Is there no way to somehow rewrite it instead of removing it fully ? I know its better to fix GroupWise, I've been *trying* to get it fixed really hard for a few weeks now, but trying to get Novell to move a little is like trying to fight a major house fire with just a supersoaker .. I need a temporary solution because people are starting to get really annoyed by their e-mail not getting where it should be. The ISP in question does not give a valid error and e-mails just vanish into thin air when they're sent there. - Arjan
>>> Op 12-11-2009 om 14:40 is door Noel Jones <njo...@megan.vbhcs.org> >>> geschreven: On 11/12/2009 3:50 AM, Arjan Melein wrote: > Hello, > > I am currently having an issue where our enterprise mail system is > bugged(groupwise8sp1) and for some unknown reason, even for Novell, it is > adding ";1:1" to the end of the TO: line and this is causing e-mails to > bounce with a certain ISP who has very strict header checking. > I'm using postfix as the MTA for our MailScanner setup so I'm currently > trying to figure out if and how it's possible to somehow strip the ;1:1 from > outgoing e-mails. > I ran into the REPLACE function for the header checks but I can't seem to > figure out how to properly use it (still working on my regexp knowledge) > Currently the header_checks only holds '/^Received:/ HOLD' so the scanner can > pick it up. > > Is there anyone who can give me some pointers on if this is possible and how ? > > My only other option is to use a 2nd hold directory and write some kind of > script to pick up e-mails and rewrite them before handing them to the scanner. > > - > Arjan > You'll need to fix this on the groupwise server. RFC822 header syntax is quite complex. I don't think it's possible to create a regular expression that correctly handles and fixes every case, especially with multiple addresses in a header. If you must fix this in postfix, your best choice is probably to just remove the header with IGNORE. # header_checks /^(To|CC): .*;1:1/ IGNORE Postfix will then insert the undisclosed_recipients_header in the message. This is a very ugly solution. Better to fix groupwise. -- Noel Jones
