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

Re: Per recipient restrictions

Expand Messages
  • Mike Morris
    ... For what it s worth I decided to go with the PCRE approach for now to get this up and running. When I have more time I ll probably look at doing this with
    Message 1 of 8 , Feb 28, 2008
    • 0 Attachment
      On 02/27/2008 08:19 AM, Mike Morris wrote:
      > Hi all,
      >
      > I would like to work on a configuration that allows individual
      > recipients to opt out of any combination of a number of restrictions. I
      > am having difficulty determining how to best do this within Postfix. I
      > have already created the interface which allows users to select up to
      > six restrictions to opt out of. It stores a row in a MySQL table with
      > the recipient address and a boolean value for each of the six checks
      > that the user can choose to opt out of.
      >
      > My initial idea was to generate CDB maps based on the information from
      > the MySQL table, one map for each of the restrictions a recipient is
      > allowed to opt out of. Each map would return a 'DUNNO' action for email
      > addresses that chose to opt out for that one check and return a
      > restriction class that enabled that restriction for everyone else. These
      > maps would then replace the respective restrictions in the appropriate
      > smtpd_*_restrictions stanza. The problem with this is, unless I am
      > missing something, there is no way to return a default action to call a
      > restriction class for hash, CDB, etc. A PCRE map could be used to do
      > what I wanted. But that got me thinking that perhaps I was going about
      > this the wrong way.
      >
      > Doubtless this has been discussed before. The closest I can find in the
      > archives is this one:
      >
      > http://groups.google.com/group/list.postfix.users/browse_thread/thread/2f452e71834581bf/65a4ad03b74a0b42?lnk=gst&q=per+recipient+restrictions#
      >
      >
      > There was a suggestion in that thread to use regex/PCRE as well. Is
      > doing so a decent enough approach? Or is there a better way to
      > accomplish what I am trying to do?
      >
      > If more information is required, let me know. Thanks.


      For what it's worth I decided to go with the PCRE approach for now to
      get this up and running. When I have more time I'll probably look at
      doing this with a policy daemon, a la Postfilter.

      Thanks for the input.
    Your message has been successfully submitted and would be delivered to recipients shortly.