I've currently got a Scheduleable - ie a class that extends org.jboss.varia.scheduler.Schedulable and it is deployed in a SAR with the following sort of jboss-service.xml
<?xml version="1.0" encoding="UTF-8"?>
<server>
<classpath codebase="lib" archives="scheduler-plugin.jar"/>
<mbean code="org.jboss.varia.scheduler.Scheduler"
name="mydomain:service=MySchedulabler">
<attribute name="StartAtStartup">true</attribute>
<attribute name="SchedulableClass">
my.example.SchedulableClass
</attribute>
<attribute name="InitialStartDate">NOW</attribute>
<attribute name="InitialRepetitions">-1</attribute>
<attribute name="SchedulePeriod">60000</attribute>
<attribute name="SchedulableArguments">
password, password
</attribute>
<attribute name="SchedulableArgumentTypes">
java.lang.String, java.lang.String
</attribute>
<attribute name="TimerName">mydomain:service=MyTimer</attribute>
<depends>
<mbean code="javax.management.timer.Timer"
name="mydomain:service=MyTimer" />
</depends>
</mbean>
</server>
But now I want to deploy this so that it is a HASingleton - ie I only want this schedulable to be kicked off on one node in a cluster, not every node.
What is the easiest way for me to convert from the configuration I have to a singleton?
I've seen some documentation around (JBoss AS Clustering ed7) that demonstrates a singleton schedulable, but only by using a defined MBean as the target action of a org.jboss.varia.scheduler.SingleScheduleProvider MBean. Is the easiest way to go from what I have now to a singleton schedulable define my task (class that currently extends Schedulable) as an MBean and then use the org.jboss.varia.scheduler.SingleScheduleProvider?
I'm just a bit confused about the right way to configure this...
Thanks,
Dion.