[CentOS] Apache wakes-up inactive Exim

Mon Sep 29 00:27:49 UTC 2014
Les Mikesell <lesmikesell at gmail.com>

On Sun, Sep 28, 2014 at 2:01 PM, Always Learning <centos at u62.u22.net> wrote:
> Had a surprising event on C 6.5.
> Exim was the only MTA installed. It was partially configured (with ACL,
> Router, Transport) and definitely not running.
> I was remotely testing a web page. A web page error condition invoked
> the embedded PHP mail() command.
> To my astonishment something in Centos woke-up Exim. Exim sent the email
> and then became inactive again. The Exim logs does not show any start-up
> lines, just
> 1. input from Apache.
> 2. output to remote server.
> 3. 'completed'.
> Hours later Logwatch, not yet customised, also caused inactive Exim to
> send an email (which got rejected by Exim because it was to local user
> 0).
> What causes Centos to temporarily activate in-active (meaning
> non-running) Exim ?

Applications often send mail by piping to the command 'sendmail' which
is enough of a standard that other MTAs normally offer a compatible
command.  With 'real' sendmail you can configure it to either deliver
in the current process or to queue for subsequent delivery by the
daemon. Not sure how the others handle it, but yours apparently
completes the delivery.

   Les Mikesell
    lesmikesell at gmail.com