[Rspamd-Users] complete spam analysis in mail header [off topic]

[Gerald Galster]

This thread was about extended_spam_headers.
Next time create a new thread please as fixing documentation
is another issue.


>>>> https://rspamd.com/doc/modules/milter_headers.html#scan-results-exposure-prevention
>>> You're right, it does not.
> 
> small but imho important error in this link
> 
> # ==========================================================================
> # service type  private unpriv  chroot  wakeup  maxproc command + args
> #               (yes)   (yes)   (no)    (never) (100)
> # ==========================================================================
> smtp      inet  n       -       n       -       1       postscreen
>  -o smtpd_discard_ehlo_keyword_address_maps=cidr:$config_directory/esmtp_access
> 
> fail is that postscreen is not smtpd_

According to https://www.postfix.org/master.5.html

  Command name + arguments
  ...
  The following command-line options have the same effect for all
  daemon programs:
  ...
  -o name=value (short form)
  Override the named main.cf configuration parameter.

this will override main.cf's smtpd_discard_ehlo_keyword_address_maps.


According to https://www.postfix.org/postconf.5.html#postscreen_discard_ehlo_keyword_address_maps

  postscreen_discard_ehlo_keyword_address_maps (default: $smtpd_discard_ehlo_keyword_address_maps)

postscreen uses that smtpd_'s value as default.

So while the naming could be more direct, I would not consider this a fail.


> so this -o line is better in the pass type

As postscreen is offering that functionality I don't see why it
should not be used at that stage.

> who will fix this error ?

https://github.com/rspamd/rspamd.com/blob/master/doc/modules/milter_headers.md

You could try to create an issue/patch/pull-request.

Best regards,
Gerald