Can not start/enable the service

I created and imported a service in the SMF. It works fine since days. But today I can't get it start. I delete this service and re-imprt it. But fail to start it. In the log I can see fowllings:

...
[ Jul 11 10:39:41 Executing start method ("/lib/svc/method/wsvws start") ]
Starting jboss-as ... cleaning the cache ...
/opt/jboss-apps/wsvws/tmp not exist
/opt/jboss-apps/wsvws/data not exist
skipping       31 lines of /var/log/jboss-apps/wsvws/server.log
[OK]
[ Jul 11 10:39:43 Method "start" exited with status 0 ]
su: Sorry
[ Jul 11 10:39:45 Stopping because all processes in service exited. ]
[ Jul 11 10:39:45 Executing stop method ("/lib/svc/method/wsvws stop") ]
Stopping jboss-as... Jboss is stopped.
[ Jul 11 10:39:45 Method "stop" exited with status 0 ]
[ Jul 11 10:39:45 Restarting too quickly, changing state to maintenance ]
...

I wonder why.

Originally Posted by thomas2004ch I created and imported a service in the SMF. It works fine since days. But today I can't get it start. I delete this service and re-imprt it. But fail to start it. In the log I can see fowllings: ... [ Jul 11 10:39:41 Executing start method ("/lib/svc/method/wsvws start") ] Starting jboss-as ... cleaning the cache ... /opt/jboss-apps/wsvws/tmp not exist /opt/jboss-apps/wsvws/data not exist skipping 31 lines of /var/log/jboss-apps/wsvws/server.log [OK] [ Jul 11 10:39:43 Method "start" exited with status 0 ] su: Sorry [ Jul 11 10:39:45 Stopping because all processes in service exited. ] [ Jul 11 10:39:45 Executing stop method ("/lib/svc/method/wsvws stop") ] Stopping jboss-as... Jboss is stopped. [ Jul 11 10:39:45 Method "stop" exited with status 0 ] [ Jul 11 10:39:45 Restarting too quickly, changing state to maintenance ] I wonder why.

Well, we can't tell you, since you don't provide any useful details. All you've told us is that you created 'a service in the SMF'. It was working "since days", and now it's not.

Short answer: something changed, so change it back to the working state. Otherwise, post details, and we can try to help you.

"su: Sorry" is dubious here:

[ Jul 11 10:39:43 Method "start" exited with status 0 ]
su: Sorry

Originally Posted by TB0ne Short answer: something changed, so change it back to the working state. Otherwise, post details, and we can try to help you.

I haven't changed anything. But I am not sure if someone else changed something on the server machine or not.

I remember what I did yesterday is: I set the 'svcadm -t disable service-name' because I want to test if the service can be started by rebooting. But since I can't reboot the machine at this moment - the other colleague are working at the machine. I try to enable it again with 'svcadm enable service-name'. Since here the service can't be started.

What details do you want? I will attach the svcs -x and svcs -l here.

root@S001AP61-TEST (vws-tst1) # svcs -xv wsvws:default
svc:/application/jboss/wsvws:default (wsvws service)
 State: maintenance since Fri Jul 12 07:33:01 2013
Reason: Restarting too quickly.
   See: http://sun.com/msg/SMF-8000-L5
   See: /var/svc/log/application-jboss-wsvws:default.log
Impact: This service is not running.

root@S001AP61-TEST (vws-tst1) # svcs -l wsvws:default
fmri         svc:/application/jboss/wsvws:default
name         wsvws service
enabled      true
state        maintenance
next_state   none
state_time   Fri Jul 12 07:33:01 2013
logfile      /var/svc/log/application-jboss-wsvws:default.log
restarter    svc:/system/svc/restarter:default
contract_id
root@S001AP61-TEST (vws-tst1) #

From the error id in svcs -x 'http://sun.com/msg/SMF-8000-L5', the description is: The service terminated and was restarted more frequently than normal. The Solaris Service Manager placed the service into maintaenance mode instead of restarting it again.

I don't understand what here "restarted more frequently than normal" means.

And here is the manifest to the service:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE service_bundle SYSTEM "/usr/share/lib/xml/dtd/service_bundle.dtd.1">
<service_bundle type='manifest' name='wsvws'>
	<service name='application/jboss/wsvws' type='service' version='1'>
		<create_default_instance enabled='false' />
		<single_instance />
		<method_context>
			<method_credential user='wsvws' group='jboss' />
		</method_context>
		<exec_method type='method' name='start' exec='/lib/svc/method/wsvws start'
			timeout_seconds='60' />
		<exec_method type='method' name='stop' exec='/lib/svc/method/wsvws stop'
			timeout_seconds='60' />
		<exec_method type='method' name='restart'
			exec='/lib/svc/method/wsvws restart' timeout_seconds='60' />
		<stability value='Evolving' />
		<template>
			<common_name>
				<loctext xml:lang='C'>wsvws service</loctext>
			</common_name>
		</template>
		
	</service>
</service_bundle>

Originally Posted by jlliagre "su: Sorry" is dubious here: [ Jul 11 10:39:43 Method "start" exited with status 0 ] su: Sorry

There is embeded command 'su' in the init-script. I've taken out this and there is no any message of "su: Sorry". But the other error message are the same as before. This means, the 'su' is not the reason.

I do further research and find interessting thing.

First here i smy init-script. I taked out a lot of codes and left the simple codes as follow:

#!/usr/bin/sh
#

start() {
	echo Starting...
}

stop() {
	echo stopping...
}

case "$1" in
  start)
      start
      sleep 2
      ;;
  stop)
      stop
      sleep 2
      ;;
  restart)
      $0 stop
      $0 start
      ;;
  status)
      status
      ;;
  cleanup)
      cleanup
      ;;
  *)
      ## If no parameters are given, print which are avaiable.
      echo "Usage: $0 {start|stop|status|restart|reload|cleanup}"
      exit 1
      ;;
esac

If I run this init-script manuelly there is no any problem.

But when I use the SMF to controll this script and want to start it with svcadm enable service-name, I can see in the log file that the service will be started and stopped 5 times and at last it changed to maintenance mode. This means, there is a loop in the SMF. This is why I got the error message "http://sun.com/msg/SMF-8000-L5".

But why?

The problem is solved. See here:
https://forums.oracle.com/thread/2560145