JAVA线程执行中断方式

Java中只能通过协作的方式取消

• 第一种是通过标志位实现，假设有个计算所有素数的任务，每次计算前检查下是否取消的标志位，如果为true则退出计算。调用方想要取消任务的话，则将标志位设为true。但这种方法无法再计算的过程中取消任务，像是一些阻塞调用无法被取消

• 第二种是中断，用于通过协作机制停止线程继续执行任务，原理是向进程发送中断请求将标记线程为Interrupted，线程会在下一个合适的时刻停止运行，阻塞的库方法例如Thread.sleep Object.wait 都会响应中断，抛出InterruptedException意味着阻塞操作因为中断结束，但不能保证响应速度。

  通常任务不会在自己拥有的线程执行，而是在某个服务的线程池中执行，因此任务应该保存阻塞状态，使得线程能处理中断。这就是库函数抛出InterruptedException的原因。同时任务也不能对线程处理中断的策略做任何假设。

  有两种处理中断的方法，千万不要catch掉异常什么也不做，除非你拥有该线程

  • 向外抛出InterruptedException
  • 调用 Thread.currentThread().interrupt();恢复中断状态

  例如在向线程池提交runnable执行时，runnable调用了阻塞库函数则需要写处理中断的逻辑，逻辑中要使用上面处理中断的两种方法之一，因为线程池会处理线程遇到阻塞，将线程从可用线程列表删除，统计审计信息等操作

  final void runWorker(Worker w) {Thread wt = Thread.currentThread();Runnable task = w.firstTask;w.firstTask = null;w.unlock(); // allow interruptsboolean completedAbruptly = true;try {while (task != null || (task = getTask()) != null) {w.lock();try {beforeExecute(wt, task);Throwable thrown = null;try {task.run(); // 执行runnable} catch (RuntimeException x) {thrown = x; throw x;} catch (Error x) {thrown = x; throw x;} catch (Throwable x) {thrown = x; throw new Error(x);} finally {afterExecute(task, thrown);}} finally {task = null;w.completedTasks++;w.unlock();}}completedAbruptly = false;} finally {processWorkerExit(w, completedAbruptly); // 处理InteruptException}}
  private void processWorkerExit(Worker w, boolean completedAbruptly) {if (completedAbruptly) // If abrupt, then workerCount wasn't adjusteddecrementWorkerCount();final ReentrantLock mainLock = this.mainLock;mainLock.lock();try {completedTaskCount += w.completedTasks;workers.remove(w);} finally {mainLock.unlock();}tryTerminate();int c = ctl.get();if (runStateLessThan(c, STOP)) {if (!completedAbruptly) {int min = allowCoreThreadTimeOut ? 0 : corePoolSize;if (min == 0 && ! workQueue.isEmpty())min = 1;if (workerCountOf(c) >= min)return; // replacement not needed}addWorker(null, false);}}
  

还有一种方法是使用future来中断任务执行，使用submit.cancel(true)，内部也是利用了Thread.interrupt

ExecutorService executor = Executors.newFixedThreadPool(1);final Future<?> submit = executor.submit(() -> {});try {final Object o = submit.get(10, TimeUnit.MILLISECONDS);} catch (TimeoutException e) {}finally {submit.cancel(true); // 正常得到结果后再cancel也无影响}

未捕获异常的处理方式

线程因为异常或者error退出后，会向System.error输出堆栈，但线上程序一般通过日志来排查问题而不是控制台输出。Java支持配置Thread.UncaughtExceptionHandler处理这种未被捕获的Throwable，程序最好自己实现该接口，至少将异常堆栈输出到日志中，下面是ES实现的Thread.UncaughtExceptionHandler(去掉了一些细节)，ES将异常信息输出到了日志文件。

class ElasticsearchUncaughtExceptionHandler implements Thread.UncaughtExceptionHandler {private static final Logger logger = LogManager.getLogger(ElasticsearchUncaughtExceptionHandler.class);@Overridepublic void uncaughtException(Thread thread, Throwable t) {if (isFatalUncaught(t)) {onFatalUncaught(thread.getName(), t);} else {onNonFatalUncaught(thread.getName(), t);}}static boolean isFatalUncaught(Throwable e) {return e instanceof Error;}void onFatalUncaught(final String threadName, final Throwable t) {final String message = "fatal error in thread [" + threadName + "], exiting";logger.error(message, t);Terminal.DEFAULT.errorPrintln(message);t.printStackTrace(Terminal.DEFAULT.getErrorWriter());// Without a final flush, the stacktrace may not be shown before ES exitsTerminal.DEFAULT.flush();}void onNonFatalUncaught(final String threadName, final Throwable t) {final String message = "uncaught exception in thread [" + threadName + "]";logger.error(message, t);Terminal.DEFAULT.errorPrintln(message);t.printStackTrace(Terminal.DEFAULT.getErrorWriter());// Without a final flush, the stacktrace may not be shown if ES goes on to exitTerminal.DEFAULT.flush();}}