Alan Premselaar wrote:
Philosophically, it makes more sense for SpamAssassin to focus on
identifying SPAM, and let another application (MTA, procmail, etc)
focus on what it was primarily designed for: processing
(delivery,rejection,etc) of said email. It's certainly no more of a
hassle to add a procmail rule to dump a blacklist hit to /dev/null
than it is to add a procmail rule for other delivery options.
There may be cases where it would be very inappropriate for *any*
mail, blacklisted or not, to be dumped to /dev/null.
I don't disagree with most of what you said. However, I don't
necessarily agree with the above because while I can add a procmail rule
to handle a specific user's blacklist I can't get back the wasted CPU
cycles which spamassassin expended ... perhaps all I am really asking is
if there is a way to allow spamassassin to just stop processing a
message that is in a blacklist to save the cycles? I am not asking for
spamassassin to become an MTA/MDA.
Privately Ed Kasky raised the good point about spoofing. I believe
whitelists have to have the full weight of a spamassassin scan to catch
messages with spoofed addresses. But blacklists ... I don't believe
suffer this problem. Yes I might receive a spam message from somebody
spoofing an arbitrary Email address. If that address happens to have
@ebay.com then I am highly unlikely to blacklist it. If it is from
@anybody.com then unless I have some reason to believe that I should
expect legitimate Email from that address, I argue that blacklisting it
does no harm and that any message found to be on a blacklist could be
processed with the minimal of effort by spamassassin. But then what do I
know? :)
--
Paul ([EMAIL PROTECTED])
