List of Spam Blocker features?

[juank]

Hello,
I tried to find any documentation about the Spam Blocker module in Untangle, but nothing so far.

This is my question: What's being used in the Spam Blocker module?
We've been using AMAVIS for a couple of years with a 99.99% (almost perfect) rate. We set the DROP rate to 6.5 with 150+ users and never had ANY complaint or need to recover from Quarantine (and we have some really picky users!).

Amavis uses SpamAssasin, razor2, pyzor, fuzzyOCR, uridnsbl, ImageCheck, ImageInfo, SARE rulesets, PDFassassin, etc.

Does anybody knows what modules are being used by the SpamBlocker and also, what are the rate numbers used by it when you set the threshold?

Thanks a lot.

[mdh]

juank,

That's all on the wiki (in the Developer Wiki section, under Upstream Projects). The link is;

http://wiki.untangle.com/index.php/Upstream_Projects

I don't know the answer to the second part of your question. Perhaps engineering will be able to comment when they have a chance.

[juank]

MDH,

Nope. Nothing of that related to my question is in the WIKI page. My question is specific about the SpamBlocker.

Thanks.

[mdh]

On the middle of that page is a reference to Spamassassin, used in SpamBlocker. I pointed you to that page rather than the one-line answer, so that if you had any questions about what was part of other Untangle modules, you would have that as a reference.

[sunflower]

I think this is what 'juank' is looking for. It's under technical details of the SpamBlocker. I don't know what is the default score set on the spamassassin configuration but you can log into the box and look around.

Under the Hood
Spam Blocker implements a rule-based architecture that leverages the open-source project SpamAssasin. No one rule can flag email as spam. It's the collection of rules that identifies email as spam. Each rule outputs a confidence level. The combination of each rule's confidence level produces a spam score. This score identifies the likelihood that the email is spam. The user-defined threshold you use to customize Spam Blocker's scanning sensitivity is based on this score.
What It Does

Transparently scans SMTP, POP and IMAP traffic and blocks, quarantines or marks spam
How It Does It

Uses SpamAssassin, Razor, Bayesian filters, DNSBL/RBLs, optical character recognition (OCR), tar pitting, custom tuning and updates

[mdh]

sunflower,

That is a much more detailed description. Thank you for helping out!