Bug#320483: exim4-base: Exim4 init script still broken in 4.63-7

[Marc Haber]

tags #320483 - patch
tags #320483 wontfix
thanks

On Fri, Oct 27, 2006 at 12:34:05PM -0500, Donald King wrote:
> 1. A combined queuerunner/listener daemon on port 25 (handles modern TLS)
> 2. A standalone inetd listener on port 465 (handles legacy SSL)
> 3. A standalone inetd listener on port 587 (handles RFC 2476 submission)
> 
> As the ports handled by 2 and 3 are rarely used in my setup, I decided
> to split them off from the main daemon.

I don't think that it is a good idea to split this off. Can you please
explain why you think that it is a good idea?

> However, I have to note that, although I can work around the problem in
> my own setup by simply deleting the inetd listeners and having the main
> daemon handle all ports,

Yes, that is my recommendation.

>  there are other configurations that are more broken.  Most notably,
>  if someone wants to use an inetd listener on port 25 with
>  QUEUERUNNER=queueonly, they will find that the queuerunner daemon
>  never runs, due to the troublesome code on lines 60-65 of
>  /etc/init.d/exim4.

Using inetd is not supported. /etc/init.d/exim4 is a dpkg-conffile,
please remove the code yourself.

> Theoretically, the inetd check should be conditional on the value of
> $QUEUERUNNER, but I'd argue that it's the duty of the users to edit
> the value of $QUEUERUNNER themselves in /etc/default/exim4, and that the
> inetd check is redundant.  If the users really want the behavior forced on
> them by the inetd check, they should specify QUEUERUNNER=nodaemon.

I am not going to make that change to the Debian exim4 packages. inetd
is not supported.

Greetings
Marc

-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Mannheim, Germany  |  lose things."    Winona Ryder | Fon: *49 621 72739834
Nordisch by Nature |  How to make an American Quilt | Fax: *49 621 72739835