RBL Example

[philsbbs]

Can some one show me some examples of RBL they have installed.

I looked in the manual and this forum but cannot see any examples, some ive tired reports dns string errors.

Thanks in advance.

[Code Crafters]

If you use either the SPAM wizard or the Add Preset RBLs button you can add all of our recommended default RBLs tested and approved by Code-Crafters. Make sure you have the latest version for the most up to date default RBLs (currently 2.62).

[philsbbs]

thanks i've installed default RBL is there any chance chris i can email you the admin access and you check how my setting look.

ive installed the mail server at the weekend and have checked nearly 2000 email and would like to have it set up to now delete spam etc and use recommend settings (as recommend by you the expert).

[Code Crafters]

Yes, of course. Just email your mail server domain, remote admin port and admin username / password (general settings) to chris@code-crafters.com and I'll have a look.

But basically, I recommend the medium level filtering via the SPAM wizard and then if possible to set up Bayesian filtering separately for any accounts that can be trusted to accurately train this. This involves users sorting mails into SPAM / non-SPAM folders for Bayesian filtering to learn from and changing SPAM filtering to set the SPAM flag and not reject (delete) the email at the SMTP SPAM filtering level so that mails can be stored for later training.

I also recommend adding the content filter preset rule "SPAM Identifier" to mark <SPAM> in the subject of SPAM detected mails for easier sorting and if you want adding an action to that rule to automatically move mails to the SPAM fodler if SPAM flag is set but this is better to do once the Bayesian filter is well trained (over 1000 SPAM and 1000 non-SPAM mails). If you do automatically move mails to the SPAM folder you need to periodically check for non-SPAMs in there that shouldn't be. You'll get less false positives as Bayesian is better trained.

We get over 1000 SPAM mails a day and our Bayesian blocks 99.5% of these meaning that rarely do any hit our Inbox and very few mails (maybe a couple a month) get filtered as SPAM when they're not so it's very accurate.