Loading ...
Sorry, an error occurred while loading the content.

Re: postfix and/or amavisd-new duplicating sender_bcc's?

Expand Messages
  • Troy Piggins
    ... Understood. Thanks for the clarification, Jeroen. -- Troy Piggins
    Message 1 of 5 , Sep 28 3:21 PM
      * Jeroen Geilman wrote :
      >* On 2011-09-28 11:57, Troy Piggins wrote:
      >>* Wietse Venema wrote :
      >>>* Troy Piggins:
      >>>> I'm using sender_bcc maps to archive/backup sent mail. Config
      >>>> extracts below. It does work, but is creating duplicates when mail
      >>>> is sent from clients other than localhost. IE if I send mail using
      >>> See http://www.postfix.org/FILTER_README.html, and look for the
      >>> examples with receive_override_options. This workaround is needed
      >>> before or after the filter.
      >> Thankyou! Had used this tutorial
      >> https://help.ubuntu.com/community/PostfixAmavisNew although most of
      >> the receive_override_options listed in that tutorial were already in
      >> the default Ubuntu package /etc/postfix/master.cf. However after
      >> reading the link you pointed me to, I added
      >> receive_override_options = no_address_mappings
      >> just after the content_filter line in /etc/postfix/main.cf,
      > If master.cf contains no empty overrides, this means you lose ALL aliasing.
      > You need to specify this option either on your main smtpd(8)
      > listener, meaning the original recipients will be sent to the
      > content_filter, or on your content_filter re-injection listener,
      > which means only final (aliased) recipients will be sent through the
      > content_filter.
      > Which one you want depends on your requirements.
      > Similarly, you normally don't want to set content_filter in main.cf;
      > instead, specify it directly on your main smtpd(8) listener.
      > This means it will not be applied on submission, nor, indeed, on
      > your re-injection listener - which creates an infinite loop.
      > Many distros hack around this by specifying an empty content_filter=
      > on the re-injection listener, but I consider this a crock. Don't
      > specify it globally in the first place.

      Understood. Thanks for the clarification, Jeroen.

      Troy Piggins
    Your message has been successfully submitted and would be delivered to recipients shortly.