![]() ![]() Debug: sieve: action pipe: running program: warden-learn-spam.shĪug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: Finished running script `/var/qmail/popuser/warden-learn-spam.svbin'Īug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: Started running script `/var/qmail/popuser/warden-learn-spam.svbin'Īug 21 11:20:32 el7p17 dovecot: service=imap, ip=. ![]() Debug: sieve: Executing script from `/var/qmail/popuser/warden-learn-spam.svbin'Īug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: binary /var/qmail/popuser/warden-learn-spam.svbin: save: not saving binary, because it is already storedĪug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: Script binary /var/qmail/popuser/warden-learn-spam.svbin successfully loadedĪug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: Loading script /var/qmail/popuser/warden-learn-spam.sieveĪug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: Opening script 1 of 1 from `/var/qmail/popuser/warden-learn-spam.sieve'Īug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: file storage: script: Opened script `warden-learn-spam' from `/var/qmail/popuser/warden-learn-spam.sieve'Īug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: sieve: file storage: Using Sieve script path: /var/qmail/popuser/warden-learn-spam.sieveĪug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: imapsieve: Matched static mailbox rule Īug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: imapsieve: mailbox INBOX.Spam: MOVE eventĪug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Debug: Mailbox INBOX.Spam: Mailbox opened because: UID moveĪug 21 11:20:32 el7p17 dovecot: service=imap, ip=. With debug mode enabled this is what you will see in the /var/log/maillog when a user marks a message as spam: Aug 21 11:20:32 el7p17 dovecot: service=imap, ip=. Restart dovecot after enabling debug mode: systemctl restart dovecot To turn on debug mode for dovecot edit the file /etc/dovecot/nf and search for the mail_debug parameter: mail_debug = yes ![]() You can check that your SpamAssassin config has no errors using the command (It should not return any errors): spamassassin -lint The dovecot learning sieve filter also expects that SpamAssassin and sa-learn is in working order. If your learning statistics are not working you can go though the setup wizard again under Settings -> Application Setup to try and fix the problem (It will re-add database connection information to /etc/mail/spamassassin/local.cf which is used by the real-time learning). When a user marks a message as spam or ham in their email client you should see the Anti-spam - Learning Statistics increment the spam or ham learned values. All Bayes and Txrep data is stored in the Warden database for improved maintenance and performance. You can view the number of trained items on the Dashboard or in the Reports area under Anti-spam - Learning Statistics. Warden will log this training in dashboard under the Anti-spam - Learning Statistics dashboard widget.īy default BAYES rules will be enabled once 200 spam and ham message are trained. If a message is marked as not spam Warden will train the message as HAM (good email). When a user marks a message as spam in their email client (or though webmail) Warden will use a dovecot IMAP service sieve filter to train SpamAssassin (the message is fed though the spamassassin sa-learn binary). ![]()
0 Comments
Leave a Reply. |