Hi, thanks for reply! >> Amavis still tries use nonexisting user, spamassassin seems simply >> ignore bayes tests in that case. > > Are you sure? Maybe the new users are just below bayes_min_ham_num and > bayes_min_spam_num. >
Yes, that override seems intended behaviour, fragment from documentation: "If this options is set the BayesStore::SQL module will override the set username with the value given." Additionally I sent same mail to different accounts, trained differently, but got same score. When issue is with not enough training material (nonexisting account is good example), bayes score is missing. Best regards, Ain
