Re: Syntax question: smtp-policy.spam

[Matti Aarnio <mea@nic.funet.fi>]

On Thu, Jun 13, 2002 at 11:20:46AM +0200, Arnt Gulbrandsen wrote:
> Matti Aarnio <mea@nic.funet.fi>
> >   If the program gives unexpected outputs (like some PERL syntax error),
> >   the policy report is considered to be "-1 550 5.7.1 ...",  which
> >   might not be quite wanted -- especially when the reason is a perl
> >   goofup..
> 
> Shouldn't the smtpserver give a 4xx code and syslog an error if the
> program behaves unexpectedly?

  Like I said, it is "fragile".  I need to have a moment to
  re-engineer it to have sensible behaviour in absense of
  proper responses.  (Response lacking leading signed integer,
  or possibly to tighten the protocol to always demand the
  initial triplet, e.g.:  "-1 550 5.7.1 blabla...", and if
  that isn't gotten, then reply with "-1 450 4.7.1 blabla.."..)

  Of course, somebody else may also lend a helpfull hand, and
  make such re-engineering, and I will apply the patch. ( ;-) )

  Also, a collegue of mine said that "550" isn't proper response
  for the end of the DATA in all conditions.  That needs checking,
  and fixing.  It isn't proper to reply with bogus codes; while
  most UNIX systems look at the first character to classify the
  response, way too many M$ platform systems look into the entire
  code, and get mighty upset (and do unexpected) if they don't see
  codes what they expect per RFC 821..

> --Arnt

-- 
/Matti Aarnio	<mea@nic.funet.fi>
-
To unsubscribe from this list: send the line "unsubscribe zmailer" in
the body of a message to majordomo@nic.funet.fi