> FTR, directed at those who valiantly continue to self-host mail/SMTP: Greylisting is not sound any more in this day and age, because the largest mail services will rarely, if ever, use the same MTA instance to retry delivery upon a soft bounce.

I had this issue with SendGrid years ago: long story short, after discussing with support and eventually an engineer it turns out they weren't just looking at the status codes, but at the status messages. I don't recall the exact patterns they used, but they will retry if the message matches a fixed set patterns, and otherwise it would just discard it.

There was some back-and-forth over this, because our customer just had greylisting with the "wrong" error message. To be fair, they did turn it off for a few hours and took the conversation serious (none of this "we have passed it along", never to be heard from again) but they got back to us they turned it back on again "because the queues got too large". I mean .... okay.... Seems rather curious to break a fundamental aspect of email because "muh queues". Not having to worry about this sort of stuff is exactly why we're using SendGrid in the first place :-/

My experiences with MailGun were also not exactly stellar. At the time at least, these people literally did not understand how encodings worked and would mangle e.g. Greek or Hebrew emails in ISO-8869-7 or -8. Why? Well, turns out that "emails should be in ASCII or UTF-8 and there is no way for us to know which encoding is used". Ehh ... there is literally a header telling you... I sent a nice detailed email explaining this: no reply. Some follow-ups over the course of a few weeks: no reply. A not-so-nice snarky email inquiring whether the entire MailGun team was suffering from a horrible debilitating disease and if there was anything I could do to help: "well, we just didn't know what else to do as there is no way to solve this"...

I'm hardly an "email purist"; I understand there are practical concerns and the RFC isn't a stone tablet from the mountain. But this was just ridiculous. There are a bunch of other cases both SendGrid and MailGun are actually quite bad at.

Dealing with email providers is always a frustrating experience.

I haven't used DCC, but it looks like an interesting anti-spam toolkit:

https://www.dcc-servers.net/dcc/

It appears to support "weak" IP matching:

> All or part of the IP address of the SMTP client can be optionally ignored by DCC clients as far as the greylist triple is concerned. This feature may be useful for legitimate mail systems that shuffle messages among SMTP clients between retransmissions. See the dccm and dccd man pages.

https://www.dcc-servers.net/dcc/greylist.shtml

It doesn't quite sound like it does a job of 100% "same email, same sender, different mx in same domain" -but I suspect it works well enough in practice?

> Usually the DCC greylist system requires that an almost identical copy of the message be retransmitted during the embargo. If weak-body is present, any message with the same triple of sender IP address, sender mail address, and target mail address ends the em-bargo, even if the body of the message differs.

> If weak-IP is present, all mail from an SMTP client at an IP address is accept after any message from the same IP address has been ac-cepted.

https://www.dcc-servers.net/dcc/dcc-tree/dccd.html#OPTION-G

I can't recall what I used for greylisting last -possibly greylistd.

Anyway, the smart play these days might be to whitelist/greylist via SPF - I'm not sure if spammers (of the variant caught by greylists) generally have SPF?

See: https://poolp.org/posts/2019-12-01/spf-aware-greylisting-and...

https://github.com/poolpOrg/filter-greylist

Ed: although if service providers like mailgun simply ignore rfc and only "sometimes" retries... wwell that's a problem.