While I don't consider our system as being too strict, we do use bl.spamcop,, spamhaus and a few others in addition to SpamAssassin.
Our "official" position is that if a client gets themselves listed (on any of them) it is their problem to get de-listed. As I understand it, spamcop lists spammers based on either reports by multiple users or receipt of UCE at spamtrap email addresses. Their system auto-removes a spammer (listing) if no spamming is done for 48 hours after being listed. If more UCE is detected or reported, then their stay on the list will exponentially lengthen. Part of the parsing of UCE that spamcop does is to report IP numbers to several open relay database keepers for testing. If the IP number fails the tests from there, they get listed additionally, while becoming more tedious and difficult to get removed. That "punishment" IMHO is well deserved, and never should the excuse of "My system spammed an hour ago, but it is not spamming now" ever be accepted. My opinion of a business who allows unpatched or unsecured systems to relay spam is not worth doing business with anyway, as they are neither ethical nor reliable. We mail server maintainers are then left with the choice of opening the floodgate and allowing all the UCE into our systems and just shouldering the costs involved, or we take a more proactive stance of blocking known spam IP numbers, thus conserving our expensive resources. In our case, we have accepted a small amount of risk, and advise a sender that if they find themselves blocked, they are directed to a page with a guide toward remedying the situation. The remedies can be from patching compromised systems, all the way to switching to another provider. Since we are a subscription service, our clients understand this methodology, and in the rare case where there is wanted email via a spammer's IP number, that is what we have whitelists for. Currently serving over 180 domains as a gateway, we have had only eleven requests for whitelisting since our origination. This with a block rate of around 96% (we aren't so strict after all) We have never considered accepting all spam and then dumping it into a catch-all box, as this would defeat our purpose of having the spam filters in the first place, that is to deny access to spammers, and conserve our resources for our clients. To do otherwise would quickly load our system to the max just be receiving all the junkmail. If the major providers would enforce their own Acceptable Use policies, all this would probably not be necessary, but they don't, and we do. There are several MAJOR providers that currently are at risk of losing their peering privileges altogether. Comcast and Attbi come to mind in many discussions I am seeing on various lists I subscribe to. Perhaps that drastic implementation will be what it takes to get the top executives at those providers to wake up and fly right. Who knows? ====================================== Our Anti-spam solution works!! http://www.clickdoug.com/mailfilter.cfm For hosting solutions http://www.clickdoug.com http://www.forta.com/cf/isp/isp.cfm?isp_id=1069 ====================================== To Unsubscribe: http://www.ipswitch.com/support/mailing-lists.html List Archive: http://www.mail-archive.com/imail_forum%40list.ipswitch.com/ Knowledge Base/FAQ: http://www.ipswitch.com/support/IMail/
