DSPAM-3.9.0 Released

interfasys

Verified User
Joined
Oct 31, 2003
Messages
2,100
Location
Switzerland
The DSPAM team is happy to announce our first official release, right on
our first celebration of the community fork!

The source tarball can be downloaded from the downloads section at
SourceForge [1]. A gpg signature by Alexander Prinsier is available in
the downloads section as well.

If you are running any of the older RC, BETA or ALPHA releases then you
can just upgrade to this release. There are no compatibility issues
compared to older ALPHA or older BETA or RC releases. Users using DSPAM
3.8.0 or older can upgrade to DSPAM 3.9.0 by following the steps
mentioned in the UPGRADING file.

A detailed changelog is available inside the tarball.

Most changes are bug fixes, stability enhancements or enhancements to
the documentation.

Other new features in this release are the various new translated
templates for the Web UI.

Our mailing lists and the bug tracker are happy to help you out if you
run into some problems.


The DSPAM Team

[1] http://sourceforge.net/projects/dspam/files/dspam/dspam-3.9.0/
 
can you explain more about this program. what is the main difference between it and SpamAssassin

tnx
 
From the website: http://dspam.nuclearelephant.com/

DSPAM is a scalable and open-source content-based spam filter designed for multi-user enterprise systems. On a properly configured system, many users experience results between 99.5% - 99.95%, or one error for every 200 to 2000 messages. DSPAM supports many different MTAs and can also be deployed as a stand-alone SMTP appliance. For developers, the DSPAM core engine (libdspam) can be easily incorporated directly into applications for drop-in filtering

DSPAM has been implemented on many large and small scale systems with the largest being reported at about 350,000 mailboxes.

DSPAM is an adaptive filter which means it is capable of learning and adapting to each user's email. Instead of working off of a list of "rules" to identify spam, DSPAM's probabilistic engine examines the content of each message and learns what type of content the user deems as spam (or nonspam). This approach to machine-learning provides much higher levels of accuracy than commercial "hodge-podge" solutions, and with minimal resources. DSPAM's best recorded levels of accuracy have included 99.991% by one avid user (2 errors in 22,786) and 99.987% by the author (1 error in 7000), which could be ten times more accurate than a human being!

It uses less resources that SA and allows users to manage theirs spams and hams directly from the webmail. Works well with Exim and Dovecot.

There is also a web interface provided if you don't want to integrate it into your webmail.
 
I wouldnt mind using it if it works better then spamass. I will wait for jlasman to incorporate it into his configs.
 
Integration depends more on JBMC Software. There is a feature request in this forum, for them to make the antispam page more flexible, allowing us to use whichever solution we want.
The next step would be the Exim mods, followed by the webmail (Horde works well) one. The dovecot mod is optional.
 
I wouldnt mind using it if it works better then spamass. I will wait for jlasman to incorporate it into his configs.
Thanks for the vote of confidence. As interfasys points out, it's a multi-step operation.

If DirectAdmin wants to install their part of the equation, then I suppose it wouild be easy enough to offer a patch to the exim.conf file.

But they'd have to do it first.

Jeff
 
Unfortunately I don't have time to do much rewriting on SpamBlocker right now, except to fix any bugs found. Anyone care to write the exim.conf side of the equation?

Jeff
 
Hello,

We've currently got other projects on the go, and SpamAssassin seems to work ok for most cases, so we won't be implementing this right now. There I don't see any reason why you couldn't swap out SA for DS, other than the fact at the DA interface is setup purely for SA.

John
 
@Jeff - There is an example in the Exim wiki and I wrote a rough howto on here a few years back. We can use the same mechanism that is in place now to detect whether the feature has been enabled for a user.
That wouldn't be very useful though without proper support from DA or custombuild. Most people wanting to implement it are either going to pay for the service or will know what they are doing and in both cases, somebody will know how to properly configure exim to make it work.

@John - One thing maybe that DA could change is the reliance on the "spamassassin" name everywhere, including the config files by renaming all these instances to "antispam" ?
 
@Jeff - There is an example in the Exim wiki and I wrote a rough howto on here a few years back. We can use the same mechanism that is in place now to detect whether the feature has been enabled for a user.
See post 11 to this thread.

:)

Jeff
 
Back
Top