset.  If you don't use
> -L, TVD> sets 1 and 3 are used, with -L it's 0 and 2.
>
> I'm on the road and so can't test this myself this week. Can anyone
> answer a quick follow-up question:
>
> If running spamassassin directly, not spamc/spamd, if we added the
> line
>
> > tflags  rule_name  net
>
> to that user's user_prefs file, could we effectively ensure that
> rule_name ran only when network tests ran, and not when -L was
> specified?

I just tested it (sorry for the dely). I put the following entries in my 
user_prefs:

>>>
tflags  MSGID_DOLLARS  net
tflags  RATWARE_ZERO_TZ net
tflags  UNWANTED_LANGUAGE_BODY  net
tflags  HTML_60_70 net
<<<

When I run spamassassin without -L the header of the mail contains:

>>>
X-Spam-Status: Yes, score=10.3 required=5.0 
tests=AWL,DATE_IN_FUTURE_06_12,
        HTML_60_70,HTML_MESSAGE,MIME_HTML_ONLY,MSGID_DOLLARS,RATWARE_ZERO_TZ,
        UNWANTED_LANGUAGE_BODY autolearn=no version=3.0.0
X-Spam-Report: =?ISO-8859-1?Q?
        *  1.2 DATE_IN_FUTURE_06_12 Absendezeit 6 bis 12 Stunden nach 
Datum in "Received"-Kopfzeilen
        *  0.0 HTML_60_70 BODY: Nachricht besteht zu 60-70% aus HTML
        *  0.0 HTML_MESSAGE BODY: Nachricht enth=e4lt HTML
        *  2.8 UNWANTED_LANGUAGE_BODY BODY: Nachrichtentext in 
unerw=fcnschter Sprache
        *  1.2 MIME_HTML_ONLY BODY: MIME-Nachricht besteht nur aus HTML
        *  2.4 RATWARE_ZERO_TZ Seltsame Zeitzone (+0000)
        *  3.1 MSGID_DOLLARS Muster in Kopfzeile "Message-ID" typisch 
f=fcr Spam        * -0.4 AWL AWL: From: address is in the auto 
white-list?=
<<<

When I run spamassassin with -L the header of the mail contains:

>>>
X-Spam-Status: Yes, score=9.2 required=5.0 
tests=AWL,DATE_IN_FUTURE_06_12,
        HTML_MESSAGE,MIME_HTML_ONLY,MSGID_DOLLARS,RATWARE_ZERO_TZ
        autolearn=no version=3.0.0
X-Spam-Report: =?ISO-8859-1?Q?
        *  1.0 DATE_IN_FUTURE_06_12 Absendezeit 6 bis 12 Stunden nach 
Datum in "Received"-Kopfzeilen
        *  0.0 HTML_MESSAGE BODY: Nachricht enth=e4lt HTML
        *  1.2 MIME_HTML_ONLY BODY: MIME-Nachricht besteht nur aus HTML
        *  3.2 RATWARE_ZERO_TZ Seltsame Zeitzone (+0000)
        *  3.0 MSGID_DOLLARS Muster in Kopfzeile "Message-ID" typisch 
f=fcr Spam        *  0.7 AWL AWL: From: address is in the auto 
white-list?=
<<<

As you can see, the rules MSGID_DOLLARS and RATWARE_ZERO_TZ
are used with -L, but the rules UNWANTED_LANGUAGE_BODY and
HTML_60_70 are not used.

It seems to depend on the rules.

> And does this also apply to mass-check runs?

I don't know.

Dietmar

Re[2]: network tests

Posted by Robert Menschel <Ro...@Menschel.net>.
Monday, October 18, 2004, 3:27:58 PM, Theo wrote:

TVD> On Tue, Oct 19, 2004 at 12:19:18AM +0200, Dietmar Lippold wrote:

>>  * Which tests (SYMBOLIC_TEST_NAMEs) are not used when I run
>> spamassassin or spamd which option "-L"? 
TVD> any that have a "tflags net" set.

>>  * When I disable network tests by entries in user_prefs, which
>> values of the score entries are used (thoses with or without
>> network tests)?  
TVD> -L is the only thing that impacts scoreset.  If you don't use -L,
TVD> sets 1 and 3 are used, with -L it's 0 and 2.

I'm on the road and so can't test this myself this week. Can anyone
answer a quick follow-up question:

If running spamassassin directly, not spamc/spamd, if we added the
line 
> tflags  rule_name  net
to that user's user_prefs file, could we effectively ensure that
rule_name ran only when network tests ran, and not when -L was
specified?

And does this also apply to mass-check runs?

Thanks.

Bob Menschel



Re: network tests

Posted by Theo Van Dinter <fe...@kluge.net>.
On Tue, Oct 19, 2004 at 12:19:18AM +0200, Dietmar Lippold wrote:
>  * Which tests (SYMBOLIC_TEST_NAMEs) are not used when I run
>    spamassassin or spamd which option "-L"?

any that have a "tflags net" set.

>  * How can I disable network tests by entries in user_prefs (e.g. when
>    I use spamc but spamd runs without option "-L")? I put these lines in
>    user_prefs:

you'd have to disable all of the net rules -- some are done via the config you
posted, others will require rule disabling (score 0)

>  * When I disable network tests by entries in user_prefs, which values
>    of the score entries are used (thoses with or without network tests)?

-L is the only thing that impacts scoreset.  If you don't use -L, sets 1 and 3
are used, with -L it's 0 and 2.

-- 
Randomly Generated Tagline:
"Your next question is 'How does this gate work?'  I don't know.  I
 don't have to know, I'm not an Electrical Engineer, I'm a Computer
 Scientist."                  - Prof. Hamel