Re: sanity check in init.d

[martin@internet-treff.uni-koeln.de (Martin Bialasinski)]

>> "MB" == Martin Bialasinski <martin@internet-treff.uni-koeln.de> writes:

>> "SG" == Steve Greenland <stevegr@debian.org> writes:

SG> Daemons *should* (I think) log their startup (or failure to do so)
SG> in /var/log/daemon.log (or whereever daemon.* is configured to
SG> log).

MB> Yes, Shaul also pointed this out. I had the analogy of usual
MB> commands in mind (no output = no error).

Another thing comes to my mind (as I took another look at the init.d
scripts):

They already report different kind of errors on start-up.

if [ ! -e /lib/modules/`uname -r`/fs/autofs.o ]; then
  echo "Error: autofs support not available."
  exit 0
fi

if [ "$IFACE" = "none" ] ; then
  echo "DHCP client daemon is disabled."
  exit
fi

if [ "$(id -u)" != "0" ]
then
  echo "You must be root to start, stop or restart gpm."
  exit 1
fi

if [ "`whoami`" != "root" ]
then
  echo "You must be root to start, stop or restart mysqld."
  exit 1
fi

etc.

These messages are a good thing IMHO. But we have to be
consistent. Either we don´t show any messages on failure or we show
messages on any failure (my preference). 

With the current situation, the user can´t be sure about success or
failure. Some give error messages, some not... This is not good.

We should issue such messages to STDOUT. Messages by the daemon (*after* 
it got started by the start-stop-daemon) should go to syslog.

Opinions?

Ciao,
	Martin