[milters] Archive

Lists Index Date Thread Search

Article: 955
From: Anthony Howe
Date: 2006-06-05 16:23:09 -0400
Subject: Re: Question regarding the merits of milter flag to "Temporarily

Removal...........: milters-request@milter.info?subject=remove
More information..: http://www.milter.info/#Support

Taylor, Grant wrote:
> I am considering configuring Sendmail to Temp Fail for many (if not all) 
> of my milters if the milter in question is not available.  However 
> before I do such I was wondering what other people's opinions on this 
> are.  I am VERY impressed with what the milters from Anthony / SnertSoft 
> will do and how well they prevent a lot of spam.  I'm impressed so much 
> so that I am almost to the point that I do not want to accept email if I 
> can not pass the email through some of my filters.  Namely, I do not 
> want to accept any email, but rather Temp Fail if I can not perform 
> sender verification and virus scan the email.

All my milters document the F=T and F=R flags concerning 
INPUT_MAIL_FILTER() settings. In the early days of milter-sender I even 
suggested that F=T be used in the sample.mc configuration. I later 
removed it for reasons that others have mentioned already (I think its 
even mentioned in the 0.x change logs some place).

I'm at a loss for adequate words of thanks for the vote of confidence so 
many of you express about my work. But please remember the early days of 
milter-sender or milter-spamc or any other milter I've written to date. 
They are not always perfect from day one and sometimes when I make grand 
sweeping revisions I still make mistakes, even after almost four years 
of milter development and many many more years as a C programmer.

Mail is important to all of us and spam/viruses our bane. As postmasters 
we walk a fine line choosing and implementing policy on behalf of our 
users that we hope is the best for all. My milters offer a wide variety 
of strategies and choices, but

	no one milter is so important that mail should be
	held up or refused in the event that a milter fails.

I do NOT recommend the use of F=T or F=R. Its far to easy to shoot 
yourself in both feet. There are too many possible points of failure, 
most of them human caused, to account for.

Anthony C Howe          Skype: SirWumpus                    SnertSoft
+33 6 11 89 73 78         AIM: SirWumpus    Sendmail Milter Solutions
http://www.snert.com/     ICQ: 7116561

Lists Index Date Thread Search