You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by tony <ne...@gmail.com> on 2013/05/31 12:52:29 UTC

increase in spams getting though

I've been running spamassassin for about 4 years on a centos 5 server,
currently running spamassassin-3.3.1-2.el5. I've always been impressed and
very pleased with the way SA works and I regularly run sa-update and I
update the bayes databases. I use plugins too - raor, pyzor, dkim etc but
over the last couple of months I've seen SA's effectiveness wane, with more
spam emails slipping though into my inbox.
There are two scenarios here, either my system is not working as it should,
or spammers have found a new clever way to beat the filters and SA is yet to
cactch up. So my first thought in posting here is, has anyone else
experienced an increase in penetration by spammers in the last couple of
months?



--
View this message in context: http://spamassassin.1065346.n5.nabble.com/increase-in-spams-getting-though-tp104964.html
Sent from the SpamAssassin - Users mailing list archive at Nabble.com.

Re: increase in spams getting though

Posted by Per Jessen <pe...@computer.org>.
Martin Gregorie wrote:

> On Fri, 2013-05-31 at 03:52 -0700, tony wrote:
>> I've been running spamassassin for about 4 years on a centos 5
>> server, currently running spamassassin-3.3.1-2.el5. I've always been
>> impressed and very pleased with the way SA works and I regularly run
>> sa-update and I update the bayes databases. I use plugins too - raor,
>> pyzor, dkim etc but over the last couple of months I've seen SA's
>> effectiveness wane, with more spam emails slipping though into my
>> inbox.
>>
> For the last few months almost all my spam stream has been Stock
> pump'n'dump scams (no particular TLD associated with it) and a small
> admixture of marital aid spam (almost all from Russia).

Completely different to what I'm seeing - of the stuff that is getting
through: very, very few pumpndump, loads of cheap loans&mortgages,
flexible hoses, new cars, car lease, learn foreign languages, Ruby
Palace, dating/mating service.



-- 
Per Jessen, Zürich (10.9°C)
http://www.dns24.ch/ - free DNS hosting, made in Switzerland.


Re: increase in spams getting though

Posted by RW <rw...@googlemail.com>.
On Fri, 31 May 2013 12:15:54 -0700 (PDT)
tony wrote:

> now I've given it some more thought it's a certain type of spam that
> gets through. I never get scams or marital aid spam, never. The spam
> mails that get through are consistently to do with hospitality,
> increased sales and business courses. Are there rules I should be
> using to combat this type of spam?

If you don't want to spend a lot of time managing your own local rules,
I'd recommend scoring in a second statistical filter, one that supports
multi-word tokenization. These can do better than Bayes with type of
spam you mentioned because they can match patterns made out of words
that are themselves neutral. I say patterns because once you get
beyond 2-words the tokens can include gaps. 

I use Bogofilter configured for 4-word tokenization and DSpam with
orthogonal sparse bigram tokenization, but I've scored down DSpam
because it was contributing to most of my FPs. 

I recommend trying Bogofilter. No need for a plugin, just pipe your mail
through it and score the x-Bogosity  header.

Re: increase in spams getting though

Posted by Martin Gregorie <ma...@gregorie.org>.
On Fri, 2013-05-31 at 12:15 -0700, tony wrote:
> now I've given it some more thought it's a certain type of spam that gets
> through. I never get scams or marital aid spam, never. The spam mails that
> get through are consistently to do with hospitality, increased sales and
> business courses. Are there rules I should be using to combat this type of
> spam?
> 
You can train Bayes to recognise these messages as spam.

Everybody's spam stream is a bit different, so you could also analyse a
moderate collection of these messages for common phrases and write a set
of rules that (a) recognise types of phrase and (b) recognise and
penalise messages that contain combinations of them.

For instance, I get fairly high reliability by writing low scoring
(0.001) rules to recognise selling phrases and product names and then
using high scoring (5.0 or more) meta rules that combine them to
recognise sales spam. Here's a set to give you the idea: 

body  SALES_TERM (low cost|available now|.....)
score SALES_TERM 0.001

body  PRODUCT    (Nike|Dior|Tommy Hilfiger|......)
score PRODUCT    0.001

meta  SALES_SPAM (SALES_TERM && PRODUCT)
score SALES_SPAM 5.0

This approach ensures that other messages can contain phrases that are
used for selling or product names without being treated as spam, but any
message that contains at least one of each will be dinged as spam.

HTH

Martin





Re: increase in spams getting though

Posted by tony <ne...@gmail.com>.
now I've given it some more thought it's a certain type of spam that gets
through. I never get scams or marital aid spam, never. The spam mails that
get through are consistently to do with hospitality, increased sales and
business courses. Are there rules I should be using to combat this type of
spam?



--
View this message in context: http://spamassassin.1065346.n5.nabble.com/increase-in-spams-getting-though-tp104964p104997.html
Sent from the SpamAssassin - Users mailing list archive at Nabble.com.

Re: increase in spams getting though

Posted by Martin Gregorie <ma...@gregorie.org>.
On Fri, 2013-05-31 at 03:52 -0700, tony wrote:
> I've been running spamassassin for about 4 years on a centos 5 server,
> currently running spamassassin-3.3.1-2.el5. I've always been impressed and
> very pleased with the way SA works and I regularly run sa-update and I
> update the bayes databases. I use plugins too - raor, pyzor, dkim etc but
> over the last couple of months I've seen SA's effectiveness wane, with more
> spam emails slipping though into my inbox.
>
For the last few months almost all my spam stream has been Stock
pump'n'dump scams (no particular TLD associated with it) and a small
admixture of marital aid spam (almost all from Russia). 

I've just done a quick scan: both types cause locally written rules to
fire, but almost all the pump'n'dump spam would be caught by standard SA
rules and/or URIBLs. I'm uncertain whether this also applies to the
marital aid spam: it triggers my "CYRILLIC" rule, which simply catches
anything with a .ru or .su TLD unless its from one of my three Russian
correspondents.

So, no, I don't agree with you: SA is working just fine here. 

My local rule collection is largely stable, is changing very slowly and
is averaging less than one misclassified ham per week - and I can easily
fish that out of my quarantine area. Currently my spam load is around
15-20% of incoming mail, which is relatively low volume as I'm a private
site.

HTH

Martin





Re: increase in spams getting though

Posted by Martin Hepworth <ma...@gmail.com>.
I'd check what rules are firing for the spam that gets through

Could be youve got a whitelist entry being abused or even something
internally generating the crud

Martin

On Friday, 31 May 2013, tony wrote:

> I've been running spamassassin for about 4 years on a centos 5 server,
> currently running spamassassin-3.3.1-2.el5. I've always been impressed and
> very pleased with the way SA works and I regularly run sa-update and I
> update the bayes databases. I use plugins too - raor, pyzor, dkim etc but
> over the last couple of months I've seen SA's effectiveness wane, with more
> spam emails slipping though into my inbox.
> There are two scenarios here, either my system is not working as it should,
> or spammers have found a new clever way to beat the filters and SA is yet
> to
> cactch up. So my first thought in posting here is, has anyone else
> experienced an increase in penetration by spammers in the last couple of
> months?
>
>
>
> --
> View this message in context:
> http://spamassassin.1065346.n5.nabble.com/increase-in-spams-getting-though-tp104964.html
> Sent from the SpamAssassin - Users mailing list archive at Nabble.com.
>


-- 
-- 
Martin Hepworth, CISSP
Oxford, UK

Re: increase in spams getting though

Posted by Matus UHLAR - fantomas <uh...@fantomas.sk>.
On 31.05.13 03:52, tony wrote:
>I've been running spamassassin for about 4 years on a centos 5 server,
>currently running spamassassin-3.3.1-2.el5. I've always been impressed and
>very pleased with the way SA works and I regularly run sa-update and I
>update the bayes databases. I use plugins too - raor, pyzor, dkim etc but
>over the last couple of months I've seen SA's effectiveness wane, with more
>spam emails slipping though into my inbox.

What about SPF, DCC plugins? Also, do you use BAYES?
Do you have RBLs allowed? Haven't they blocked you due to excessive queries?
(how many mail do your servers process daily?)

>There are two scenarios here, either my system is not working as it should,
>or spammers have found a new clever way to beat the filters and SA is yet to
>cactch up. So my first thought in posting here is, has anyone else
>experienced an increase in penetration by spammers in the last couple of
>months?

I could say so, but that's apparently mostly due to rfc-ignorant.org
deactivation. 

Yes, spammers try to avoid those techniques. Do you update your rules on
regular basis?

-- 
Matus UHLAR - fantomas, uhlar@fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
I wonder how much deeper the ocean would be without sponges.