Knowledge Essentials - 3Essentials Hosting

Spam Protection: Overview of 3Essentials Spam Protection

Article ID: 87

 Back to Search

OVERVIEW OF 3ESSENTIALS SPAM FILTERING - for shared hosting customers

Spam has become an increasingly annoying and costly problem for both 3Essentials network, systems and most importantly our customers.  To help protect our shared hosting customers, 3Essentials employs a 2 tier protection mechanism for spam:

  • An enterprise-wide spam gateway through which all mail flows.  The enterprise gateway blocks based on objective measures which are considered industry standard: rate control, internationally recognized blacklists (spamhaus, spamcop), lack of conformance to industry wide email standards, and other objective measures.  This gateway blocks/rejects mail that it finds fails these tests. 
    • There are no user-specific or domain-specific settings available for the enterprise gateway.
    • This enterprise-wide spam filtering is required for all shared hosting domains in order to protect our shared hosting infrastructure.  This layer cannot be disabled upon customer request.  This layer is not provided to dedicated server hosting, as with dedicated servers only they carry the risk of not implementing such infrastructure protection.  Other protection mechanisms are available for dedicated server hosting customers.
    • The false positive rate for this objective spam protection layer is extremely low.  However, if a legitmate email to your 3essentials hosted domain has been blocked, please submit a support ticket, and we will identify and provide feedback on the  reason.
    • For every 1,000,000 emails received, over 600,000 are deemed as spam, and blocked to protect our customers. 
  • Host based spam protection: Depending on which server you're on, you may have another level of spam filtering that's integrated with the mail server on your hosted server.  This layer would be considered the "subjective" layer of filtering, i.e., what's spam to one person is not spam to another.  As such, you control spam filtering settings to meet your specific needs.  There are three components to host based spam protection: content filtering, bayesian analysis, and image scanning 
    • Content filtering: a very powerful tool to allow you to filter messages based on sender, header info, subject, text, etc. (virtually any property of the mail), and configure a specific action for mail meeting that criteria.  Even more useful, content filters can be set at a domain-wide level AND users can set their own user-specific filters.  Find more details in the following KB article: Spam Protection: Content Filtering.
    • Bayesian analysis: learning by behavior... you tag the mails that are spam, the bayesian filter gets smarter at identifying what is spam and what isn't. Find out more details in the following KB article:Spam Protection: Training the bayesian spam filter.
    • Image scanning: spammers have gotten smarter... they know you're filtering their mail based on keywords... so now they are sending IMAGES that contain their advertisements, knowing that standard text filters can't pick these up.  3Essentials uses image scanning to perform OCR scanning of embedded images to determine if an image contains spam.  Find out more details in the following KB article: Spam Protection: Enabling 3E's image scanning spam filtering 
    • Spam Free Trusted Mail: For the ultimate in spam protection, 3Essentials offers Spam Free Trusted Mail.  Enabling this protection guarantees you virtually 100% spam protection.  Find more details in the following KB article: Spam Protection: 3Essentials Spam Free Trusted Mail

RECENT CHANGES

  • June 1 2007:  3Essentials implements Spam Free Trusted Mail
  • May 11 2007:  3Essentials implements Sender Policy Framework (SPF) /  Sender ID to enhance spam protection. 
  • March 19 2007: Based on customer feedback, the remaining subjective component of the mailgateway, image scanning of embedded images, will be reconfigured to no longer block mail it determines is possibly spam, and instead will insert a custom email header, which mail clients can set custom rules to handle.  As such, customers will be able to choose whether or not they wish to utilize this component, and they can chose their own disposition of email tagged by this component.  Please see KB article Spam Protection: Enabling 3E's image scanning spam filtering for more details on enabling this feature.
    • As a result of disabling the automatic blocking of mail by this component, customers may receive more spam mail that utilizes embedded images containing advertising.  Please see the following KB article which provides instructions to re-enable this feature for your domain: Spam Protection: Enabling 3E's image scanning spam filtering
  • March 16 2007: Enterprise Spam Gateway services enabled for shared hosting servers in our Dallas, TX datacenter: web14, web16.  Previously, our enterprise spam gateway services were only active for our Raleigh, NC datacenter.  As of this date, we have completed implementation of an enterprise spam gateway infrastructure for our Dallas, TX datacenter as well.  
  • February 2007:  3Essentials changes enterprise spam gateway solutions.  The previous gateway solution included blocking and tagging based on bayesian intent analysis (at the enterprise gateway level).  The bayesian intent analysis tagged emails with **SPAM***.  Based on customer feedback regarding inaccuracy and lack of user-specific control, 3Essentials implemented a new enterprise gateway solution removing the bayesian intent analasis and tagging.  Instead, primarily objective tests are performed, and if failed, mail is blocked/rejected.  One subjective test, image scanning of embedded images, is left as part of this gateway to test/evaluate usefulness to customers. 

 

 
Downloads Associated With This Article
No downloads are currently associated with this article.