Milter-greylist

19 \u0434\u0435\u043a\u0430\u0431\u0440\u044f 2016�\u0433. 0:30:00 CET, "Greg Troxel gdt@... [milter-greylist]" <milter-greylist@yahoogroups.com> \u043f\u0438\u0448\u0435\u0442:
>
>"Marcus Schopen lists-yahoogroups@... [milter-greylist]"
><milter-greylist@yahoogroups.com> writes:
>
>>> On the other hand, a host being on a bad blocklist leading to big
>>> delays and getting taken off is going to cause all sorts of
>problems,
>>> and people delaying mail for 12h instead of 15m does not seem likely
>>> to rise to the top.
>>
>> Could you technically explain the last point please?
>
>What I mean is that if a host gets put on an RBL that indicates such
>bad
>behavior that people want to delay mail for 12h, then it's likely that
>other anti-spam software will apply other penalties to that host, such
>as high scores, just rejecting SMTP connections, etc.    So if the big
>problem is that when it's taken off the list you think the greylist
>timer should drop to 15m immediately, but that it still gets 12h from
>the time it first tried, that's perhaps not as serious as what other
>people will be doing based on having been on this blocklist.

On a similar note: sometimes we have spam bursts where a "legit" but abused mail server sent out spam. It may have got greylisted, but got autowhited on timeout and resubmission of a message, and only some hours later that IP got into DNS RBLs because its infestation continues and spam level got to a critical threshold of the honeypots. But our relay already trusts the host as previously autowhited (and the trust continues as we get new spams regularly).

Is there a way to do enforce some checks (like DNS RBL) for *auto*whited hosts as well (but avoid the re-check for manual/ldap/dnswl etc. whitelists)?

Jim
--
Typos courtesy of K-9 Mail on my Samsung Android