Quartz是OpenSymphony开源组织在任务调度领域的一个开源项目,完全基于Java实现。Quartz具备以下特点:
- 强大的调度功能,例如支持丰富多样的调度方法,可以满足各种常规及特殊需求
- 灵活的应用方式,例如支持任务和调度的多种组合方式,支持调度数据的多种存储方式;
- 集群能力
- 能够很轻易的就与spring集成在一起
一、基础配置
1. 引入jar包
org.quartz-scheduler
quartz
2.2.3
org.quartz-scheduler
quartz-jobs
2.2.3
2. xml配置
使用quartz需要配置job、trigger、scheduler,其中:
job:任务的执行类,需要在xml中指明名称、组别和类
trigger:任务的触发器,需要指明名称、组别以及关联的job。一个触发器只能对应一个job;触发器有两种,分别为CronTriggerFactoryBean和SimpleTriggerFactoryBean,前者支持cron表达式,后者只支持一些简单的配置。
scheduler:调度器,需要将trigger配置在scheduler的triggers中,可以配置多个。这里自定义实现了jobFactory,可以在job中自动注入spring bean;applicationContextSchedulerContextKey属性用于在job中获取spring 的上下文
3. 代码清单
3.1. 定义SpringBeanJobFactory
package com.sawyer.job;
import org.quartz.spi.TriggerFiredBundle;
import org.springframework.beans.factory.config.AutowireCapableBeanFactory;
import org.springframework.context.ApplicationContext;
import org.springframework.context.ApplicationContextAware;
import org.springframework.scheduling.quartz.SpringBeanJobFactory;
public class AutowiringSpringBeanJobFactory extends SpringBeanJobFactory implements ApplicationContextAware {
private transient AutowireCapableBeanFactory beanFactory;
public void setApplicationContext(final ApplicationContext context) {
beanFactory = context.getAutowireCapableBeanFactory();
}
@Override
public Object createJobInstance(final TriggerFiredBundle bundle) throws Exception {
final Object job = super.createJobInstance(bundle);
beanFactory.autowireBean(job);
return job;
}
}
自定义实现了JobFactory,就可以在job中自动注入bean
3.2 AbstractJob
package com.sawyer.job;
import org.quartz.Job;
import org.quartz.JobExecutionContext;
import org.quartz.JobExecutionException;
/**
* 抽象类,job需要继承该类,通过这种方式可以做一些自定义处理
*/
public abstract class AbstractJob implements Job {
@Override
public final void execute(JobExecutionContext jobExecutionContext) throws JobExecutionException {
try {
safeExecute(jobExecutionContext);
} catch (Exception e) {
JobExecutionException e2 = new JobExecutionException(e);
if (!ignoreException()) {
if (isRefireImmediatelyWhenException()) {
//立即重新运行当前job
e2.setRefireImmediately(true);
} else {
//立即停止与当前Job有关的所有触发器,当前job不会再运行
e2.setUnscheduleAllTriggers(true);
}
}
throw e2;
}
}
public abstract void safeExecute(JobExecutionContext context) throws Exception;
/**
* 是否忽略job运行时产生的异常
*/
public abstract boolean ignoreException();
/**
* 发生异常时是否立即重新执行JOB或将JOB挂起.
*
*
* @return {@code true} Job运行产生异常时,立即重新执行JOB.
* {@code false} Job运行产生异常时,挂起JOB等候管理员处理.
*/
public abstract boolean isRefireImmediatelyWhenException();
}
AbstractJob是一个抽象类,凡是job都应该继承该类,通过这种方式可以job做一些自定义处理。如图中代码所示,AbstractJob中有4个方法:
ignoreException:是否忽略异常,返回true时忽略异常,否则必须处理。
isRefireImmediatelyWhenException:当出现异常时,是否立即重新执行。返回true时,立即重新执行,否则将会挂起所有与该job有关的trigger,不再执行。
safeExecute:子类需要实现该方法,在方法中定义具体的实现逻辑。
execute:接口job的实现方法,job最终是在该方法中执行的。只有这个方法才是Job自带的,其他的都是自定义方法。
3.3 TestJob
package com.sawyer.job;
import org.quartz.DisallowConcurrentExecution;
import org.quartz.JobDataMap;
import org.quartz.JobExecutionContext;
import org.quartz.PersistJobDataAfterExecution;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.context.ApplicationContext;
import org.springframework.data.redis.core.RedisTemplate;
import java.util.Date;
@PersistJobDataAfterExecution
@DisallowConcurrentExecution
public class TestJob extends AbstractJob {
private static final String COUNT_KEY = "count";
private static final String APPLICATION_CONTEXT_KEY = "applicationContext";
@Autowired
private RedisTemplate redisTemplate;
@Override
public void safeExecute(JobExecutionContext context) throws Exception {
//统计执行次数
JobDataMap jobDataMap = context.getJobDetail().getJobDataMap();
int count = 0;
if (jobDataMap.containsKey(COUNT_KEY)) {
count = (int) jobDataMap.get(COUNT_KEY);
}
count++;
jobDataMap.put(COUNT_KEY, count);
//打印出redisTemplate,测试是否能自动注入
System.out.println("autowiring spring bean :" + redisTemplate);
//获取spring上下文
ApplicationContext applicationContext = (ApplicationContext) context.getScheduler().getContext().get(APPLICATION_CONTEXT_KEY);
System.out.println("spring context :" + applicationContext);
if (10 == count) {
int m = 1 / 0;
}
System.out.println("********current date :" + new Date() + " and the thread is :" + Thread.currentThread().getName() + " and the count is :" + count + " **********");
}
@Override
public boolean isRefireImmediatelyWhenException() {
return false;
}
@Override
public boolean ignoreException() {
return false;
}
}
Quartz默认是支持并发的,即上一个任务未完成的时候就开始了下一个任务,并且每次都是生成一个新的job实例。图中例子由于需要共享count,统计次数,所在TestJob上面加了两个注解@PersistJobDataExecution和@DisallowConcurrentExecution,这是用于持久化JobData和禁用并发的。加了这两个注解就能够让testJob成为一个有状态的job,并且上一次任务执行完后,才开始下一次任务,这样一来就可以通过jobDataMap共享数据,反之,如果不加注解的话,job就是一个无状态的job,每次运行时都会产生一个新的job实例,数据无法共享。而禁用并发是为了防止并发产生数据紊乱的问题。
TestJob中isRefireImmediatelyWhenException和ignoreException都为false,那么当count等于10时,会抛出异常,系统就会将于当前job有关有的所有trigger挂起,不再执行job。
二、Quartz 存储与持久化
Quartz提供了两种作业存储类型,分别为RAMJobSore和JDBCJobStore。
RASMJobStore:将作业的调度信息存储到内存中,不需要配置外置数据库,配置简单,运行速度快;但是由于调度信息存储在内存中,当应用程序停止时,作业的调度信息将会丢失,此外一旦作业运行期间崩溃,将无法恢复事故现场,比如原定执行30次,执行到第15次是崩溃了,那么系统重启时,将会从0开始。
JDBCJobStore:将作业的调度信息存储到数据库中,该种方式支持集群,调度信息不会丢失,并且可以手动恢复意外停止的job;但是这种方式会较为复杂。
Quartz默认使用的就是RAMJobStore,下面开始介绍持久化配置
1. 数据表初始化
- 从官网中下载quartz包:http://www.quartz-scheduler.org/downloads
- 在quartz-2.2.3/docs/dbTables目录下找到与数据库对应的sql文件,我使用的是mysql数据库,所以这里选择tables.mysql.sql文件,在数据库中执行
2. 引入 quartz.properties ,并根据业务需要进行配置
org.quartz.jobStore.isClustered = true
org.quartz.jobStore.clusterCheckinInterval = 20000
org.quartz.scheduler.instanceId = AUTO
org.quartz.scheduler.skipUpdateCheck = true
# 集群配置
org.quartz.scheduler.instanceName = DefaultQuartzScheduler
org.quartz.scheduler.rmi.export = false
org.quartz.scheduler.rmi.proxy = false
org.quartz.scheduler.wrapJobExecutionInUserTransaction = false
# 线程池的实现类
org.quartz.threadPool.class = org.quartz.simpl.SimpleThreadPool
# 线程数量
org.quartz.threadPool.threadCount = 10
# 线程优先级,最大值为10,最小值为1
org.quartz.threadPool.threadPriority = 5
#
org.quartz.threadPool.threadsInheritContextClassLoaderOfInitializingThread = true
# 持久化配置
#org.quartz.jobStore.class = org.quartz.simpl.RAMJobStore
org.quartz.jobStore.class = org.quartz.impl.jdbcjobstore.JobStoreTX
#org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.HSQLDBDelegate
org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.StdJDBCDelegate
org.quartz.scheduler.classLoadHelper.class=org.quartz.simpl.CascadingClassLoadHelper
#org.quartz.jobStore.useProperties = true
org.quartz.jobStore.tablePrefix = QRTZ_
org.quartz.jobStore.maxMisfiresToHandleAtATime=1
org.quartz.jobStore.selectWithLockSQL=SELECT * FROM {0}LOCKS UPDLOCK WHERE LOCK_NAME = ?
org.quartz.jobStore.misfireThreshold = 60000
#============================================================================
# 配置插件
#============================================================================
#org.quartz.plugin.triggHistory.class=org.quartz.plugins.history.LoggingJobHistoryPlugin
org.quartz.plugin.runningListener.class=com.sawyer.job.RunningListenerPlugin
org.quartz.plugin.runningListener.LogRunningInfo=true
如上图所示,在quartz.properties中,配置了一个自定义的插件:com.sawyer.job.RunningListenerPlugin,可以在这里做一些比较有意义的事情。另外就是,我并没有将数据库信息配置在quart.properties中,而是选择另行配置,目的为了将数据库信息集中到一起,方便操作,实际项目中,也推荐使用这种方式
3. application-datasource.xml
4. application-job.xml
这一步的scheduler配置与上文的基本一致,只是配置了dataSource和引入了quartz.properties文件
5. 自定义插件 RunningListenerPlugin
package com.sawyer.job;
import org.quartz.ListenerManager;
import org.quartz.Scheduler;
import org.quartz.SchedulerException;
import org.quartz.impl.matchers.EverythingMatcher;
import org.quartz.spi.ClassLoadHelper;
import org.quartz.spi.SchedulerPlugin;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.context.ApplicationContext;
public class RunningListenerPlugin implements SchedulerPlugin {
private final Logger log = LoggerFactory.getLogger(getClass());
private Scheduler scheduler;
private boolean isLogRunningInfo;
@Override
public void initialize(String s, Scheduler scheduler, ClassLoadHelper classLoadHelper) throws SchedulerException {
this.scheduler = scheduler;
System.out.println("################## 调度器初始化 ##################");
}
@Override
public void start() {
System.out.println("################## 调度器启动 ###################");
try {
ApplicationContext applicationContext = (ApplicationContext) scheduler.getContext().get("applicationContext");
System.out.println("获取到的 ApplicationContext :" + applicationContext);
ListenerManager listenerManager = scheduler.getListenerManager();
if (isLogRunningInfo()) {
listenerManager.addJobListener(new JobRunningListener(applicationContext), EverythingMatcher.allJobs());
listenerManager.addSchedulerListener(new SchedulerRunningListener(applicationContext));
}
} catch (SchedulerException e) {
throw new RuntimeException(e);
}
}
@Override
public void shutdown() {
System.out.println("############# 调度器关闭 ##################");
}
public boolean isLogRunningInfo() {
return isLogRunningInfo;
}
public void setLogRunningInfo(boolean logRunningInfo) {
isLogRunningInfo = logRunningInfo;
}
}
可以添加多个监听器,实现各种各样的业务需求,比如用于任务结束后发邮件,记录更为详细的运行信息等。只需要实现响应的接口即可,这里我选择了添加JobRunningListener和SchedulerRunningListener
6. JobRunningListener
package com.sawyer.job;
import org.quartz.*;
import org.springframework.context.ApplicationContext;
public class JobRunningListener implements JobListener {
private final static String LISTENER_NAME = "JobRunningListener";
private ApplicationContext applicationContext;
public JobRunningListener(ApplicationContext applicationContext) {
this.applicationContext = applicationContext;
}
@Override
public String getName() {
return LISTENER_NAME;
}
@Override
public void jobToBeExecuted(JobExecutionContext jobExecutionContext) {
System.out.println("######### job 准备开始执行 ##########");
}
@Override
public void jobExecutionVetoed(JobExecutionContext jobExecutionContext) {
System.out.println("######### job 被否决了 ##########");
}
@Override
public void jobWasExecuted(JobExecutionContext jobExecutionContext, JobExecutionException e) {
System.out.println("######### job 执行完毕 ##########");
}
}
7. SchdulerRunningListener
package com.sawyer.job;
import org.quartz.*;
import org.springframework.context.ApplicationContext;
public class SchedulerRunningListener implements SchedulerListener {
private final ApplicationContext applicationContext;
public SchedulerRunningListener(ApplicationContext applicationContext) {
this.applicationContext = applicationContext;
}
@Override
public void jobScheduled(Trigger trigger) {
}
@Override
public void jobUnscheduled(TriggerKey triggerKey) {
}
@Override
public void triggerFinalized(Trigger trigger) {
}
@Override
public void triggerPaused(TriggerKey triggerKey) {
}
@Override
public void triggersPaused(String s) {
}
@Override
public void triggerResumed(TriggerKey triggerKey) {
}
@Override
public void triggersResumed(String s) {
}
@Override
public void jobAdded(JobDetail jobDetail) {
}
@Override
public void jobDeleted(JobKey jobKey) {
System.out.println("############ " + jobKey.getGroup() + " 下的 " + jobKey.getName() + "已被删除 ###################");
}
@Override
public void jobPaused(JobKey jobKey) {
}
@Override
public void jobsPaused(String s) {
}
@Override
public void jobResumed(JobKey jobKey) {
}
@Override
public void jobsResumed(String s) {
}
@Override
public void schedulerError(String s, SchedulerException e) {
}
@Override
public void schedulerInStandbyMode() {
}
@Override
public void schedulerStarted() {
}
@Override
public void schedulerStarting() {
}
@Override
public void schedulerShutdown() {
}
@Override
public void schedulerShuttingdown() {
}
@Override
public void schedulingDataCleared() {
}
}
8. 执行job
package com.sawyer;
import com.sawyer.job.AbstractJob;
import org.quartz.*;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.context.ApplicationContext;
import org.springframework.context.support.ClassPathXmlApplicationContext;
public class Main {
private static final Logger logger = LoggerFactory.getLogger(Main.class);
public static void main(String[] args) {
String configLocation = "classpath:spring/applicationContext*.xml";
ClassPathXmlApplicationContext context = new ClassPathXmlApplicationContext(configLocation);
try {
createJob(context);
} catch (SchedulerException e) {
e.printStackTrace();
}
logger.info("Container has been startup.");
}
public static void createJob(ApplicationContext context) throws SchedulerException {
final String jobClassName = "com.sawyer.job.TestJob";
final String jobName = "test_job";
final String jobGroup = "test_group";
final String jobDescription = "这是TestJob的描述信息";
final String triggerName = "test_trigger";
final String triggerGroup = "test_trigger_group";
final String cron = "0/5 * * * * ?";
// 加载job类,并判断 job类的父类是否为AbstractJob
boolean assignableFrom = false;
Class forName = null;
try {
forName = Class.forName(jobClassName);
assignableFrom = AbstractJob.class.isAssignableFrom(forName);
} catch (ClassNotFoundException e) {
if (logger.isErrorEnabled()) {
logger.error(e.getMessage(), e);
}
}
JobBuilder jb = JobBuilder.newJob(forName).withIdentity(jobName, jobGroup)
.withDescription(jobDescription);
//放入数据
JobDataMap data = new JobDataMap();
data.put("count", 5);
jb = jb.usingJobData(data);
JobDetail jobDetail = jb.build();
TriggerBuilder triggerBuilder = TriggerBuilder.newTrigger()
.withIdentity(triggerName, triggerGroup).forJob(jobDetail);
ScheduleBuilder sche = CronScheduleBuilder.cronSchedule(cron);
Trigger trigger = triggerBuilder.withSchedule(sche).build();
Scheduler quartzScheduler = (Scheduler) context.getBean("quartzScheduler");
quartzScheduler.scheduleJob(jobDetail, trigger);
}
}
这里使用的job类仍然是上文中的TestJob,只是改用程序生成调用而已,另外这里的count我手动赋值成了5
到此,quartz持久化配置就完成了。
三、web管理
quartz是通过jobName和jobGroup来区分job的,所以job的name和group一定要填写,然后通过这个就可以对job进行添加、暂停、恢复和删除了。
添加:quartzScheduler.scheduleJob(jobDetail,trigger)
暂停:quartzScheduler.pauseJob(JobKey.jobKey(jobName,jobGroup))
恢复:quartzSchduler.resumeJob(JobKey.jobKey(jobName,jobGroup))
删除:quartScheduler.deleteJob(JobKey.jobKey(jobName,jobGroup))
可以根据需要定制相关的UI界面,通过界面去触发job操作