Systemd service life cycle

[Ralf Mardorf]

On Sun, 10 Apr 2016 08:31:38 +0100, Colin Law wrote:
>OK, but I still don't understand why oneshot is more appropriate than
>forking.  With oneshot would the service notice if the tunnel failed
>for example, and restart the service?

I don't know, for me all English systemd explanations are vague and
there seem to be no German systemd explanations available that
unambiguous explain units. I would like to read a clear explanation in
my native language, perhaps I can't find such an explanation in my
native language, because systemd is fishy and units don't allow clear
explanations.

When I used a PPPoE script to get the interface up and to pon to
establish an ADSL connection, the connection was automatically new
established, after it was interrupted every 24 hours. I don't know if
pppd or systemd reconnected. Assumed pppd shouldn't do this, then
systemd must have done this.