You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by James <bj...@lockie.ca> on 2015/07/11 04:00:10 UTC
non-English sender and body
I get a lot of spam from Chinese senders and Chinese subjects but only
an image for the body.
I want to mark as spam any non-English sender names and subjects.
I tried TextCat but either I did it wrong or it only looks at the Body.
Re: non-English sender and body
Posted by "chino@antennex.com" <ch...@antennex.com>.
From: James
Date: 2015-07-10 21:00
To: users
Subject: non-English sender and body
I get a lot of spam from Chinese senders and Chinese subjects but only
an image for the body.
I want to mark as spam any non-English sender names and subjects.
I tried TextCat but either I did it wrong or it only looks at the Body.
-------------------------------------------------------
James, I have been using "milter-regex" for years and one of its knobs will refuse non-english characters (and mnuch, much more) and those emails won't even reach as far as spamassassin. On my system, it's the first filter right after Spamhaus and other reject stuff. Low overhead.
It's in the ports....
Warmest regards,
Mark Chino
--
Re: non-English sender and body
Posted by Mauricio Tavares <ra...@gmail.com>.
On Jul 12, 2015 5:32 PM, "James" <bj...@lockie.ca> wrote:
>
>
>
> On 07/12/15 00:22, chino@antennex.com wrote:
>>>
>>>
>>>
>>> Why not run:
>>> sa-learn --dump magic
>>>
>>>
>>> And you'll see the number of spams vs tyhe number of hams learned?
>
> The problem is finding out which directory the running spamassassin uses,
I can't seen to train the one it expects.
>
Just as a FYI, some distros tend to package programs with
non-standard file paths. Also, if you run it command line in debug mode,
spamassassin -D -t < spam2.eml
it should tell you the config files it used.
> I put this in my /etc/spamassassin/local.cf:
> bayes_path /var/spamassassin/bayes_db/bayes
> bayes_file_mode 0777
> and it seems to pick up the right db.
>
> $ sudo sa-learn --dump magic
> 0.000 0 3 0 non-token data: bayes db version
> 0.000 0 6794 0 non-token data: nspam
> 0.000 0 3115 0 non-token data: nham
> 0.000 0 142011 0 non-token data: ntokens
> 0.000 0 1414709537 0 non-token data: oldest atime
> 0.000 0 1436703769 0 non-token data: newest atime
> 0.000 0 0 0 non-token data: last journal sync
atime
> 0.000 0 1436684568 0 non-token data: last expiry atime
> 0.000 0 22118400 0 non-token data: last expire atime
delta
> 0.000 0 206830 0 non-token data: last expire
reduction count
>
> $ sudo sa-learn --dump magic --db /var/spamassassin/bayes_db/
> 0.000 0 3 0 non-token data: bayes db version
> 0.000 0 6794 0 non-token data: nspam
> 0.000 0 3115 0 non-token data: nham
> 0.000 0 142011 0 non-token data: ntokens
> 0.000 0 1414709537 0 non-token data: oldest atime
> 0.000 0 1436703769 0 non-token data: newest atime
> 0.000 0 0 0 non-token data: last journal sync
atime
> 0.000 0 1436684568 0 non-token data: last expiry atime
> 0.000 0 22118400 0 non-token data: last expire atime
delta
> 0.000 0 206830 0 non-token data: last expire
reduction count
>
> Time will tell if the Chinese picture spam is gone.
Good luck!
>
>>>
>>> I always heard it best to try and even out those two.
>>>
>>>
>>> At the moment, my new server's bayes is 8x ham to spam. Thus I need
more spam.
>>>
>>>
>>> /Jason
>
>
Re: non-English sender and body
Posted by Antony Stone <An...@spamassassin.open.source.it>.
On Sunday 12 July 2015 at 23:25:04 (EU time), James wrote:
> > Am 12.07.2015 um 21:40 schrieb Bill Cole:
> >> On 12 Jul 2015, at 11:28, James wrote:
> >>>
> >>> I put this in my /etc/spamassassin/local.cf:
> >>> bayes_path /var/spamassassin/bayes_db/bayes
> >>> bayes_file_mode 0777
> >>
> >> It's heartwarming (in a sense) to see that there are people so confident
> >> in the absolute security of their systems that they are happy to let any
> >> process on a server write to files that other processes are supposed to
> >> trust.
> >>
> >> For folks who don't actually have that sort of confidence, it is
> >> generally a better approach to figure out the set of users who actually
> >> require read and/or write access to the Bayes DB and use ownership,
> >> groups, ACLs, su, sudo or whatever you might need to get rid of that
> >> third 7
> I saw the instructions on
> http://wiki.apache.org/spamassassin/SiteWideBayesSetup
"many use world RWX to simplify this, but this is insecure and not
recommended"
You think maybe that should be put in bold so people might then read it?
Antony.
--
Tinned food was developed for the British Navy in 1813.
The tin opener was not invented until 1858.
Please reply to the list;
please *don't* CC me.
Re: non-English sender and body
Posted by James <bj...@lockie.ca>.
On 07/12/15 17:15, Reindl Harald wrote:
>
> Am 12.07.2015 um 21:40 schrieb Bill Cole:
>> On 12 Jul 2015, at 11:28, James wrote:
>>
>>> The problem is finding out which directory the running spamassassin
>>> uses, I can't seen to train the one it expects.
>>>
>>> I put this in my /etc/spamassassin/local.cf:
>>> bayes_path /var/spamassassin/bayes_db/bayes
>>> bayes_file_mode 0777
>>
>> It's heartwarming (in a sense) to see that there are people so confident
>> in the absolute security of their systems that they are happy to let any
>> process on a server write to files that other processes are supposed to
>> trust.
>>
>> For folks who don't actually have that sort of confidence, it is
>> generally a better approach to figure out the set of users who actually
>> require read and/or write access to the Bayes DB and use ownership,
>> groups, ACLs, su, sudo or whatever you might need to get rid of that
>> third 7
>
> well, when someone switches from start "sa-learn" as root to "suodo
> sa-learn" after multiple "don't do that as root" it makes no difference...
Actually I never noticed the wide open permissions.
I saw the instructions on
http://wiki.apache.org/spamassassin/SiteWideBayesSetup
Re: non-English sender and body
Posted by Reindl Harald <h....@thelounge.net>.
Am 12.07.2015 um 21:40 schrieb Bill Cole:
> On 12 Jul 2015, at 11:28, James wrote:
>
>> The problem is finding out which directory the running spamassassin
>> uses, I can't seen to train the one it expects.
>>
>> I put this in my /etc/spamassassin/local.cf:
>> bayes_path /var/spamassassin/bayes_db/bayes
>> bayes_file_mode 0777
>
> It's heartwarming (in a sense) to see that there are people so confident
> in the absolute security of their systems that they are happy to let any
> process on a server write to files that other processes are supposed to
> trust.
>
> For folks who don't actually have that sort of confidence, it is
> generally a better approach to figure out the set of users who actually
> require read and/or write access to the Bayes DB and use ownership,
> groups, ACLs, su, sudo or whatever you might need to get rid of that
> third 7
well, when someone switches from start "sa-learn" as root to "suodo
sa-learn" after multiple "don't do that as root" it makes no difference...
Re: non-English sender and body
Posted by Bill Cole <sa...@billmail.scconsult.com>.
On 12 Jul 2015, at 11:28, James wrote:
> The problem is finding out which directory the running spamassassin
> uses, I can't seen to train the one it expects.
>
> I put this in my /etc/spamassassin/local.cf:
> bayes_path /var/spamassassin/bayes_db/bayes
> bayes_file_mode 0777
It's heartwarming (in a sense) to see that there are people so confident
in the absolute security of their systems that they are happy to let any
process on a server write to files that other processes are supposed to
trust.
For folks who don't actually have that sort of confidence, it is
generally a better approach to figure out the set of users who actually
require read and/or write access to the Bayes DB and use ownership,
groups, ACLs, su, sudo or whatever you might need to get rid of that
third 7.
Re: non-English sender and body
Posted by James <bj...@lockie.ca>.
On 07/12/15 00:22, chino@antennex.com wrote:
>
>
> Why not run:
> sa-learn --dump magic
>
>
> And you'll see the number of spams vs tyhe number of hams learned?
>
The problem is finding out which directory the running spamassassin
uses, I can't seen to train the one it expects.
I put this in my /etc/spamassassin/local.cf:
bayes_path /var/spamassassin/bayes_db/bayes
bayes_file_mode 0777
and it seems to pick up the right db.
$ sudo sa-learn --dump magic
0.000 0 3 0 non-token data: bayes db version
0.000 0 6794 0 non-token data: nspam
0.000 0 3115 0 non-token data: nham
0.000 0 142011 0 non-token data: ntokens
0.000 0 1414709537 0 non-token data: oldest atime
0.000 0 1436703769 0 non-token data: newest atime
0.000 0 0 0 non-token data: last journal
sync atime
0.000 0 1436684568 0 non-token data: last expiry atime
0.000 0 22118400 0 non-token data: last expire
atime delta
0.000 0 206830 0 non-token data: last expire
reduction count
$ sudo sa-learn --dump magic --db /var/spamassassin/bayes_db/
0.000 0 3 0 non-token data: bayes db version
0.000 0 6794 0 non-token data: nspam
0.000 0 3115 0 non-token data: nham
0.000 0 142011 0 non-token data: ntokens
0.000 0 1414709537 0 non-token data: oldest atime
0.000 0 1436703769 0 non-token data: newest atime
0.000 0 0 0 non-token data: last journal
sync atime
0.000 0 1436684568 0 non-token data: last expiry atime
0.000 0 22118400 0 non-token data: last expire
atime delta
0.000 0 206830 0 non-token data: last expire
reduction count
Time will tell if the Chinese picture spam is gone.
>
>
> I always heard it best to try and even out those two.
>
>
> At the moment, my new server's bayes is 8x ham to spam. Thus I
> need more spam.
>
>
> /Jason
>
Re: Re: non-English sender and body
Posted by "chino@antennex.com" <ch...@antennex.com>.
From: James
Date: 2015-07-11 10:41
To: users
Subject: Re: non-English sender and body
On 07/11/15 09:49, chino@antennex.com wrote:
From: chino@antennex.com
Date: 2015-07-11 08:32
To: RW; USERS-SPAMASSASSIN
Subject: Re: Re: non-English sender and body
From: RW
Date: 2015-07-11 08:28
To: users
Subject: Re: non-English sender and body
On Fri, 10 Jul 2015 22:00:10 -0400
James wrote:
> I get a lot of spam from Chinese senders and Chinese subjects but
> only an image for the body.
> I want to mark as spam any non-English sender names and subjects.
> I tried TextCat but either I did it wrong or it only looks at the
> Body.
Does Bayes catch them?
=======================================================
I don't know since I don't get any non-english emails. But, you could collect and feed them to "sa-learn spam" and see it that helps.
Warmest regards,
Mark Chino
===============================================================
Ouch! One non-english sneaked through, but still tagged as spam. Here's one of several rules fired and notice it looked at the "subject."
| 4.0 NOT_IN_ENGLISH Subject Contains Non English Characters
/Jason
That is the rule I want.
--------------------------------------------------------------
Why not run:
sa-learn --dump magic
And you'll see the number of spams vs tyhe number of hams learned?
I always heard it best to try and even out those two.
At the moment, my new server's bayes is 8x ham to spam. Thus I need more spam.
/Jason
Re: non-English sender and body
Posted by James <bj...@lockie.ca>.
On 07/11/15 09:49, chino@antennex.com wrote:
>
> *From:* chino@antennex.com <ma...@antennex.com>
> *Date:* 2015-07-11 08:32
> *To:* RW <ma...@googlemail.com>; USERS-SPAMASSASSIN
> <ma...@spamassassin.apache.org>
> *Subject:* Re: Re: non-English sender and body
>
> *From:* RW <ma...@googlemail.com>
> *Date:* 2015-07-11 08:28
> *To:* users <ma...@spamassassin.apache.org>
> *Subject:* Re: non-English sender and body
> On Fri, 10 Jul 2015 22:00:10 -0400
> James wrote:
> > I get a lot of spam from Chinese senders and Chinese
> subjects but
> > only an image for the body.
> > I want to mark as spam any non-English sender names and
> subjects.
> > I tried TextCat but either I did it wrong or it only looks
> at the
> > Body.
> Does Bayes catch them?
>
> =======================================================
>
> I don't know since I don't get any non-english emails. But,
> you could collect and feed them to "sa-learn spam" and see it
> that helps.
>
> Warmest regards,
> Mark Chino
> ===============================================================
>
> Ouch! One non-english sneaked through, but still tagged as
> spam. Here's one of several rules fired and notice it looked
> at the "subject."
>
> | 4.0 NOT_IN_ENGLISH Subject Contains Non English Characters
>
> /Jason
>
That is the rule I want.
Re: Re: non-English sender and body
Posted by "chino@antennex.com" <ch...@antennex.com>.
From: chino@antennex.com
Date: 2015-07-11 08:32
To: RW; USERS-SPAMASSASSIN
Subject: Re: Re: non-English sender and body
From: RW
Date: 2015-07-11 08:28
To: users
Subject: Re: non-English sender and body
On Fri, 10 Jul 2015 22:00:10 -0400
James wrote:
> I get a lot of spam from Chinese senders and Chinese subjects but
> only an image for the body.
> I want to mark as spam any non-English sender names and subjects.
> I tried TextCat but either I did it wrong or it only looks at the
> Body.
Does Bayes catch them?
=======================================================
I don't know since I don't get any non-english emails. But, you could collect and feed them to "sa-learn spam" and see it that helps.
Warmest regards,
Mark Chino
===============================================================
Ouch! One non-english sneaked through, but still tagged as spam. Here's one of several rules fired and notice it looked at the "subject."
| 4.0 NOT_IN_ENGLISH Subject Contains Non English Characters
/Jason
Re: non-English sender and body
Posted by Antony Stone <An...@spamassassin.open.source.it>.
On Sunday 12 July 2015 at 11:07:30 (EU time), Reindl Harald wrote:
> Am 12.07.2015 um 10:59 schrieb Antony Stone:
> > On Sunday 12 July 2015 at 10:48:28 (EU time), Reindl Harald wrote:
> >> the root of all evil is when people start calling tools as root
> >
> > Since this appears to be such a common problem, would it perhaps be a
> > good idea to include a warning in sa-learn if it is run as UID=0, along
> > the lines of "You are running this command as root, which is almost
> > certainly NOT what you should be doing. This command should be run by
> > the same user ID as spamassassin itself runs under."
>
> frankly it should refuse to start at all not only warn
That's certainly an alternative, however it doesn't really conform to the Unix
approach of "do what the user requested, even if it's a bit stupid".
Maybe a good compromise would be to emit the above warning message and then
stop, which behaviour can be overridden with a "--root-anyway" or similar
option (I don't know when that could genuinely be useful, but I really don't
like the idea of a command which will not do what a user requests, even when
it's what the user really wants).
> on the other hand if you start a cli tool which analyzes bad mail
> including malware by definition with root permissions you should
> consider refrain from maintain a public mailserver and that's not
> disrespectfully
Ah, but the very word "consider" in your comment above means the person needs
to be aware that this is something to think about. If the tools do not warn
the user when they're being run under the wrong UID, how does the novice email
admin learn what's right and what's wrong?
- please don't answer along the lines of "nobody should be allowed to run an
email server until they've had at least 5 years experience of doing so" :)
Regards,
Antony.
--
"A person lives in the UK, but commutes to France daily for work.
He belongs in the UK."
- From UK Revenue & Customs notice 741, page 13, paragraph 3.5.1
- http://tinyurl.com/o7gnm4
Please reply to the list;
please *don't* CC me.
Re: non-English sender and body
Posted by Reindl Harald <h....@thelounge.net>.
Am 12.07.2015 um 10:59 schrieb Antony Stone:
> On Sunday 12 July 2015 at 10:48:28 (EU time), Reindl Harald wrote:
>
>> the root of all evil is when people start calling tools as root
>
> Since this appears to be such a common problem, would it perhaps be a good
> idea to include a warning in sa-learn if it is run as UID=0, along the lines
> of "You are running this command as root, which is almost certainly NOT what
> you should be doing. This command should be run by the same user ID as
> spamassassin itself runs under."
frankly it should refuse to start at all not only warn
on the other hand if you start a cli tool which analyzes bad mail
including malware by definition with root permissions you should
consider refrain from maintain a public mailserver and that's not
disrespectfully
don't run *any* tool as root if not needed, most of my shell helpers
have explicit "su -c 'command' roleaccount" so that they can be started
as root, ensure pemissions of files and folders but give away
permissions for anything which don't need them
Re: non-English sender and body
Posted by Reindl Harald <h....@thelounge.net>.
Am 12.07.2015 um 11:10 schrieb Axb:
> On 12.07.2015 10:59, Antony Stone wrote:
>> On Sunday 12 July 2015 at 10:48:28 (EU time), Reindl Harald wrote:
>>
>>> the root of all evil is when people start calling tools as root
>>
>> Since this appears to be such a common problem, would it perhaps be a
>> good
>> idea to include a warning in sa-learn if it is run as UID=0, along the
>> lines
>> of "You are running this command as root, which is almost certainly
>> NOT what
>> you should be doing. This command should be run by the same user ID as
>> spamassassin itself runs under."
>
> if you have the bayes path defined in local.cf you can run sa-learn as
> root. same applies if you use SQL or Redis.
and how does that help new users with default settings coming here with
always the same problems - besides that there is no single reason to run
sa-learn with root permissions and with saying "you can run sa-learn as
root" you make a big mistake and forget resposibility
there are many tools out there which refuse to start as root at all or
until a command-line switch is given that you know what you are doing
Re: non-English sender and body
Posted by Axb <ax...@gmail.com>.
On 12.07.2015 10:59, Antony Stone wrote:
> On Sunday 12 July 2015 at 10:48:28 (EU time), Reindl Harald wrote:
>
>> the root of all evil is when people start calling tools as root
>
> Since this appears to be such a common problem, would it perhaps be a good
> idea to include a warning in sa-learn if it is run as UID=0, along the lines
> of "You are running this command as root, which is almost certainly NOT what
> you should be doing. This command should be run by the same user ID as
> spamassassin itself runs under."
>
if you have the bayes path defined in local.cf you can run sa-learn as root.
same applies if you use SQL or Redis.
Re: non-English sender and body
Posted by Antony Stone <An...@spamassassin.open.source.it>.
On Sunday 12 July 2015 at 10:48:28 (EU time), Reindl Harald wrote:
> the root of all evil is when people start calling tools as root
Since this appears to be such a common problem, would it perhaps be a good
idea to include a warning in sa-learn if it is run as UID=0, along the lines
of "You are running this command as root, which is almost certainly NOT what
you should be doing. This command should be run by the same user ID as
spamassassin itself runs under."
?
Regards,
Antony.
--
There's no such thing as bad weather - only the wrong clothes.
- Billy Connolly
Please reply to the list;
please *don't* CC me.
Re: non-English sender and body
Posted by Reindl Harald <h....@thelounge.net>.
Am 12.07.2015 um 04:59 schrieb James:
> On 07/11/15 19:55, Reindl Harald wrote:
>> Am 11.07.2015 um 17:36 schrieb James:
>>>> Does Bayes catch them?
>>>>
>>>> =======================================================
>>>>
>>>> I don't know since I don't get any non-english emails. But, you
>>>> could collect and feed them to "sa-learn spam" and see it that
>>>> helps.
>>>>
>>>> Warmest regards,
>>>> Mark Chino
>>>>
>>> It doesn't help (I think I feed sa-learn to much spam).
>>> They are always close.
>>> score=3.9 required=4.5
>>
>> and where is the problem to rasie the bayesc scores in loal.cf if you
>> *really* trained it?
>
> Mmm.
>
> X-Spam-Status: No, score=3.9 required=4.5 tests=HTML_MESSAGE,
> HTML_OBFUSCATE_05_10,RCVD_IN_PBL,RDNS_DYNAMIC autolearn=no
> version=3.3.2
>
> Looks like my bayes is not working again.
> I had a lot of problems getting it right
then you found the real problem, any spamfilter without bayes has a real
problem but i don't understand "lot of problems getting it right"
* you need some hundret ham *and* spam samples
* always call "sa-learn" as exactly the user SA runs later
* two cli calls to each sample folder, done
the root of all evil is when people start calling tools as root
Re: non-English sender and body
Posted by James <bj...@lockie.ca>.
On 07/11/15 19:55, Reindl Harald wrote:
>
>
> Am 11.07.2015 um 17:36 schrieb James:
>>> Does Bayes catch them?
>>>
>>> =======================================================
>>>
>>> I don't know since I don't get any non-english emails. But, you
>>> could collect and feed them to "sa-learn spam" and see it that
>>> helps.
>>>
>>> Warmest regards,
>>> Mark Chino
>>>
>> It doesn't help (I think I feed sa-learn to much spam).
>> They are always close.
>> score=3.9 required=4.5
>
> and where is the problem to rasie the bayesc scores in loal.cf if you
> *really* trained it?
Mmm.
X-Spam-Status: No, score=3.9 required=4.5 tests=HTML_MESSAGE,
HTML_OBFUSCATE_05_10,RCVD_IN_PBL,RDNS_DYNAMIC autolearn=no version=3.3.2
Looks like my bayes is not working again.
I had a lot of problems getting it right.
Re: non-English sender and body
Posted by Mauricio Tavares <ra...@gmail.com>.
On Sat, Jul 11, 2015 at 7:55 PM, Reindl Harald <h....@thelounge.net> wrote:
>
>
> Am 11.07.2015 um 17:36 schrieb James:
>>>
>>> Does Bayes catch them?
>>>
>>> =======================================================
>>>
>>> I don't know since I don't get any non-english emails. But, you
>>> could collect and feed them to "sa-learn spam" and see it that helps.
>>>
>>> Warmest regards,
>>> Mark Chino
>>>
>> It doesn't help (I think I feed sa-learn to much spam).
>> They are always close.
>> score=3.9 required=4.5
>
>
> and where is the problem to rasie the bayesc scores in loal.cf if you
> *really* trained it?
>
> [root@mail-gw:~]$ cat spamd-local.conf | grep BAYES
> score BAYES_00 -3.5
> score BAYES_05 -2.0
> score BAYES_20 -1.0
> score BAYES_40 -0.5
> score BAYES_50 1.8
> score BAYES_60 3.5
> score BAYES_80 5.0
> score BAYES_95 6.5
> score BAYES_99 7.5
> score BAYES_999 0.4
>
> milter reject starts with 8.0 with no single complaint the last 12 months
>
What if the OP change the weight/score values for textcat?
Re: non-English sender and body
Posted by Reindl Harald <h....@thelounge.net>.
Am 11.07.2015 um 17:36 schrieb James:
>> Does Bayes catch them?
>>
>> =======================================================
>>
>> I don't know since I don't get any non-english emails. But, you
>> could collect and feed them to "sa-learn spam" and see it that helps.
>>
>> Warmest regards,
>> Mark Chino
>>
> It doesn't help (I think I feed sa-learn to much spam).
> They are always close.
> score=3.9 required=4.5
and where is the problem to rasie the bayesc scores in loal.cf if you
*really* trained it?
[root@mail-gw:~]$ cat spamd-local.conf | grep BAYES
score BAYES_00 -3.5
score BAYES_05 -2.0
score BAYES_20 -1.0
score BAYES_40 -0.5
score BAYES_50 1.8
score BAYES_60 3.5
score BAYES_80 5.0
score BAYES_95 6.5
score BAYES_99 7.5
score BAYES_999 0.4
milter reject starts with 8.0 with no single complaint the last 12 months
Re: non-English sender and body
Posted by James <bj...@lockie.ca>.
On 07/11/15 09:32, chino@antennex.com wrote:
>
> *From:* RW <ma...@googlemail.com>
> *Date:* 2015-07-11 08:28
> *To:* users <ma...@spamassassin.apache.org>
> *Subject:* Re: non-English sender and body
> On Fri, 10 Jul 2015 22:00:10 -0400
> James wrote:
> > I get a lot of spam from Chinese senders and Chinese subjects but
> > only an image for the body.
> > I want to mark as spam any non-English sender names and subjects.
> > I tried TextCat but either I did it wrong or it only looks at the
> > Body.
> Does Bayes catch them?
>
> =======================================================
>
> I don't know since I don't get any non-english emails. But, you
> could collect and feed them to "sa-learn spam" and see it that helps.
>
> Warmest regards,
> Mark Chino
>
It doesn't help (I think I feed sa-learn to much spam).
They are always close.
score=3.9 required=4.5
Re: non-English sender and body
Posted by James <bj...@lockie.ca>.
On 07/11/15 09:59, chino@antennex.com wrote:
>
>
> ------------------------------------------------------------------------
> Warmest regards,
> Mark Chino
> --
> chino@antennex.com <ma...@antennex.com>
> www.antennex.com
>
> *From:* RW <ma...@googlemail.com>
> *Date:* 2015-07-11 08:48
> *To:* users <ma...@spamassassin.apache.org>
> *Subject:* Re: non-English sender and body
> On Sat, 11 Jul 2015 08:32:44 -0500
> chino@antennex.com wrote:
> >
> > From: RW
> > Date: 2015-07-11 08:28
> > To: users
> > Subject: Re: non-English sender and body
> > On Fri, 10 Jul 2015 22:00:10 -0400
> > James wrote:
> >
> > > I get a lot of spam from Chinese senders and Chinese subjects but
> > > only an image for the body.
> > > I want to mark as spam any non-English sender names and subjects.
> > > I tried TextCat but either I did it wrong or it only looks at the
> > > Body.
> >
> > Does Bayes catch them?
> >
> > =======================================================
> >
> > I don't know since I don't get any non-english emails.
> That's why I was addressed the question to the OP.
> BTW I forgot to mention TextCat needs a minimum amount of text to work
> with.
> =================================================
>
> Have you googled? Here's one that may help:
> http://email.about.com/cs/spamassassintips/qt/et032504.htm
>
> /Jason
>
I looked at that and it doesn't seem to work on non-text bodies (I get
pictures of Chinese writing).
That is why I asked if something looks at the language of the
subject/sender.
Re: Re: non-English sender and body
Posted by "chino@antennex.com" <ch...@antennex.com>.
Warmest regards,
Mark Chino
--
chino@antennex.com
www.antennex.com
From: RW
Date: 2015-07-11 08:48
To: users
Subject: Re: non-English sender and body
On Sat, 11 Jul 2015 08:32:44 -0500
chino@antennex.com wrote:
>
> From: RW
> Date: 2015-07-11 08:28
> To: users
> Subject: Re: non-English sender and body
> On Fri, 10 Jul 2015 22:00:10 -0400
> James wrote:
>
> > I get a lot of spam from Chinese senders and Chinese subjects but
> > only an image for the body.
> > I want to mark as spam any non-English sender names and subjects.
> > I tried TextCat but either I did it wrong or it only looks at the
> > Body.
>
> Does Bayes catch them?
>
> =======================================================
>
> I don't know since I don't get any non-english emails.
That's why I was addressed the question to the OP.
BTW I forgot to mention TextCat needs a minimum amount of text to work
with.
=================================================
Have you googled? Here's one that may help:
http://email.about.com/cs/spamassassintips/qt/et032504.htm
/Jason
Re: non-English sender and body
Posted by RW <rw...@googlemail.com>.
On Sat, 11 Jul 2015 08:32:44 -0500
chino@antennex.com wrote:
>
> From: RW
> Date: 2015-07-11 08:28
> To: users
> Subject: Re: non-English sender and body
> On Fri, 10 Jul 2015 22:00:10 -0400
> James wrote:
>
> > I get a lot of spam from Chinese senders and Chinese subjects but
> > only an image for the body.
> > I want to mark as spam any non-English sender names and subjects.
> > I tried TextCat but either I did it wrong or it only looks at the
> > Body.
>
> Does Bayes catch them?
>
> =======================================================
>
> I don't know since I don't get any non-english emails.
That's why I was addressed the question to the OP.
BTW I forgot to mention TextCat needs a minimum amount of text to work
with.
Re: Re: non-English sender and body
Posted by "chino@antennex.com" <ch...@antennex.com>.
From: RW
Date: 2015-07-11 08:28
To: users
Subject: Re: non-English sender and body
On Fri, 10 Jul 2015 22:00:10 -0400
James wrote:
> I get a lot of spam from Chinese senders and Chinese subjects but
> only an image for the body.
> I want to mark as spam any non-English sender names and subjects.
> I tried TextCat but either I did it wrong or it only looks at the
> Body.
Does Bayes catch them?
=======================================================
I don't know since I don't get any non-english emails. But, you could collect and feed them to "sa-learn spam" and see it that helps.
Warmest regards,
Mark Chino
Re: non-English sender and body
Posted by RW <rw...@googlemail.com>.
On Fri, 10 Jul 2015 22:00:10 -0400
James wrote:
> I get a lot of spam from Chinese senders and Chinese subjects but
> only an image for the body.
> I want to mark as spam any non-English sender names and subjects.
> I tried TextCat but either I did it wrong or it only looks at the
> Body.
Does Bayes catch them?