Search results

since i've enabled the "before queue" filtering, i get notifications of smtp-errors every now and then: the sender/receiver/sending-host tuple looks like totally legit (and includes "trusted" mailserverse; "trusted" in the sense that i trust them to do a good job and not send out spam; but not...

did you add the additional domains in the *Configuration->Mail Proxy->Relay Domains* setting?

so you subscribed to legit newsletters, and receive the emails you asked for and then you wonder why? for me it sounds like everything is working as expected. (a "newsletter" is not necessarily spam; esp. if you can sign up manually)

yes of course. journalctl and /var/log/mail.info are full of lines like: i was only talking about the webinterface.

ich hatte das gleiche problem, und habe es dann gelöst, indem ich im eigentlichen (terminierenden) mailserver, das header-feld umgeschrieben habe. siehe Thread: using spam-bar instead of spam-stars

I ended up with rewriting the header-field in the receiving (final) mailserver, replacing * with +. Something like the following: - In PMG, add a rule that adds __SPAMSTARS__ to the X-proxmox-Spam-Level header-field. - in exim, add a data-ACL (acl_check_data, or via CHECK_DATA_LOCAL_ACL_FILE)...

is there any way to draw attention to this issue? as of now, the charts on the #pmgGeneralMailStatistics page are only useful to gather information about spam, but not about non-spam mails.

to be sure: i would *love* to see the rejected emails in my log-files. but since this is no regression from the milter-solution, i immediately adapted "before queue".

for me, the current behaviour is already an improvement: - it provides the same core-feature as the milter-solution: hard reject (544) at SMTP-time instead of discarding the email (this was the reason why i deployed heutger's setup in the first place) - it has the same caveats as the...

i would add 'SOLVED' myself, but i'm not the owner of this thread, since i only continued what @grefabu started...

oh sorry, my last post was actually meant to say: "yes it's working now. it seems i did something wrong in my first try". thanks.

DoH, i stand corrected. it seems that a reboot does not fully include a `pmgconfig sync --restart 1`. My main.cf was slightly modified, to use spamassassin-milter (following heutger's tutorial), but that part was commented out before doing the reboot (so I'm pretty sure that there were no...

this is what i see in PMG's tracking center: Dec 2 13:01:27 mailgate postfix/smtpd[4953]: connect from anotherdomain.test[89.106.215.196] Dec 2 13:01:27 mailgate postfix/smtpd[4953]: 5944420014: client=anotherdomain.test[89.106.215.196] Dec 2 13:01:27 mailgate postfix/cleanup[5040]: 5944420014...

thanks. i tried adding before_queue_filtering 1 and it does "something" (as in: adding `--smtpd_proxy_filter` and similar lines to postfix/master.cf), but how can i test whether it actually works? i tried sending a GTUBE-email to the gateway, but it was silently BLOCKED (aka: discarded). is...

after upgrading to PMG-5.1, my ClamAV installation complains about being outdated. E.g. this is what I get when I manually click the "Update Now" button in the web-frontend: ClamAV update process started at Mon Oct 15 14:55:46 2018 WARNING: Your ClamAV installation is OUTDATED! WARNING: Local...

no. you somehow have to call `sa-learn` on good and bad emails. the problem is mainly how to get the spam/ham mails from the users to the pmg. one option is to use imap, another one to use rsync, a third one to use an USB-stick,....

would it be feasible to train the bayes-database with emails the user released from the quarantine as ham? - sometimes false-positives get caught by the quarantine. - the users will usually notice, and release these mails from the quarantine. - my idea is to use this information to train the...

the bayes won't get trained unless you have a spam-score that is greater than 12. ("10-12 points" qualifies as less than 12, so those mails don't automatically train the database). so yes: you should train the bayes-db manually.

when an email gets analysed for SPAM, it gets a Spam-Score assigned. this Spam-Score is used for various filtering (e.g. whether to put the email into quarantine; or whether to block it completely; or...) this Spam-Score is also used to automatically train the bayes filter. since spamassassin...

i think the question is more targeted towards: "which checks must be disabled to accept connections from IPs with incorrect public DNS entries?"