![]() ![]() While testing, I became aware that (almost) nothing is happening. I initially had other sieve_dir and sieve in nf, but it always wrote to the above, so I adapted nf accordingly. Sieve = /var/mail/sieve/%d/%u/.dovecot.sieve In /var/mail//username/.dovecot.sieveĪnd in /var/mail//username/sieve/managesieve.sieve or anyfiltername i chose within roundcube. I activated said plugin and it writes sieve files as following, no matter what I define in nf. Posted: Sun 3:31 pm Post subject: roundcube managesieve plugin interfears with rspamd learning Gentoo Forums Forum Index Networking & Security Roundcube managesieve plugin interfears with rspamd learning By the way, I’m using AirMail not Apple Mail as my client and unlike the rest of the world, I’m still using the email server for my web host.Gentoo Forums :: View topic - roundcube managesieve plugin interfears with rspamd learning I would pay up to $200 for a class (in person or online) for working with spam software to optimize my spam filtering. The question becomes whether I should turn off the server spam filtering, in which case, what to do about the emails coming to my phone? In the raw email source, I see X-Spam-Status: Yes, score=4.1, so perhaps the Spam Filtering on my email server is marking it as spam. There are several entries in the whitelist for various Washington Post emails and yet, again this morning, I find another Washington Post email marked as spam. I regularly get emails from The Washington Post and I have trained SpamSieve over many months by selecting these emails and choosing “Train As Good”. Just this morning, I got two misclassified emails. And yes, the statistical analysis is not working correctly. I agree that I should not be training every single address, although somewhere it says that for SpamSieve to work optimally, you should be training regularly. If you’re getting a lot of these, there’s probably something wrong with the setup or training on your Mac, and it’s better to fix the source of the problem, so that it all works automatically, than to make lots of domain-specific rules. Good messages misclassified as spam should be very rare. SpamSieve automatically learns all the addresses that it sees and also learns, statistically, about whole domains. You should not be training every single address. ![]() But it’s not a common enough action to clutter and confuse the user interface and risk spams getting through because of inappropriate whitelist rules. You can do so if you want by creating rules yourself in the Whitelist window. I think it’s only been asked a handful of times in the last 15 years because it’s rarely necessary to whitelist whole domains. Is there some reason why this rather obvious functionality isn’t available? I use SpamSieve (current version) with AirMail and I have two pop-up menu choices: Train As Good and Train As Spam. This applies to SO MANY SITES that I cannot tell you how important it is. However, for many sites and services, I may receive or The point is that I just need to white list the domain. The Train As Good option adds the specific email address to the White List. ![]() I’ve been using SpamSieve for over 6 months now. I’m sorry if this is a perennial question. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |