As it says on the tin, really. The /etc/init.d/ntpd init script does not remove the pidfile when told to 'stop'. This is possibly due to the start-stop-daemon script not having privlidges to remove a pidfile (ch)owned by ntp:ntp? Just a guess. Result is that ntpd cannot be restarted until the old pidfile is manually removed. Reproducible: Always Steps to Reproduce: 1./etc/init.d/ntpd start 2./etc/init.d/ntpd stop 3./etc/init.d/ntpd start (fails) 4.Manually delete /var/run/ntpd.pid 5./etc/init.d/ntpd start (succeeds) Actual Results: Script says ntpd cannot be started because it's already running. (it's not) Expected Results: Started ntpd.
a sort of bug #45199 duplicate. the issue affects all scripts and in general runscript.sh may be the routines have to be reviewed regards ph
you never said what version of ntpd you're using
ara@hobbes ara $ qpkg -I ntp -vv /var/db/pkg/net-misc/ntp-4.1.2/ntp-4.1.2.ebuild I did an emerge rsync && emerge -pU world before reporting the bug, and no pending updates were listed, so I presumed this problem still exists.
See bug #46058 for more details. Bug #45199 is invalid, since it is not really required to delete a PID file, if start routine was a little bit smarted.
*** Bug 46058 has been marked as a duplicate of this bug. ***
fixed in cvs
