Using sa-learn

[Errol Neal]

Hi,

I am trying to figure out a way to get spam that makes it through our
mailscanners fed into sa-learn. Since I have no idea
what MUA's our users are using (because we are a service provider) I need
to come up with a sure shot way that will work
with every MUA.

I saw some earlier discussions on the list about forwarding messages
being  no no because the original mail headers are modified,
IE the message that was spam has now changed. Another gentlemen suggested
that users be trained to forward the orignial message
as an attachment and use scripts to extract the original message from the
attachment and feed it into sa-learn. I think this may
be the best approach but, I cannot say that all of my users are using
Outlook. I was wondering about the redirect feature of Eudora,
If i redirect a message from eudora, the headers are slightly changed. Not
mangled like they are if I were to do a forward, but they are somewhat
changed. Will that corrupt what I am trying to achieve?


Errol