procmail
[Top] [All Lists]

Re: Mail Redirection need help!

2004-11-30 02:30:29
On 30 Nov 2004 Klaus Johannes Rusch (KlausRusch(_at_)atmedia(_dot_)net) wrote:
While you check for the X-Loop header, you do not actually add an X-Loop header so the test never matches. From the description of your problem it sounds like

:0
* ^From:(_dot_)*one(_at_)yahoo(_dot_)com
* ^TO(_dot_)*two(_at_)consulting(_dot_)com
! three(_at_)bullhorn(_dot_)com

Since the OP is using qmail it is probably better (i.e., more accurate!) to use the Delivered-To header instead of procmail's ^TO macro. For example, something like this might work:

:0
* ^From:(.*\<)?one(_at_)yahoo\(_dot_)com
* ^Delivered-To:(.*\<)?two(_at_)consulting\(_dot_)com
! three(_at_)bullhorn(_dot_)com

But there are all kinds of potential problems here because (as Sean so wisely said): PROCMAIL IS NOT AN MTA!

Potential problems include:

* From: headers are easily forged so you need a better way to get the original MAIL FROM address

* Delivered-To: is easily forged and, in any case, is not necessarily the original recipient address

* Mail loops and all the other problems that can happen with forwarding.

I have some information about these issues in these sections of my Procmail Quick Start:

Important Warning: Procmail Cannot Reliably Route Users' Messages in a Virtual Domain
 <http://www.ii.com/internet/robots/procmail/qs/#virtualDomain>

 The Problems With Forwarding
 <http://www.ii.com/internet/robots/procmail/qs/#forwardingProbs>

Hope this helps,
NM

--
Infinite Ink: <http://www.ii.com/>
Reverse Spam Filtering: <http://www.ii.com/internet/messaging/spam/>
IMAP Service Providers: <http://www.ii.com/internet/messaging/imap/isps/>
All About Pine: <http://www.ii.com/internet/messaging/pine/>

____________________________________________________________
procmail mailing list   Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail

<Prev in Thread] Current Thread [Next in Thread>