You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Robert Leonard <RL...@Opto22.com> on 2004/08/11 19:21:43 UTC
SA Users whitelisting/Bayes question
Many of the emails coming to me from the SA and SURBL discussions are being
tagged as spam by SA.. What is the best way to whitelist these groups, and
secondly, is there a way to prevent the Bayes system from learning from them
(I use autolearn)..
Thanks!
Re: SA Users whitelisting/Bayes question
Posted by Steven Dickenson <st...@mrchuckles.net>.
>> Robert Leonard <RL...@Opto22.com> writes:
>> The best way is to not pass mail from these lists through
>> SpamAssassin. I use a procmail recipe which does not call
>> SpamAssassin (I actually use spamc) if the List-ID header matches.
all_spam_to spamassassin-users@incubator.apache.org
Works great here.
Steven
--
Steven Dickenson <st...@mrchuckles.net>
http://www.mrchuckles.net
Re: SA Users whitelisting/Bayes question
Posted by LuKreme <kr...@kreme.com>.
On 11 Aug 2004, at 20:37, Graham Murray wrote:
> Robert Leonard <RL...@Opto22.com> writes:
>
>> Many of the emails coming to me from the SA and SURBL discussions are
>> being tagged as spam by SA.. What is the best way to whitelist these
>> groups, and secondly, is there a way to prevent the Bayes system from
>> learning from them (I use autolearn)..
>
> The best way is to not pass mail from these lists through
> SpamAssassin. I use a procmail recipe which does not call
> SpamAssassin (I actually use spamc) if the List-ID header matches.
I just threw 5 months worth of SA-users at sa-learn --ham and haven't
had problems since. On very rare occasions I get 1 or 2 messages
mismatched, thought the way my filters work they get filtered into the
list mail anyway.
--
In England 100 miles is a long distance. In the US 100 years is a long
time
Re: SA Users whitelisting/Bayes question
Posted by Graham Murray <gr...@gmurray.org.uk>.
Robert Leonard <RL...@Opto22.com> writes:
> Many of the emails coming to me from the SA and SURBL discussions are
> being tagged as spam by SA.. What is the best way to whitelist these
> groups, and secondly, is there a way to prevent the Bayes system from
> learning from them (I use autolearn)..
The best way is to not pass mail from these lists through
SpamAssassin. I use a procmail recipe which does not call
SpamAssassin (I actually use spamc) if the List-ID header matches.
Re: SA Users whitelisting/Bayes question
Posted by Ryan Thompson <ry...@sasknow.com>.
Robert Leonard wrote to spamassassin-users@incubator.apache.org:
> Many of the emails coming to me from the SA and SURBL discussions are
> being tagged as spam by SA.. What is the best way to whitelist these
> groups, and secondly, is there a way to prevent the Bayes system from
> learning from them (I use autolearn)..
Yes, there is. As others have suggested, you could move these with
procmail, but, you're right, you need to stop them from being
autolearned, and you need SA to do that, and whitelist at the same time.
Here's what I use:
whitelist_from_rcvd spamassassin-*@incubator.apache.org apache.org
bayes_ignore_from spamassassin-*@incubator.apache.org
However, it's a maybe little easier (and perhaps more reliable) if you
have your list mail delivered to a different account. Then, you simply
need:
all_spam_to spam_destination@yourdomain.com
bayes_ignore_to spam_destination@yourdomain.com
Hope this helps,
- Ryan
--
Ryan Thompson <ry...@sasknow.com>
SaskNow Technologies - http://www.sasknow.com
901-1st Avenue North - Saskatoon, SK - S7K 1Y4
Tel: 306-664-3600 Fax: 306-244-7037 Saskatoon
Toll-Free: 877-727-5669 (877-SASKNOW) North America
Re: SA Users whitelisting/Bayes question
Posted by Dougie Nisbet <do...@highmoor.co.uk>.
On Wednesday 11 August 2004 18:46, Bob McClure Jr wrote:
[ ... ]
>
> Depends on your system. My mail is delivered by procmail, so I have a
> stanza in my .procmailrc that exclude SA list mail from SA scanning:
>
> # Don't let SA filter this traffic as it could poison the Bayes filter.
>
> :0:
>
> * ^List-Id: \"SpamAssassin Users\" <users@spamassassin\.apache\.org>
> /var/spool/mail/bob
I do something similar. When I installed Mailscanner a couple of weeks ago I
was in a bit of a quandry about whether to allow it to do the spamassassin
stuff too, but decided in the end to continue with my current procmail setup
because it allows the flexibility of selectively processing the incoming
mail. I let mailscanner deal entirely with virus scanning. I notice you use
the List-Id as your matching expression - I didn't know you could do that. It
looks slightly more elegant than my stanza:
# I don't want spamassassin processing stuff to the
# spamassassin mailing list.
:0
* ^TO_spamassassin-users@incubator.apache.org
! dougie
Re: SA Users whitelisting/Bayes question
Posted by Bob McClure Jr <ro...@earthlink.net>.
On Wed, Aug 11, 2004 at 10:21:43AM -0700, Robert Leonard wrote:
> Many of the emails coming to me from the SA and SURBL discussions are being
> tagged as spam by SA.. What is the best way to whitelist these groups, and
> secondly, is there a way to prevent the Bayes system from learning from them
> (I use autolearn)..
>
> Thanks!
Depends on your system. My mail is delivered by procmail, so I have a
stanza in my .procmailrc that exclude SA list mail from SA scanning:
# Don't let SA filter this traffic as it could poison the Bayes filter.
:0:
* ^List-Id: \"SpamAssassin Users\" <users@spamassassin\.apache\.org>
/var/spool/mail/bob
:0fw
| spamc
# Any spam with 9 or more * will be summarily punted.
:0 H
* ^X-Spam-Level: \*\*\*\*\*\*\*\*\*
/dev/null
# For testing the above
# $HOME/Mail/reallyspam
:0:
* ^X-Spam-Status: Yes
$HOME/Mail/caughtspam
Cheers,
--
Bob McClure, Jr. Bobcat Open Systems, Inc.
robertmcclure@earthlink.net http://www.bobcatos.com
Grace happens.