[Rspamd-Users] False positive and learn ham

[Tobias Kirchhofer]

Hi everybody, we are coming from spamassassin, new to rspamd.

Currently we test a setup for our new mailserver with postfix, dovecot, 
rspamd, redis. One domain is already running there and we experience 
difficulties with the behaviour of rspamd.

Some newsletter are tagged as spam false positive and sieve is sorting 
them out into the users spam folder. For giving the user a chance to 
tell us what is spam and what is not spam We have configured sieve and 
dovecot like this:

```bash
### INBOX/Spam
# From elsewhere to INBOX/Spam folder
imapsieve_mailbox1_name = INBOX/Spam
imapsieve_mailbox1_causes = COPY
imapsieve_mailbox1_before = 
file:/var/vmail/sieve/global/learn-spam.sieve
 From INBOX/Spam folder to elsewhere
imapsieve_mailbox2_name = *
imapsieve_mailbox2_from = INBOX/Spam
imapsieve_mailbox2_causes = COPY
imapsieve_mailbox2_before = file:/var/vmail/sieve/global/learn-ham.sieve
```

If we move a false positive from the spam folder to eq. inbox rspamd 
indicates in the log that it learend the mail as ham:

```bash
2018-12-02 09:21:46 #743(controller) rspamd_log_reopen_priv: log file 
reopened
2018-12-02 09:21:46 #743(controller) <221190>; csession; bayes_classify: 
skip classification as spam class has not enough learns: 4, 200 required
2018-12-02 09:21:46 #743(controller) <221190>; csession; 
rspamd_controller_learn_fin_task: <127.0.0.1> learned message as ham: 
0000BBE2.5BFBAA15 at 192.168.1.100
```

Next time the newsletter arrives it is still tagged as spam.

We have a basic rspamd setup inspired by 
https://thomas-leister.de/mailserver-debian-stretch/

Could someone pleas give us a hint?

Thank you,
Tobias

-- 
collect at shift.agency