Am 17.02.2015 um 23:37 schrieb LuKreme:
On 17 Feb 2015, at 08:27 , Robert Schetterer <r...@sys4.de> wrote:Am 17.02.2015 um 16:13 schrieb LuKreme:OK, so I have spamass-milter running, but I need to train it. What is the proper way to do this?you dont train spamass-milter, you should train spamassassinspamassassin has existing user-specific training already in place. Spamass-milter isn’t using the user DBs.
because in a default milter-setup the one and only user is the user which SA and the miler service are running as, hence my script which needs maybe small adjustments for your environment (--no-sync and so on depend on the config, directories needs to exist and permissions for samples needs to be correct)
no mysql and what not, just a default bayes-db, two traing-folders for ham and spam and the script for feed new eml-samples as well as the option for ac omplete rebuild based on the current samples, the corpus will stay forever on that machines and samples are named YYYY-mm-dd-number.eml
hence we start spamd as well as the milter service with the same user, both on a high port and directly with the correct user and group with systemd
after that all relevant is in milter-home below .spamassassin and the global settings as usal in "local.cf"
that setup with the global bayes is now running for 6 months replacing a commercial solution after 10 years and after all the training and adjustment effort i have never faced a more accurate filter in case of spam as well of ham to prevent false positives
signature.asc
Description: OpenPGP digital signature
