Re: [anti-spam-wg@ripe.net] Antivirus bounces

[der Mouse <]

> I think a combination of serveral solution mentioned in this thread
> would be good.

> 1. Standard human and machine readable bounces
> 2. This bounces should contain the orign header of the virus mail.

We already have a standard for this, which is widely ignored.  (My
automated filters know how to pick apart multipart/report bounces, but
I get an awful lot that aren't.)

> 3. The mail server shouldn't bounce mails, if the virus is known to
>    forged the return path.

What's your stance on simply refusing the mail?  That's what I
currently do, and while it does in some sense cause bounces, I'm
certainly not generating them.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B