Sky Cree writes:
> I,Ab??(Bm experiencing a problem with an SMF stop method on Solaris
> x86. The daemon in question is NTP. I am also experiencing the same issue
> with my SSH daemon but will use NTP for the purpose of this question. These
> are both scripts that I copied from my sparc implementation but I cannot
> figure out why SMF is having issue with the stop method since the script by
> itself stops the daemon just fine. I am by no means an expert with SMF or
> Solaris so forgive me for overlooking something simple.
>
> Executing "svcadm disable ntpd" puts it into maintenance mode.
>
> Output from ,Ab??(Bsvcs ,Ab??(Bx,Ab??(B is as follows:
>
> bash-3.2# svcs -x ntpd
> svc:/network/ntpd:default (NTPD)
> State: maintenance since Tue Jun 09 20:30:09 2009
> Reason: Method failed repeatedly.
> See: http://sun.com/msg/SMF-8000-8Q
> See: /var/svc/log/network-ntpd:default.log
> Impact: This service is not running.
>
> I have checked the web page recommended above but it is not very helpful
> as to what the underlying cause could be. Processes related to ntpd and
> ntpddate are not running after the stop method is issued.
>
> Clearing and enabling the service with svcadm starts the service without
> issue.
>
> Logfile:
> /var/svc/log/network-ntpd\:default.log
> [ Jun 9 20:30:09 Stopping because service disabled. ]
> [ Jun 9 20:30:09 Executing stop method ("/lib/svc/method/ntpd stop") ]
> [ Jun 9 20:30:09 Method "stop" exited with status 208 ]
> [ Jun 9 20:30:09 Executing stop method ("/lib/svc/method/ntpd stop") ]
> [ Jun 9 20:30:09 Method "stop" exited with status 208 ]
> [ Jun 9 20:30:09 Executing stop method ("/lib/svc/method/ntpd stop") ]
> [ Jun 9 20:30:09 Method "stop" exited with status 208 ]
> [ Jun 9 20:32:13 Leaving maintenance because clear requested. ]
> [ Jun 9 20:32:13 Enabled. ]
>
> svcs ,Ab?(B?l produces:
> bash-3.2# svcs -l ntpd
> fmri svc:/network/ntpd:default
> name NTPD
> enabled false
> state maintenance
> next_state none
> state_time Tue Jun 09 21:03:52 2009
> logfile /var/svc/log/network-ntpd:default.log
> restarter svc:/system/svc/restarter:default
> contract_id 263
> dependency require_all/none svc:/system/filesystem/local (online)
> dependency require_all/none svc:/network/loopback (online)
> dependency require_all/none svc:/network/physical (online)
> dependency require_all/restart file://localhost//usr/local/etc/ntpd.conf
> (online)
> dependency require_all/none svc:/milestone/network:default (online)
>
> The method script:
>
> #
> # Start/stop NTPD
> #
> . /lib/svc/share/smf_include.sh
> stop_ntpd()
> {
> /usr/bin/pkill ntpddate
> /usr/bin/pkill ntpd
> }
> case "$1" in
> 'start')
> if [ -x /usr/local/bin/ntpd -a -f /usr/local/etc/ntpd.conf ] ; then
> if [ -x /usr/local/bin/ntpdate ] ; then
> /usr/local/bin/ntpdate timehost ; sleep 2
> fi
> /usr/local/bin/ntpd -c /usr/local/etc/ntpd.conf &
> fi
> ;;
> 'stop')
> stop_ntpd
> ;;
> *)
> echo "Usage: $0 { start | stop }"
> exit 1
> ;;
> esac
> exit $SMF_EXIT_OK
>
> The manifest:
>
> The tags of xml file are not displayed correctly through the forum
> interface. If someone absolutely needs them I can figure out a way to
> post it.
>
> Thanks
> --
> This message posted from opensolaris.org
> _______________________________________________
> smf-discuss mailing list
> smf-discuss at opensolaris.org
What version of Solaris are you running? I've looked at Solaris 10,
OpenSolaris and Nevada systems. They all have an ntp service, but none
have an ntpd service. In addition the start script for ntp looks nothing
like the one that you have posted, and these systems do not have a stop
script.
The log file that you have posted indicates that the stop script is exiting
with 208 which does not appear to be possible with the script that you have
posted. Thus, there appears to be a disconnect.
Could you post the output of
svcprop ntpd
tom
