synchronisation of tasks/jobs how to

[msduk]

I am new to spring batch and quartz but have a few examples working well but I cant see how to proceed more generally with the whole set of requirements. I think I might be missing the interplay between Quartz Triggers and Spring Batch.

Requirements -:
I have a set of tasks to run sequentially every night.
I also have a polling task that needs to run every minute or so.
These must not run in parallel.

The question is how do I achieve this? I tried two separate quartz triggers but I cant see how to ensure only one runs at a time. Of course the answer might be to synchronise elsewhere.

Regards

[Benny Jiang]

In Spring Batch, the jobLauncher in default is using synchronized taskExecutor, which means the jobs you run with this launcher will run in sequential.

[msduk]

In Spring Batch, the jobLauncher in default is using synchronized taskExecutor, which means the jobs you run with this launcher will run in sequential.

Hi thanks. This is what I thought would happen, but in testing I fire up to 10 jobs (I believe this is the thread limit).

<bean class="org.springframework.scheduling.quartz.SchedulerFactoryBean">
		<property name="triggers">
			<list>
				<bean id="repeat" class="org.springframework.scheduling.quartz.SimpleTriggerBean">
					<property name="jobDetail" ref="repeatDetail" />
					<property name="startDelay" value="5001" />
					<property name="repeatInterval" value="100"/>
				</bean>
			</list>
		</property>
	</bean>


<bean id="minuteRepeatDetail" class="org.springframework.scheduling.quartz.JobDetailBean">
		<property name="jobClass" value="foo.JobLauncherDetails" />
		<property name="group" value="repeat" />
		<property name="jobDataAsMap">
			<map>
				<entry key="jobName" value="minuteRepeatJob"/>
				<entry key="jobLocator" value-ref="jobRegistry"/>
				<entry key="jobLauncher" value-ref="jobLauncher"/>
			</map>
		</property>
	</bean>
	
	<batch:job id="minuteRepeatJob" restartable="true" incrementer="runIdIncrementer">
		<batch:step id="harvestSchedule" next="explodeXMLRecords">
			<batch:tasklet ref="harvestScheduler" />
		</batch:step>
		<batch:step id="explodeXMLRecords">
			<batch:tasklet>
				<batch:chunk reader="xmlExploderItemReader" processor="xmlExploderItemProcessor" writer="xmlExploderItemWriter" commit-interval="1000" />
			</batch:tasklet>
		</batch:step>
	</batch:job>

<bean id="jobLauncher" class="org.springframework.batch.core.launch.support.SimpleJobLauncher">
		<property name="jobRepository" ref="jobRepository" />
		<property name="taskExecutor">
			<bean class="org.springframework.core.task.SyncTaskExecutor" />
		</property>
	</bean>

10:45:28,718 [INFO ] org.springframework.batch.core.launch.support.Simp leJobLauncher [run():118] - Job: [FlowJob: [name=minuteRepeatJob]] launched with the following parameters: [{run.id=1281087928093}]
10:45:28,718 [INFO ] org.springframework.batch.core.launch.support.Simp leJobLauncher [run():118] - Job: [FlowJob: [name=minuteRepeatJob]] launched with the following parameters: [{run.id=1281087927343}]
10:45:28,765 [INFO ] org.springframework.batch.core.launch.support.Simp leJobLauncher [run():118] - Job: [FlowJob: [name=minuteRepeatJob]] launched with the following parameters: [{run.id=1281087928687}]

Is this maybe something to do with the incrementor I am using or is the fact I trigger using quartz the culprit?

[msduk]

In Spring Batch, the jobLauncher in default is using synchronized taskExecutor, which means the jobs you run with this launcher will run in sequential.

Ah it seems quartz uses a threaded pool as a default. Adding this fixed my problems. I guess it is a bit unfair to expect jobs to be called synchronously despite several calling threads. Thanks.

<bean class="org.springframework.scheduling.quartz.SchedulerFactoryBean"><property name="taskExecutor">
<bean class="org.springframework.core.task.SyncTaskExecutor" />
</property>
</bean>