> On Jun 8, 2020, at 3:08 AM, Matus UHLAR - fantomas <uh...@fantomas.sk> wrote: > > Please avoid private copies of mail, thank you.
Odd, no idea why the reply-to didn’t fire instead... > >>>> I wonder that two very new documents describe something that has been long >>>> recommended to avoid: postgrey > >>> On Jun 7, 2020, at 8:03 AM, Laura Smith >>> <n5d9xq3ti233xiyif...@protonmail.ch> wrote: >>> I agree. Greylisting is a primitive, last century "sledgehammer to crack a >>> nut". >>> >>> It has no place in 2020's anti-spam. > > On 07.06.20 14:10, Charles Sprickman wrote: >> I’m going to have thoughts on this next week when I trial it. >> >> RIght now there is no other option for “pausing” spammers until they show up >> on my DNSBLs… >> >> I tried postscreen with the after-220 checks that implement a very brief >> “greylist”, but it was largely ineffective. > > I don't think greylisting would be more effective than this. I found > postscreen more effective (and less problemativ) than postgrey. > > Of course, you need to allow pre-greet checks, where especially pregreet > text (postscreen_greet_action=enforce) does great job. > > ... and it does not introduce delays. I’ll just note a few things that I put in my original thread: - We’ve trialed all the premium lists and observed no huge change - Even if that did work and a combo of 3-4 paid lists plus free somehow caught things, there’s not budget for this, it’s a value-add on connectivity - The reason I’m looking at greylisting is because the last premium trial we ran gave me feedback on the “leaked” spam and it was clear that we simply get spam before the hosts are blacklisted (why? 20+ year old domain maybe? beginning of the alphabet? I have no idea) - “Pausing” new senders would resolve a large portion of the leaked spam (see above) - Whitelisting known good senders seems like a reasonable option to avoid delays for real mail (and at this point, we have more user feedback on spam leaks than delays) - I’m running the pregreet stuff now, and there does seem to be a small improvement, but the short delay is not long enough to get ahead of the delay in the host being listed A new thing I noticed with our problematic spam senders is that while the majority do have valid DKIM, they are NOT publishing SPF records. In another thread there was talk of simply skipping greylisting on senders with valid SPF records, so that may actually be a great compromise in my case. Charles > -- > Matus UHLAR - fantomas, uh...@fantomas.sk ; http://www.fantomas.sk/ > Warning: I wish NOT to receive e-mail advertising to this address. > Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu. > Remember half the people you know are below average.
