Hi All,
Setup: RH9.0 using sendmail with procmail and spamassassin (3.0.4) for site wide use. Clients: Windows XP, using Microsoft Oulook 2000, connecting to RH9.0 server using POP3. Have created 2 users (
[EMAIL PROTECTED] and [EMAIL PROTECTED]) which users can send False Negatives to. Have a cron job which then runs sa-learn against the respective ham/spam mbox files.
RH server with spamassassin works OK moving spam to mbox file in each users $HOME directory as does cron job.
Overview: Recently switched on bayes auto learning and have 1000+ ham/spam messages respectively. At this stage I am still getting some False Negatives appearing in my/our Inbox(es). According to the spamassassin web site I (all users) should be able to click "tools/resend message" in Outlook and then send the message to the spam/ham accounts.
Issue: Using the above method of sending the messages to the ham/spam accounts changes the header fields in the message.
Question/Advice: How can I/we get around this, resolve this - making sure that the original header stays intact?
Thank in advance
Mark
