Anthony Warren wrote:
> My organisation is concerned with the risk of
> false positives. For those that use spam reduction
> techniques - how do you cope with the risk of FP's ?
- We only /tag/ spam; it's up to the user to configure his/her inbox
agents.
- we use RBLs and own black/whitelists (eg mail from yahoo.* is
rejected)
- all refused mail is rejected at RCPT time - the sending MTA should
inform the user about the bounce.
- no mail is forwarded to :blackhole: - either the user gets it or the
sender gets a 5xx error at RCPT time.
- mail for non-existing users is bounced immediately (exchange and notes
are too stupid to do this at RCPT time and then spend hours sending
bounce messages to non-existing recipients <giggle>)
- user list is periodically exported from ldap -> plain text file (real
time ldap was too expensive and unnecessary IMHO)
