One likely scenario may be that the spammer managed to hack into an existing
account, then use it to send out their garbage. One way to fix that is to
ensure all humans with computer access always employ best practices for
choosing and protecting secure passwords.
Another possible scenario is
It doesn't really work that way. Bayes is just one part of the picture and in
order to get good results you have to turn the full toolkit loose on the
problem; I'm not sure Bayes by itself should be expected to achieve 95%
recognition anyway. The main flaw in your current plan is that once you
It "applies" all the rules, in the sense of testing each message for
each condition. If a message matches the conditions of a rule then that
is considered a "hit" on that rule. Rules that "hit" on a message are
listed in the report. Messages that appear, to the human observer, to
be very similar
If what you presented in your message is actually the command you used, then it
might be looking for some input from the keyboard - you don't illustrate having
specified the particular file you want it to use following the '--mbox' option,
you have "--ham" in that position on the line. I have n
In this situation I believe Spock would say "Insufficient Data" . . .
What o/s are you running? What is your mail handling software? How does that
mail handling software interface to SpamAssassin? Are you sure the items were
not scanned, or are you simply bothered that they were not marked as
That doesn't look much like a SpamAssassin option there, to me.
Perhaps you may get more useful responses if you give us more detail about your
system configuration.
What mailserver are you running?
How does it invoke SpamAssassin?
Do you have a virus scanner installed?
What operating system
SpamAssassin does not "handle" mail. SpamAssassin analyzes a message and
returns a score/report to whatever asked for the analysis. That is all.
Other products "do things" with mail - store/reject/accept/deliver, etc. - and
some of those products use a SpamAssassin score as part of the basis f
Let me see if I follow you correctly there . . . you are administrator of an
email server, but you do not like to read and write email?
Also, I am not a lawyer, but I think I read something somewhere a while ago
that there is some intellectual property rights ownership associated with
'spamassa
(apologies for top posting, but the email software here does not really do
quoting in a way that works out well otherwise)
If your mail contains SpamAssassin headers then it was (obviously) processed
through SpamAssassin. Just because you have BL checks in your MTA does not
necessarily mean th
It means that if you were using BL at MTA level your SA might never have seen
the message at all.
No your rule would not be "overlooked" 'because the site is in a blacklist'
*unless* you were using the BL in your MTA and rejected the transaction from a
blacklisted IP address and, thus, never su
The most obvious problem is that you are re-using the rule name. While the
configuration is parsed the 2nd line replaces the first then the 3rd line
replaces the 2nd line. If you want three rules give them three different
names, for example: whitelist_from_luser1 whitelist_from_luser2
whiteli
>>> "Benny Pedersen" 06/28/09 12:42 AM >>>
>On Sun, June 28, 2009 05:38, Cory Hawkless wrote:
>> I agree, wouldn't it be easier to uniformly feed all of these type of URL's
>> though the already existing SA filters. As Jason suggested maybe by
>> collapsing whitespaces?
>
>lets redefine how a url
Well I suppose you could always take the product that you dislike so badly back
to the store and ask for a refund of your purchase price. Sometimes it really
amazes me how much, and how severely, some people will gripe about free
products that exist only because other people volunteer their tim
Well maybe you should figure out what is going on with these two: RE_PASSWORD
100.00, RE_PASSWORDV 100.00
since your choice of "-100" (it is not a magic pass value, just another factor
in the arithmetic) for your manual whitelist only counteracts one of them ...
or run your manual whitelist scor
Yes, the learn client does not try to keep up with what it has done, or not
done, before - that is handled by the server (the Bayes engine).
I believe there is no reasonable way for the client to achieve this, anyway -
it cannot reliably modify your maildir in such a way that it can be assured o
Artificial intelligence will never overcome natural stupidity (or the clever
ingenuity of criminals) ... if people actually DO that (copy the "url" and
remove the spaces) there is some temptation to say they get what they deserve
... but on the other hand most of the spam/scam stuff out there is
ecipient that the email had
been blocked as such. From my understanding there is no option available
short of writing a program. Would this be correct?
Kevin Parris-2 wrote:
>
> Spamassassin did not put the message in the spam folder. SA does not know
> if the item is going to be put in a fold
Spamassassin did not put the message in the spam folder. SA does not know if
the item is going to be put in a folder (spam or otherwise) or tossed into the
bit bucket. SA doesn't even, necessarily, actually know who the item is to -
or which of possible multiple recipients might want to be not
>>> support <[EMAIL PROTECTED]> 12/11/08 2:52 AM >>>
Prempting some responses:
What about external remote workers?
What about those who email stuff to themselves?
I hear this kind of thing all the time when people moan about spoofing.
On Wed, 2008-12-10 at 12:19 -
You do not have a SpamAssassin problem, you have a Communigate problem.
Present this issue to your support resources for that product.
The basics of what you want to do are something like this:
When a message is arriving from the internet, and has your own domain in the
Return-path, it should
Maybe this is a completely crazy notion, but if the mail for these accounts is
in fact actually flowing into/through your system, and being sent through SA
already, you might create a rule so that any item with one of those addresses
in it gets a high score so in turn your auto-learn threshold w
You could write yourself a rawbody rule to match on the string: td>NEVOB>> "Dietmar Maurer" <[EMAIL PROTECTED]> 08/14/08 1:53 AM >>>
Recently there are tons of simple mails like:
ftp://pve.proxmox.com/tmp/sample-spam1.txt
ftp://pve.proxmox.com/tmp/sample-spam2.txt
Seems that they trigger some
Sample posted here: http://pastebin.com/m7d993dc7
Have seen several similar to this, the message contains only random words, no
images, no web links. What's the point? It's not advertising, or trying to
lure victims to a site, or carrying any payload. Commentary anyone?
Spammers operate on the premise that lots of stupid people read email. For
example, only stupid people would actually respond to an offer to sell
medications, from a service that does not spell the product name correctly
(they are either too stupid to recognize the deviant spelling even though
Well now, if a spambot actually does start recognizing and avoiding his system,
doesn't that mean he wins and the spammer loses?
>>> John Hardin <[EMAIL PROTECTED]> 05/08/08 12:11 PM >>>
On Thu, 8 May 2008, Marc Perkel wrote:
> To participate all you have to do is set your highest numbered MX t
Maybe this is overstating the excruciatingly obvious, but why don't you just
compose the email you anticipate sending, and SEND IT to yourself so that it
comes into your email box by way of your SpamAssassin with score report headers
added, and take a peek at that result? Or if you're working w
If I have followed the discussion correctly so far, the explanation for
manual-learn not being distinguished from auto-learn is this: no matter what
mode of learning caused a token to appear in the database, if there is ongoing
mail traffic that "hits" on the token then said token will not expi
I think it might be easier if you would simply have a conversation with
the techy folks at your customers- invite them to configure THEIR system
so that either everything from YOUR system is OK no matter what spam
status it has (they can route it to bit-bucket or whatever) or turn off
the reject-no
28 matches
Mail list logo
