Topic: spam training as one user, but scanned and auto trained by another
==== REQUIRED BASIC INFO OF YOUR IREDMAIL SERVER ====
- iRedMail version (check /etc/iredmail-release): 0.9.8 MARIADB edition
- Linux/BSD distribution name and version: CentOS 7
- Store mail accounts in which backend (LDAP/MySQL/PGSQL): MySQL
- Web server (Apache or Nginx): Nginx
- Manage mail accounts with iRedAdmin-Pro? evaluating for purchase
- [IMPORTANT] Related original log or error message is required if you're experiencing an issue.
I'm experiencing a serious spam/ham filtering and training issue.
spam/ham data is trained and stored in SQL under the individual user names, such as "SomeEmail@MyDomain.com", using the sieve rules created with the iRedMail doc links.
This works for user training but this is not used for scanning incoming email. Instead, all incoming email seems to be scanned with and auto-trained under user "amavis" instead of "SomeEmail@MyDomain.com".
This means that all training errors can not be corrected by users, and incoming mail uses an error-filled Bayes database under "amavis".
from my /etc/mail/spamassassin/local.cf:
# Override the username used for storing data in the database. # This could be used to group users together to share bayesian filter data. # You can also use this config option to trick sa-learn to learn data as a specific user. #bayes_sql_override_username vmail
I do not override this because I would like it to use individual user training, and indeed it does - when users train with IMAP sieve, it trains to their own DB.
But when email arrives, it does not use that username for scanning or auto-training, it uses "amavis".
Does anyone have this working properly?
Can anyone tell me what/where to set this so it uses the recipient SQL for Bayes scanning and auto-learning?
many many thanks