tobias.burnus at physik.fu-berlin.de
Thu Aug 14 07:46:47 PDT 2003
On Tue, 12 Aug 2003, Tobias Burnus wrote:
> > The result is: If a service supports SIGHUP and is not running, a
> > "force-reload" will give you an error. If a service does not support
> > SIGHUP, a "force-reload" will start the service, if it is not running
> > before.
> > So I suggest do change "force-reload" to:
> > "cause the configuration to be reloaded if the service supports this,
> > otherwise restart the service if it is running".
> - Is the exit status code clear for force-reload if the service is not
> running? Given the discussion concerning "try-restart" it might not be
> the case. The attached patch adds
> "- running "reload" on a service already stopped or not running",
> "- running "force-reload" on a service already stopped or not running"
> to the list of actions which shall be considered successful.
After looking at several init scripts (RedHat, SuSE, Debian) and some
[minor] discussion I tend to go back to the previous behaviour:
Return a not-running 7 and not a sucess 0.
(Otherwise the not-running 7 doesn't make any sense and providing more
details rather than less is not so bad either.)
--- initactions.sgml 12 Aug 2003 21:31:28 -0000 1.10
+++ initactions.sgml 14 Aug 2003 14:41:29 -0000
@@ -93,6 +92,0 @@
-running "reload" on a service already stopped or not running
-running "force-reload" on a service already stopped or not running
What are the opinions of the others?
More information about the lsb-discuss