Last month (our) SpamAssassin (serverfarm) caught 103 million spam messages.
If you can block 70% of that with greylisting, does it really make sense to waste cpu(+ram) cycles processing / scanning and wasting bandwidth on garbage?
Yes, it might. It depends on your priorities. If you wait until you've had a chance to SpamAssassin-ify the body, you might discover that the email has a spam score of -10.0 without greylisting. In this instance it would make
This doesn't happen to the majority of the mail..
perfect sense to skip greylisting and just accept the email outright.
Exactly how many of those cases do you find/have? ;)
If CPU, RAM, or bandwidth are a priority, I agree with you. But there are
I have no idea what type of company you are envolved with or what your position is, but if you do work for a large company with many clients, are you willing to put the majority of them at risk by doing pre-queue scanning?
1 attack and you'll be singing a different tune ;)
plenty of installations out there that place a higher priority on timeliness-of-delivery, which is hurt by greylisting.
Running your greylisting daemon in learn-only mode for a week or two solves the larger majority of your problems without having to reject a single mail..
Dont get me wrong, i do infact see the valid points you are making.. There is however more than 1 way to skin a cat without putting all your other clients at risk..
Cami
