+++ This bug was initially created as a clone of Bug #522111 +++ Description of problem: Testing init script according to guidelines at [1],[2],[3] shows that radiusd does not recognize 'force-reload' action, as required by LSB. Version-Release number of selected component (if applicable): freeradius RPM version: 2.1.6-6.fc12 Actual results: # service radiusd force-reload ; echo $? Usage: /etc/init.d/radiusd {start|stop|status|restart|reload|condrestart} 1 # service radiusd hello; echo $? Usage: /etc/init.d/radiusd {start|stop|status|restart|reload|condrestart} 1 Also it would be nice to return 3, rather than 1 (unimplemented feature). Expected results: recognize the force-reload action and according to [1] do: * force-reload: "force-reload cause the configuration to be reloaded if the service supports this, otherwise restart the service if it is running." Additional info: [1] - https://fedoraproject.org/wiki/FCNewInit/Initscripts [2] - https://fedoraproject.org/wiki/Packaging/SysVInitScript [3] - http://refspecs.linux-foundation.org/LSB_3.2.0/LSB-Core-generic/LSB-Core-generic/iniscrptact.html --- Additional comment from ebenes on 2009-09-09 10:10:08 EDT --- Ohh! Having another look at the init script, this is REALLY a bad practice and should be fixed in /etc/rc.d/init.d/radiusd: [ -f $RADIUSD ] || exit 0 [ -f $CONFIG ] || exit 0 Please make it conform to following guidelines, thanks. [1] - https://fedoraproject.org/wiki/FCNewInit/Initscripts [2] - https://fedoraproject.org/wiki/Packaging/SysVInitScript [3] - http://refspecs.linux-foundation.org/LSB_3.2.0/LSB-Core-generic/LSB-Core-generic/iniscrptact.html
I can see changing all the initscripts in Fedora to be compliant, but I seriously question this in RHEL because of our guarantee of stability in RHEL. Do you have a documented mandate from RHEL engineering and product management this should be done? If so please provide a link to it.
(In reply to comment #1) > I can see changing all the initscripts in Fedora to be compliant, but I > seriously question this in RHEL because of our guarantee of stability in RHEL. > Do you have a documented mandate from RHEL engineering and product management > this should be done? Unfortunately, I don't. Adding David to CC, as he might have something to add to this conversation.
Closing this as WONTFIX because I'm not comfortable in changing the behavior (including potentially returning different status codes) of an initscript in a RHEL update. That seems like asking for trouble when the motivation is not a customer issue, but rather compliance requirements in a totally different product. I will of course update the Fedora initscript in preparation for RHEL 6.