Hi
all,
I just got a string
of 13 messages with the following header.
Message-ID:
<[EMAIL PROTECTED]>
The content of
each message varies a little so the Bayes score varies from BAYES_50 to
BAYES_90, so I wanted to sa-learn
them. However, since the message-ids are
all the same Bayes refuses to learn the messages. Is there any way
to get these messages learned?
I've also received other messages with similar M-Ids, but the
domain name changes.
It's pretty easy to make
a rule to catch this ratware oops, but does this point out a weakness in the
current (2.63) Bayes implementation?
Here's my test rule,
comments?
header STI_RMID Message-Id =~
/\.\$rand\d{1,2}m\.\$rand\d{1,2}m-\$rand\d{1,2}m\@/
describe STI_RMID Ratware oops, $rand Message-Id
describe STI_RMID Ratware oops, $rand Message-Id
Thanks,
Balam
