To do this, you just need to find the IP address of the remote server that sent the message (e.g. from the topmost "Received" header), then set the TCPREMOTEIP environment variable before starting spamdyke. Fetchmail has the ability to pass incoming messages through scripts, so it should be possible. There's no need to strip message headers on spamdyke's behalf; it doesn't check headers anyway. You may, however, need to remove headers if you're going to use other filters like SpamAssassin or ClamAV.
However, it won't work. Even if you can make a message _appear_ to be delivered from the remote server, what will happen when spamdyke rejects it? Normally the remote server is responsible for retrying or bouncing the message, but in this setup your script will have to figure out what to do. If spamdyke generates a permanent rejection code, your script can simply delete the message but legitimate senders (false positives) will receive no notification that their message was lost. If it generates a bounce message instead, it'll create backscatter. If spamdyke generates a temporary rejection code, will your script store the message and retry it later? It'll have to be pretty smart to discard the messages after they've been retried a number of times; I think you'll end up reimplementing most of a mail server's functionality. Some filters, especially graylisting, will be completely useless because your script will _always_ retry delivery, even if the original server would not have done so. spamdyke will be unable to require authentication or enforce relaying rules. All in all, I think spamdyke can't help. It depends on running during the original message delivery so its messages will be sent to the remote server. Trying to move it deep inside a mail delivery system defeats most of its design. -- Sam Clippinger Faris Raouf wrote: > Hi all, > > I have a few email addresses that are not running on servers that I control. > A lot of them are getting high levels of spam sent to them (coincidentally, > mostly ones where the FROM and TO are both the same and are my email > address). > > What I want to somehow do is arrange things so that this email is magically > piped to one of the servers we have running spamdyke, in a way that appears > to spamdyke as though it came directly from the original sender. > > I initially thought of using fetchmail for this but that wouldn't work, > would it? I need to effectively strip the headers belonging to the host that > accepted the email in the first place in order to get anywhere close to what > I want, and fetchmail isn't designed for that as far as I can tell. > > So, am I stuffed or might there be some other Linux magic available > somewhere that might help me? > > Faris. > > > > > > _______________________________________________ > spamdyke-users mailing list > [email protected] > http://www.spamdyke.org/mailman/listinfo/spamdyke-users > _______________________________________________ spamdyke-users mailing list [email protected] http://www.spamdyke.org/mailman/listinfo/spamdyke-users
