Description of problem:
vhostmd-0.4-2.el5_3
While doing some RAS test cases I noticed a wrong behaviour of the vhostmd
startup script '/etc/init.d/vhostmd'.

# kill -11 vhostmd
# service vhostmd status
vhostmd dead but pid file exists
# service vhostmd stop
# service vhostmd start
Starting vhostmd:                                          [  OK  ]
# service vhostmd status
vhostmd dead but subsys locked

I noticed that the ' /etc/init.d/vhostmd' stop() function has logic to remove
the PID file in /var/run/.
However it is not executed during stop action, because an addtional status
check is done.

stop)
        rh_status_q || exit 0
        $1
        ;;

We have RAS logic that relies on the correct behaviour of the service startup
scripts.


Version-Release number of selected component (if applicable):vhostmd-0.4-2.el5_3

How reproducible:
start vmhost

Steps to Reproduce:
1.
2.
3.
  
Actual results:
see above


Expected results:
vmhost should delet PID file in /var/run/

Additional info:
IBM Tester: thomas.weber.com
IBM Project Manager: ralf.thelen.com (opening this bug)

We think, we know what needs to be done in vhostmd. If you need this technical information let Ralf Thelen (know).