[Mimedefang] fetchmail and mimedefang for antivirus

[Jason Englander]

On Tue, 25 Jun 2002, Jim Shewmaker wrote:

>   Ok, so I've got fetchmail rigged to collect from email accounts, the
> problem is, it attemps to dump the mail at port 25, mimedefang does its job
> and rejects it.  The problem is that fetchmail can't flush the old message
> from the original server since it can't finish it's job, and it tries to get
> it again and again, every time sending my virus notification email to the
> originator.
<SNIP>

I haven't used fetchmail in a while, but semi-recent versions should have
the 'antispam' option:

       -Z <nnn>, --antispam <nnn[, nnn]...>
              (Keyword:  antispam)  Specifies the list of numeric
              SMTP errors that are to be interpreted as  a  spam-
              block  response  from  the listener.  A value of -1
              disables this option.  For the command-line option,
              the list values should be comma-separated.

My sendmail returns 554 on a bounce, which is one of the default codes.
See fetchmail -V for the ones fetchmail thinks are "spam block responses".

Also the 'spambounce' option might be one to check:

       If the spambounce option is on, mail that is  spam-blocked
       triggers  an RFC1892 bounce message informing the origina-
       tor that we do not accept mail from it.

So, you'd think that the default antispam of 571, 550, 501,
554 and spambounce off would take care of the problem.  Do you have
something else set for those?

  Jason

-- 
Jason Englander
jason at englanders.cc