前言
前序章节:springboot基础(82):分布式定时任务解决方案shedlock
如果你不清楚shedlock,建议先阅读前序章节,再来查看本文。
如果我们不在spring环境下,如何使用shedlock实现分布式互斥执行?
我们可以使用shedlock为我们提供的DefaultLockingTaskExecutor来实现手动调用。
示例
void executeWithLock(@NonNull Runnable var1, @NonNull LockConfiguration var2)
java
@GetMapping("/testRunnable")
public R testRunnable(HttpServletRequest request) {
log.info("进入方法");
String name = request.getParameter("name");
LockingTaskExecutor executor = new DefaultLockingTaskExecutor(lockProvider);
Instant now = Instant.now();
try {
executor.executeWithLock(new Runnable() {
@Override
public void run() {
log.info("执行");
helloService.helloCn(name);
}
}, new LockConfiguration(now, "testRunnable", Duration.ofSeconds(30), Duration.ofSeconds(5)));
log.info("end");
return R.ok("success", null);
} catch (Throwable throwable) {
throwable.printStackTrace();
}
return R.fail("fail");
}
LockingTaskExecutor.TaskResult executeWithLock(@NonNull LockingTaskExecutor.TaskWithResult task, @NonNull LockConfiguration lockConfig)
利用此API,我们可以让一个方法不能在一个时间只能有一次实例在执行,排斥调用者,且其它调用者的调用失败,这是与分布式锁不一样的地方。
java
@GetMapping("/testTaskWithResult")
public R testTaskWithResult(HttpServletRequest request) {
log.info("进入方法");
String name = request.getParameter("name");
LockingTaskExecutor executor = new DefaultLockingTaskExecutor(lockProvider);
Instant now = Instant.now();
try {
LockingTaskExecutor.TaskResult taskResult = executor.executeWithLock(new LockingTaskExecutor.TaskWithResult() {
@Override
public Object call() throws Throwable {
log.info("执行");
return helloService.helloCn(name);
}
}, new LockConfiguration(now, "testTaskWithResult", Duration.ofSeconds(30), Duration.ofSeconds(5)));
boolean flag = taskResult.wasExecuted();
log.info("end");
return R.ok("任务是否被执行:" + flag, taskResult.getResult());
} catch (Throwable throwable) {
throwable.printStackTrace();
}
return R.fail("fail");
}