On Sat, 15 Feb 2003, Gary wrote: >--On Saturday, February 15, 2003 07:10:06 PM -0500 Charlie Brady ><[EMAIL PROTECTED]> wrote: >> qconfirm (http://smarden.org/qconfirm/) would be another excellent >> option (and lighter weight). >I use qconfirm, and it works perfectly... However, this guy apparently >joined the list, so it would pass through qconfirm, as it did mine. >(anything from the list goes through, based on envelope address, and >anyone else who uses that email address, gets held off until a reply.)
Here's the cut from the qmail log: 2003-02-15 22:48:21.470179500 new msg 1853050 2003-02-15 22:48:21.470189500 info msg 1853050: bytes 43806 from <[EMAIL PROTECTED]> qp 13082 uid 100 2003-02-15 22:48:21.472056500 starting delivery 26835: msg 1853050 to local [EMAIL PROTECTED] 2003-02-15 22:48:21.472067500 status: local 1/10 remote 0/20 2003-02-15 22:48:21.472072500 starting delivery 26836: msg 1853050 to local [EMAIL PROTECTED] 2003-02-15 22:48:21.472079500 status: local 2/10 remote 0/20 This really pisses me off (sorry for the bad language). But I understand the spammer's reasoning. By sending spam to a mailing list, you can reach a lot of real mailboxes with little effort. Still, I am tempted to switch on per-message confirmation with qconfirm. I will let it pass for now, simply because I have _never_ seen any spammer have the nerve to do something as stupid as this (message is easily traceable). But the next spam that goes to this list will make me install per-message qconfirm immediately. I will never install any content filtering on this list, simply because of its false positives. I'd rather move the whole list to a Silc channel, or require all senders to use a personal key to sign their messages. ;) Once again, legitimate users lose to the overwhelming power of teams of annoying people who will do anything to destroy the SMTP protocol. Andy -- Andreas Aardal Hanssen | http://www.andreas.hanssen.name/gpg Author of Binc IMAP | Nil desperandum
