Extension development
Mail filtering add on

SaneSecurity signatures are not for spam control.  They are additional antivirus/antimalware signatures that can utilized by ClamAV.

Spam control is is done using the built-in Spamhalter or a third party app like POPFile.  Both utilize bayesian filtering through which incoming mail passes.  Both add a header on which you can filter.  Both must be trained. 

I can't recommend one over the other because I've only used POPFile but my sense is that both are equally effective.  SpamHalter is built-in so is the logical one to utilize.  Support is available on this forum.

Edit:  Correction... Support is available on the Mercury Community Support forum

Also, There is a guide to SpamHalter available here:  http://www.vandenbogaerde.net/pegasusmail/pf_sh_index.html.  It is based on the Pegasus Mail helpfile but is applicable to Mercury in most respects.

<p>SaneSecurity signatures are not for spam control.  They are additional antivirus/antimalware signatures that can utilized by ClamAV.</p><p>Spam control is is done using the built-in Spamhalter or a third party app like POPFile.  Both utilize bayesian filtering through which incoming mail passes.  Both add a header on which you can filter.  Both must be trained.  </p><p>I can't recommend one over the other because I've only used POPFile but my sense is that both are equally effective.  SpamHalter is built-in so is the logical one to utilize.  Support is available on this forum.</p><p>Edit:  Correction... Support is available on the Mercury Community Support forum </p><p>Also, There is a guide to SpamHalter available here:  http://www.vandenbogaerde.net/pegasusmail/pf_sh_index.html.  It is based on the Pegasus Mail helpfile but is applicable to Mercury in most respects. </p>

Hi peeps,

 After a long time of using an Excuse of a mail server, I've reverted to Mercury, and it feels good to be back! Running 4.80 as a service (bought license)...

The only "problem" I have, like most people I suppose, is that of spam.  I've got Spamhalter ticked, configured MercuryS to use Spamhaus, abuseat, spamcop etc and also added transaction filters for certain words in the MercuryS exceptions.

It's probably me not setting things up correctly, but these don't seem to cut the mustard.

So... I've written a little program that does what I need and it is triggered as a Policy on each mail that comes in.

 We have a filter.rules file which specifies words to test each line for and the weight achieved when *all* are matched. For example, this is a simple rules file:

#weight, list of words separated by commas

100,subject:,flight,simulator

-1000,from:,@mydomain.com

 *All* of the words have to appear for the weighting to be applied, so if "flight" and "simulator" are in the body, and not the subject, this won't trigger. Similarly if it is sent from @mydomain.com the weighting is reduced so the rule doesn't trigger

This program allows other lines within the email text to be tested not just the ones provided for in the transaction filter.

It is used like this:

Create a new policy

 Type of task is "Run a program and examine the return code"

Command line is

mercuryfilter ~A TotalWeight AppendLog

  for example:

mercuryfilter ~A 100 true

This will  cause the program to trigger a policy failure if the total weighting of the all the rule triggers is equal to or greater than 100. It also creates a log file which can either be replaced by each new email or appended by each one. If you want to keep a log of all the emails passed through, set the AppendLog flag to true. There's a different log file for each day.

I also have step 4 set to forward the emails that fail to me, just so I can check... 

 If you want to have a try, it can be downloaded at http://www.ictis.net./downloads/MercuryFiltersInstaller.msi

 Use with caution - no warranty is provided or implied! :-) It seems to do what I need it to here...

 The program installs to the c:\mercury by default and needs DotNet 4.5.2 to be installed.

 Hopefully I've not reinvented the wheel, but I'd appreciate any comments - especially if it's useful to someone else!

 

Tym

 

<p>Hi peeps,</p><p> After a long time of using an Excuse of a mail server, I've reverted to Mercury, and it feels good to be back!<span style="font-size: 10pt;"> Running 4.80 as a service (bought license)...</span></p><p>The only "problem" I have, like most people I suppose, is that of spam.  I've got Spamhalter ticked, configured MercuryS to use Spamhaus, abuseat, spamcop etc and also added transaction filters for certain words in the MercuryS exceptions.</p><p><span style="font-size: 10pt;">It's probably me not setting things up correctly, but these don't seem to cut the mustard.</span></p><p>So... I've written a little program that does what I need and it is triggered as a Policy on each mail that comes in.</p><p> We have a filter.rules file which specifies words to test each line for and the weight achieved when *all* are matched. For example, this is a simple rules file:</p><p><span style="white-space:pre"> </span>#weight, list of words separated by commas</p><p><span style="white-space:pre"> </span>100,subject:,flight,simulator</p><p><span style="white-space:pre"> </span>-1000,from:,@mydomain.com</p><p> *All* of the words have to appear for the weighting to be applied, so if "<span style="font-size: 13.3333px;">flight" and "simulator" </span>are in the body, and not the subject, this won't trigger. Similarly if it is sent from @mydomain.com the weighting is reduced so the rule doesn't trigger</p><p>This program allows other lines within the email text to be tested not just the ones provided for in the transaction filter.</p><p><span style="font-size: 10pt;">It is used like this:</span></p><p><span style="white-space:pre"> </span>Create a new policy</p><p><span style="white-space:pre"> </span> Type of task is "Run a program and examine the return code"</p><p><span style="font-size: 10pt;"><span style="white-space:pre"> </span>Command line is</span></p><p><span style="white-space:pre"> </span>mercuryfilter ~A TotalWeight AppendLog</p><p> <span style="font-size: 10pt; white-space: pre;"> </span><span style="font-size: 10pt;">for example:</span></p><p><span style="white-space:pre"> </span>mercuryfilter ~A 100 true</p><p><span style="font-size: 10pt;">This will  cause the program to trigger a policy failure if the total weighting of the all the rule triggers is equal to or greater than 100. It also creates a log file which can either be replaced by each new email or appended by each one. If you want to keep a log of all the emails passed through, set the AppendLog flag to true. There's a different log file for each day.</span></p><p>I also have step 4 set to forward the emails that fail to me, just so I can check... </p><p> <span style="font-size: 10pt;">If you want to have a try, it can be downloaded at http://www.ictis.net./downloads/MercuryFiltersInstaller.msi</span></p><p> <span style="font-size: 10pt;">Use with caution - no warranty is provided or implied! :-) It seems to do what I need it to here...</span></p><p> <span style="font-size: 10pt;">The program installs to the c:\mercury by default and needs DotNet 4.5.2 to be installed.</span></p><p> Hopefully I've not reinvented the wheel, but I'd appreciate any comments - especially if it's useful to someone else!</p><p> </p><p>Tym</p><p> </p>

Interesting, and thanks for sharing it.

They always say that no two folks' spam is the same and I find that not much gets through my Mercury with a similar setup.  The main difference is that I use Clamwall with extra signatures from Sanesecurity.com, and that leaves little for a well-trained Spamhalter to do.

It's useful to have a log and the ability to forward spam mails as you do - just in case of false positives. 

<p>Interesting, and thanks for sharing it.</p><p>They always say that no two folks' spam is the same and I find that not much gets through my Mercury with a similar setup.  The main difference is that I use Clamwall with extra signatures from Sanesecurity.com, and that leaves little for a well-trained Spamhalter to do.</p><p>It's useful to have a log and the ability to forward spam mails as you do - just in case of false positives. </p>

I'm use Clamwall with extra SaneSecurity signatures as well in conjunction with POPFile instead of Spamhalter.

All spam tagged messages are filtered to a Spam mailbox for review.  The rare false positives get fowarded to the appropriate user from there.

<p>I'm use Clamwall with extra SaneSecurity signatures as well in conjunction with POPFile instead of Spamhalter.</p><p>All spam tagged messages are filtered to a Spam mailbox for review.  The rare false positives get fowarded to the appropriate user from there. </p>

I've just had a look at Sanesecurity.com and to be honest, it's confused me! 

I'd appreciate any guidance on how you set this up please as the spam email issue is rather annoying...

<p>I've just had a look at <span style="font-family: Tahoma, Arial, Helvetica; font-size: 12.096px;">Sanesecurity.com and to be honest, it's confused me! </span></p><p><span style="font-size: 10pt;">I'd appreciate any guidance on how you set this up please as the spam email issue is rather annoying...</span></p>
live preview
enter atleast 10 characters
WARNING: You mentioned %MENTIONS%, but they cannot see this message and will not be notified
Saving...
Saved
With selected deselect posts show selected posts
All posts under this topic will be deleted ?
Pending draft ... Click to resume editing
Discard draft