Yes...Would be great.  
We operate a couple of email systems to where you can identify a specific email as a "honeypot".   And again... ANY email that hits that.. the sending address, and I/P get marked higher marks of spam for X amount of minutes.

 

 

Sign us up !  I like that.

Rick, this is a great idea.  I have added this to our features request list for further consideration by the development team.

Thumbs up!

Well I'll pass on a little trick that works for me, but I won't go so far as recommending it to anyone.  We have multiple MX Records for all domains. We discovered that many spammers send to the lowest priority MX Record (highest numerical weight).  We also found that while the highest priority SMTP servers were online there were no legitimate messages sent to the lowest priority MX Record (highest numerical weight) server.

 

For example if you have:

IN MX 10 mainserver1.domain.com

IN MX 20 mainserver2.domain.com

IN MX 30 backupserver.domain.com

You will find that nothing but spam will be sent to backupserver.domain.com and will be sent to your main server for processing.

 

But if you change the above to:

This has been implemented in SmarterMail 17 as a spam filter.

Just some notes.

-You can block on the connection level if the IP matches one in the honey pot, granted the IP is not in the whitelist, is not a gateway, and was not IP Bypassed.

-There are 3 levels of spam score, Passed, Triggered, Listed.

-You can set any number of honeypot addresses. If these are found in the recipients of an email the email is given the "Triggered" score.

-The honeypot addresses you set will automatically be accepted in SMTP regardless if this address is in use on your server or not.

-"Triggered" means that the email had a recipient that matched a honeypot address.

-"Listed" means that IP was found in the honeypot.

-"Passed" means the IP was not found in the honeypot nor was the email being delivered to a honeypot address.

 

If you have any questions at all let me know and I will do my best to respond. I'm happy we were able to include this, and other spam checks (stay tuned), in the SmarterMail 17 release!

 

Would also be cool if we could get some global system working with all SM 17 installs that use this feature. IE a Pass Weight, Listed Weight, Triggered Weight and maybe a Global Weight, where SM every say hour or so would pull down and also share its list of IPs to some central system.

I found nothing in the documentation on how to use this, set it up, etc. I just installed Xenforo forum'ware and it had a link on how to get a honey pot API key. I'd like to use this but really am not sure and again nothing in the docs outlining this system and how to properly configure it. I see nothing on "yourapikey" in the spam config. I'll keep digging.

Go to Anti-Spam settings, turn on HoneyPot put in a fake email address, any deliveries to that address will be treated like a normal session, except given a triggered honeypot score. Any further deliveries from the sender address will be given the honeypot score.

So for example if we created honeypot@smartertools.com and I sent from test@example.com to honeypot@smartertools.com, that session and any further deliveries in your system from test@example.com will be given a score.

Make sure your fake email is a domain that you serve but that the username does not exist already. If you need further help, lemme know and I can explain further.

It seems this feature and many others in SM17 are not documented.  When will the documentation be updated to reflect this latest, production release? Thank you.

ST does itself a disservice when they add new features without explaining how AND why they should be used.

The documentation has always been minimalistic sadly. It's enough to say "we have documentation" but is useful? I don't typically find answers within their docs. If they want good docs they should move to a system such as with Atlassian Confluence and have a tech writer do a great job with the docs. Make it a great resource, not just we have docs.

I haven't upgraded to v17 yet but I'm glad to see new antispam features being added. As for honeypot email addresses, would you just create a generic bob@domainname.com (for example) and hope for the best? Or would you somehow advertise the honeypot email address on a dummy web page etc. that search engines crawl?

@Matthew,

Yea, using something generic and stick it maybe on the website. If you search for "honeypot css" on Google, the first couple results have resources on how to place and hide honeypot addresses on a website.

Without proper documentation and an intuitive UI, no one knows how to properly configure this. Here is a screenshot of how Xenforo forums does it in their anti-spam area as I just migrated my forums to their latest release. Notice a helpful description, a link to the honeypot web site, etc. I have no clue on this project honeypot, your UI should link me to where I need to go whether your docs, honeypot, etc. and help me use it without googling endlessly.

I've tried implementing the Honeypot feature on a test basis, and found it just doesn't work. Messages that go to a honeypot address are identified, but subsequent emails from that sender are not given any additional spam score from what I can tell.

@Joe Wolf; the behavior you outlined caught my attention (regarding lower priority MX records.) I believe this is likely because email providers will keep a lower security environment set up as a best effort delivery server listed in MX records to catch email coming from older email servers, email servers in less secure environments, mass mailers, etc. These servers are typically gateway machines, and spammers are probably targeting these records specifically due to potentially lax security. I just wanted to share my thoughts on this.