Configuring non-delivery warning messages - timeouts and multiple warning mails?

2012-10-09 Thread Heiko Wundram
Hey! My searching through the Postfix documentation didn't turn up anything relevant, so I thought I'd ask on the list: which parameter(s) control whether (and if possible: how many/more than one?) warning messages are sent in the case that a mail can't be delivered for a specified amount of

Re: Configuring non-delivery warning messages - timeouts and multiple warning mails?

2012-10-09 Thread /dev/rob0
On Tue, Oct 09, 2012 at 10:36:57AM +0200, Heiko Wundram wrote: > My searching through the Postfix documentation didn't turn up > anything relevant, so I thought I'd ask on the list: which > parameter(s) control whether (and if possible: how many/more than > one?) warning messages are sent in the

mime_header_checks physical lines vs logical headers

2012-10-09 Thread Vincent Lefevre
Hi, With postfix 2.9.3-2~12.04.3 under Ubuntu: The header_checks(5) man page says: Note: message headers are examined one logical header at a time, even when a message header spans multiple lines. Body lines are always examined one line at a time. But after doing various tests on mime

Re: mime_header_checks physical lines vs logical headers

2012-10-09 Thread Wietse Venema
Vincent Lefevre: > Hi, > > With postfix 2.9.3-2~12.04.3 under Ubuntu: > > The header_checks(5) man page says: > >Note: message headers are examined one logical header at a time, >even when a message header spans multiple lines. Body lines are >always examined one line at a time. >

Re: mime_header_checks physical lines vs logical headers

2012-10-09 Thread Vincent Lefevre
On 2012-10-09 09:47:08 -0400, Wietse Venema wrote: > Vincent Lefevre: > > Hi, > > > > With postfix 2.9.3-2~12.04.3 under Ubuntu: > > > > The header_checks(5) man page says: > > > >Note: message headers are examined one logical header at a time, > >even when a message header spans multipl

Re: mime_header_checks physical lines vs logical headers

2012-10-09 Thread Wietse Venema
Vincent Lefevre: > > From a@b.c Tue Oct 9 16:16:44 2012 > Date: Tue, 9 Oct 2012 16:16:40 +0200 ... > $ postmap -m -q - pcre:mime_header_checks.pcre < msg >From SPACE is not a header line, according to RFC 5322, "Internet Message Format"

Re: mime_header_checks physical lines vs logical headers

2012-10-09 Thread Vincent Lefevre
On 2012-10-09 10:51:31 -0400, Wietse Venema wrote: > Vincent Lefevre: > > > > From a@b.c Tue Oct 9 16:16:44 2012 > > Date: Tue, 9 Oct 2012 16:16:40 +0200 > ... > > $ postmap -m -q - pcre:mime_header_checks.pcre < msg > > From SPACE is n

Re: mime_header_checks physical lines vs logical headers

2012-10-09 Thread Viktor Dukhovni
On Tue, Oct 09, 2012 at 05:32:13PM +0200, Vincent Lefevre wrote: > Do you mean that messages must completely follow RFC 5322, > i.e. with each line ending with CRLF (not just LF)? No, UNIX line endings are accepted with sendmail(1), postmap(1), ... since that's how messages are stored on Unix. O

Re: mime_header_checks physical lines vs logical headers

2012-10-09 Thread Vincent Lefevre
On 2012-10-09 15:52:36 +, Viktor Dukhovni wrote: > The postmap(1) utility feeds the MIME parser exactly what you ask > it to process. I think this is best, it is a testing tool, not a > message normalizer. OK, but wouldn't an error message in case of invalid message be a nice feature? > > Any

Re: mime_header_checks physical lines vs logical headers

2012-10-09 Thread Wietse Venema
As documented the postmap "-m" option is defined only with "-h" or "-b". If you use this option otherwise, then what happens is undefined. For example, daemons may fly out your nose. Wietse

Is postscreen really this good?

2012-10-09 Thread The Stovebolt Geek
I've been running postfix with policyd-weight and spamassassin for years on a small hobby domain that I manage. I usually have a few hundred spam messages in the spam folder after a few days. Recently I found out about postscreen on this list. After reading about it, I implemented it in pret