On 2014/06/24 20:56, [email protected] wrote:
> The nsd rc script returns an exit code other than 1 when nsd is
> not running. A problem arises when using a configuration management
> (CM) system such as ansible to control system services. With this
> non-standard exit code, the CM system cannot determine if the service
> is running or not. The nsd rc script runs 'nsd-control status', which
> documentation states will exit 3 when nsd is not running. I'd argue that
> the rc script should catch that, and instead exit 1.
If it was just for a CM system I'd argue that it should probably be
more flexible, but it also goes against the documentation in rc.subr(8),
so I think this should be fixed.
I'll commit this soon unless there are any objections.
> Index: nsd
> ===================================================================
> RCS file: /cvs/src/etc/rc.d/nsd,v
> retrieving revision 1.7
> diff -u -p -u -r1.7 nsd
> --- nsd 7 May 2014 02:46:05 -0000 1.7
> +++ nsd 25 Jun 2014 03:53:50 -0000
> @ -25,6 +25,12 @@ rc_start() {
>
> rc_check() {
> ${daemon} ${daemon_flags} status
> + ret=$?
> + if [[ $ret != 0 ]]; then
> + return 1
> + else
> + return 0
> + fi
> }
>
> rc_reload() {
>
