• The BIND DNS server has already been deprecated and removed from Plesk for Windows.
    If a Plesk for Windows server is still using BIND, the upgrade to Plesk Obsidian 18.0.70 will be unavailable until the administrator switches the DNS server to Microsoft DNS. We strongly recommend transitioning to Microsoft DNS within the next 6 weeks, before the Plesk 18.0.70 release.
  • The Horde component is removed from Plesk Installer. We recommend switching to another webmail software supported in Plesk.

Help understanding the postfix+spamassassin mailflow

P

Paul Martin

Guest
Ok, so in a previous post, I explained how to allow spamassassin to learn messages over 256k. Well, recompiling the spamc binary worked well, and it allowed >256k emails to get learned, but it had a nasty side effect - all of the learning was saved to the "popuser" account, not the actual account where it was sent to. If I replace my new binary with the old one, then it works fine. Since NOTHING ELSE has changed, this leads me to believe that the version of spamc that Parallels ships with 9.2.2 is either compiled differently or has additions made to the source that allows it to auto-determine the user to learn the spam as.

After digging into any and all spamassassin/postfix emails that I can find, I don't see where postfix sends the email to spamassassin for spam checking. If I check the maillog, I obviously see lots of references to it, but I don't see in /etc/postfix/main.cf or /etc/postfix/master.cf where postfix basically says "Ok, you got email, it's going to a legit user, now take it and run it through spamassassin prior to delivery". Can anyone help me locate that? I think I can take it from there.

Maybe.

Or, if anyone has the changes that Parallels made to the spamc source code, I'd be happy to integrate that as well. It'd be a more elegant solution. Come to think of it, since spamassassin is an open-sourced under the apache license, aren't they required to release all modifications, or am I completely out in left field here?

Thanks for any insights anyone can share.
 
That documentation doesn't exactly help me, but I appreciate the answer. I want to create a handler that allows me to pass in messages that are larger than the hardcoded value that Parallels has set. Most of my spam is larger than 128k (or whatever the value is) so it's not getting checked. I want to rectify that. Can you assist me with that?
 
Back
Top