Am Freitag, 11. März 2022, 13:31:22 CET schrieb Zakaria via Exim-users:
> Yet, this is my experience so far with spam. I started my mail server setup
> in mind to configure Greylisting, and once I came to it, I decided to drop
> it down. I just found it unnecessary with SpamAssassin and Pyzor scanning
> along with EXIM DNS black lists test, DANE, DKIM, SPF, DMARC and ARC
> validation. Whenever any of these validation measures fails, I add unique
> relevant “ could be spam " header, and use pigeonhole to rewrite subject
> and forward it to spam folder. I get very rarely spam emails, and if I do,
> I know what kind of failure it ended up with in spam, mostly marked spam
> content from SA or DKIM, SPF and recently ARC verifying fails.
...for us, greylisting (with SA at SMTP time) - beside all of this mechanisms
incl. further weighted DNSBL results - still is a important part of the
chain, even if mainly higher volume and/or more widely known / "older" mail-
addresses / mailboxes profit from (which get a high amount of spam).
Most of these mechs only validate the "technical source", but not if it's
spam. There is atill lot of spam (at least spam for our users) outgpoing from
mass hosters like hotmail, gmail and Co. which provide "valid" email senders
/ connections.
As typical, we lead "grey" stuff only to greylisting - with "self learning" to
avoid further greylist delays for known white "connections" (sender/
recipient). This typicall lead to a few more delayed emails at the beginning
for new email users - but after time only a very small percentage of emails
are delayed to the user thorugh the greylisting mech.
just my .02$,
niels.
--
---
Niels Dettenbach
Syndicat IT & Internet
https://www.syndicat.com
PGP:
https://syndicat.com/pub_key.asc
---
