On 2020/12/16 09:03, Dr. Christopher Kunz via mailop wrote: > We are still using Spam Assassin on our main setup, but I feel that it's not > aggressive enough to cope with current spam patterns, especially with > regards to its rather conservative bayesian learning parameters. > > Is it generally being superseded by other OSS solutions, or should I be > looking into fine-tuning it? It's a shared cluster, so per-mailbox bayesian > learning is an important feature for us.
Another +1 for rspamd. It is pretty good even with close-to-default configuration. > Is it generally best practice to also scan all outgoing e-mail on a shared > e-mail cluster for spamminess? It is definitely helpful to pass outgoing mail through the same system in order that it can watch message-ids and reduce score for replies to locally originated mail. It's also a good place in the pipeline to sign dkim if needed. Whether to have it score outgoing mail for spaminess really depends on local policies. Even if it's not blocking anything it maybe useful to have it as an alerting mechanism to pick up on potentially dubious accounts. _______________________________________________ mailop mailing list [email protected] https://list.mailop.org/listinfo/mailop
