Malcolm J Harwood on 20 Jan 2005 17:47:27 -0000 |
On Thursday 20 January 2005 2:17 am, Doug Crompton wrote: > I am curious what methods others are using to fight spam? I have been > using SpamAssassin for a few years and I got tired of actually accepting > this junk rather then rejecting it. I recently set up spam filtering rules > in sendmail - access.db using spamlist.org, also my own local additions > and also dnsbl's - abuseat.org and relays.orbs.org and some custom rules I > found to enhance access.db lookup. I'm using postfix, not sendmail, so I don't know if any of this helps: - I'm getting about 14,000 dictionary attack spam a day. I have the global account turned off, but I also keep a list of all the addresses used (periodially generated from the logs) so that any mail to those addresses can be dropped at the initial phase (and doesn't eat CPU going through spamassassin, or bandwidth actually accepting the mail for processing). - Greylisting. The server sends a temporary delivery error on the first attempt from any machine-from-to triplet, and lets any that is resent after a few minutes through. As most of the spambots aren't running through real mailservers that retry mails, this blocks most of the spam but not anything coming from a legitmate server (with the exception of yahoo, which sends listmail out with unique bounce addresses every time, so they have to be whitelisted). This has reduced inbound spam significantly. (At least with postfix, it's in the auth phase, so it's not even checking headers, so much lower bandwidth usage). - Amavisd with spamassassin, (version 3, makes quite a diffference; with a subset of rules-je-jour) and razor. Razor gets a lot of it. That generally takes care of most of the rest. I've gone from a few hundred spam a day to live addresses, half of which got through the filters, to maybe one a week getting through. - I'm also not getting any false positives at this point. > My real reason in sending this is to ask if anyone has lists they would > like to share, especially of domain blocks. I am currrently blocking a lot > of the world outside of the US and my list is growing! I don't use any DNS blocks (though I do have spamassassin set up to use content-based domain black holes - which hits a significant fraction of the spam). > I also actively send abuse messages to mostly US ISP's that I don't block > after looking at headers. I am not sure how much good that does but I like > doing it! I've never gotten a response from doing such. Does anyone know if it makes any difference at all? > I firmly believe that the way to eliminate this crap is to NOT accept it > rather then accept it, test it, and then dump it. As long as we accept it > it will keep coming! That's where greylisting makes a big difference. -- "If you're never scared, you're not trying." - Llewellyn (www.ozyandmillie.org/2003/om20030124.html) ___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug
|
|