解决线程池中ThreadGroup的坑
线程池中ThreadGroup的坑
在Java中每一个线程都归属于某个线程组管理的一员,例如在主函数main()主工作流程中产生一个线程,则产生的线程属于main这个线程组管理的一员。简单地说,线程组(ThreadGroup)就是由线程组成的管理线程的类,这个类是java.lang.ThreadGroup类。
定义一个线程组,通过以下代码可以实现。
ThreadGroup group=new ThreadGroup(“groupName”); Thread thread=new Thread(group,”the first thread of group”);
ThreadGroup类中的某些方法,可以对线程组中的线程产生作用。例如,setMaxPriority()方法可以设定线程组中的所有线程拥有最大的优先权。
所有线程都隶属于一个线程组。那可以是一个默认线程组(不指定group),亦可是一个创建线程时明确指定的组。在创建之初,线程被限制到一个组里,而且不能改变到一个不同的组。每个应用都至少有一个线程从属于系统线程组。若创建多个线程而不指定一个组,它们就会自动归属于系统线程组。
线程组也必须从属于其他线程组。必须在构建器里指定新线程组从属于哪个线程组。若在创建一个线程组的时候没有指定它的归属,则同样会自动成为系统线程组的一名属下。因此,一个应用程序中的所有线程组最终都会将系统线程组作为自己的“父”。
那么假如我们需要在线程池中实现一个带自定义ThreadGroup的线程分组,该怎么实现呢?
我们在给线程池(ThreadPoolExecutor)提交任务的时候可以通过execute(Runnable command)来将一个线程任务加入到该线程池,那么我们是否可以通过new一个指定了ThreadGroup的Thread实例来加入线程池来达到前面说到的目的呢?
ThreadGroup是否可行
通过new Thread(threadGroup,runnable)实现线程池中任务分组
public static void main(String[] args) { ThreadPoolExecutor pool = (ThreadPoolExecutor) Executors.newCachedThreadPool(); final ThreadGroup group = new ThreadGroup("Main_Test_Group"); for (int i = 0; i < 5; i++) { Thread thread = new Thread(group, new Runnable() { @Override public void run() { int sleep = (int)(Math.random() * 10); try { Thread.sleep(1000 * 3); System.out.println(Thread.currentThread().getName()+"执行完毕"); System.out.println("当前线程组中的运行线程数"+group.activeCount()); } catch (InterruptedException e) { e.printStackTrace(); } } }, group.getName()+" #"+i+""); pool.execute(thread); } }
运行结果
pool-1-thread-3执行完毕
pool-1-thread-1执行完毕
当前线程组中的运行线程数0
pool-1-thread-2执行完毕
当前线程组中的运行线程数0
当前线程组中的运行线程数0
pool-1-thread-4执行完毕
pool-1-thread-5执行完毕
当前线程组中的运行线程数0
当前线程组中的运行线程数0
运行结果中可以看到group中的线程并没有因为线程池启动了这个线程任务而运行起来.因此通过线程组来对线程池中的线层任务分组不可行.
从java.util.concurrent.ThreadPoolExecutor源码中可以看到如下构造函数:
public ThreadPoolExecutor(int corePoolSize, int maximumPoolSize, long keepAliveTime, TimeUnit unit, BlockingQueue<Runnable> workQueue) { this(corePoolSize, maximumPoolSize, keepAliveTime, unit, workQueue, Executors.defaultThreadFactory(), defaultHandler); }
如果我们在实例化ThreadPoolExecutor时不指定ThreadFactory,那么将以默认的ThreadFactory来创建Thread.
Executors内部类DefaultThreadFactory
下面的源码即是默认的Thread工厂
static class DefaultThreadFactory implements ThreadFactory { private static final AtomicInteger poolNumber = new AtomicInteger(1); private final ThreadGroup group; private final AtomicInteger threadNumber = new AtomicInteger(1); private final String namePrefix; DefaultThreadFactory() { SecurityManager s = System.getSecurityManager(); group = (s != null) ? s.getThreadGroup() : Thread.currentThread().getThreadGroup(); namePrefix = "pool-" + poolNumber.getAndIncrement() + "-thread-"; } public Thread newThread(Runnable r) { Thread t = new Thread(group, r, namePrefix + threadNumber.getAndIncrement(), 0); if (t.isDaemon()) t.setDaemon(false); if (t.getPriority() != Thread.NORM_PRIORITY) t.setPriority(Thread.NORM_PRIORITY); return t; } }
从唯一的构造函数可以看到DefaultThreadFactory以SecurityManager 实例中的ThreadGroup来指定线程的group,如果SecurityManager 获取到的ThreadGroup为null才默认以当前线程的group来指定.public Thread newThread(Runnable r) 则以group来new 一个Thead.这样我们可以在实例化ThreadPoolExecutor对象的时候在其构造函数内传入自定义的ThreadFactory实例即可达到目的.
public class MyTheadFactory implements ThreadFactory { private static final AtomicInteger poolNumber = new AtomicInteger(1); private final AtomicInteger threadNumber = new AtomicInteger(1); private final String namePrefix; private ThreadGroup defaultGroup; public MyTheadFactory() { SecurityManager s = System.getSecurityManager(); defaultGroup = (s != null) ? s.getThreadGroup() : Thread.currentThread().getThreadGroup(); namePrefix = "pool-" + poolNumber.getAndIncrement() + "-thread-"; } public MyTheadFactory(ThreadGroup group) { this.defaultGroup = group; namePrefix = "pool-" + poolNumber.getAndIncrement() + "-thread-"; } public Thread newThread(Runnable r) { Thread t = new Thread(defaultGroup, null, namePrefix + threadNumber.getAndIncrement(), 0); if (t.isDaemon()) t.setDaemon(false); if (t.getPriority() != Thread.NORM_PRIORITY) t.setPriority(Thread.NORM_PRIORITY); return t; } }
ThreadGroup的使用及手写线程池
监听线程异常关闭
以下代码在window下不方便测试,需在linux 上 测试
// 以下线程如果强制关闭的话,是无法打印`线程被杀掉了` // 模拟关闭 kill PID public static void main(String[] args) { Runtime.getRuntime().addShutdownHook(new Thread( () -> { System.out.println("线程被杀掉了"); })); while(true){ System.out.println("i am working ..."); try { Thread.sleep(1000); } catch (InterruptedException e) { e.printStackTrace(); } } }
如何拿到Thread线程中异常
public static void main(String[] args) { Thread thread = new Thread(() -> { try { Thread.sleep(1000); int i = 10/0; } catch (InterruptedException e) { e.printStackTrace(); } }); thread.setUncaughtExceptionHandler((t,e)->{ System.out.println("线程的名字"+ t.getName()); System.out.println(e); }); // 通过注入接口的方式 thread.start(); }
ThreadGroup
注意: threadGroup 设置为isDaemon 后,会随最后一个线程结束而销毁,如果没有设置isDaemon ,则需要手动调用 destory()
线程池使用
自己搭建的简单线程池实现
其中ThreadGroup 的应用没有写,但是我们可以观察线程关闭后,检查ThreadGroup 中是否还有活跃的线程等,具体参考ThreadGroup API
import java.util.ArrayList; import java.util.Iterator; import java.util.LinkedList; import java.util.List; import java.util.stream.IntStream; /** * @Author: shengjm * @Date: 2020/2/10 9:52 * @Description: */ public class SimpleThreadPool extends Thread{ /** * 线程数量 */ private int size; private final int queueSize; /** * 默认线程队列数量 */ private final static int DEFAULR_TASK_QUEUE_SIZE = 2000; private static volatile int seq = 0; private final static String THREAD_PREFIX = "SIMPLE_THREAD_POLL_"; private final static ThreadGroup GROUP = new ThreadGroup("Pool_Group"); private final static LinkedList<Runnable> TASK_QUEUE = new LinkedList<>(); private final static List<WorkerTask> THREAD_QUEUE = new ArrayList<>(); private final DiscardPolicy discardPolicy; private volatile boolean destory = false; private int min; private int max; private int active; /** * 定义异常策略的实现 */ private final static DiscardPolicy DEFAULT_DISCARD_POLICY = () -> { throw new DiscardException("线程池已经被撑爆了,后继多余的人将丢失"); }; /** * */ public SimpleThreadPool(){ this(4,8,12,DEFAULR_TASK_QUEUE_SIZE,DEFAULT_DISCARD_POLICY); } /** * */ public SimpleThreadPool(int min , int active , int max , int queueSize,DiscardPolicy discardPolicy) { this.min = min; this.active = active; this.max = max; this.queueSize = queueSize; this.discardPolicy = discardPolicy; init(); } /** * 初始化 */ private void init() { for(int i = 0; i < min; i++){ createWorkTask(); } this.size = min; this.start(); } private void createWorkTask(){ WorkerTask task = new WorkerTask(GROUP,THREAD_PREFIX+(seq++)); task.start(); THREAD_QUEUE.add(task); } /** * 线程池自动扩充 */ @Override public void run() { while(!destory){ System.out.println(this.min +" --- "+this.active+" --- "+this.max + " --- "+ this.size + " --- "+ TASK_QUEUE.size()); try { Thread.sleep(1000); if(TASK_QUEUE.size() > active && size < active){ for (int i = size; i < active;i++){ createWorkTask(); } size = active; }else if(TASK_QUEUE.size() > max && size < max){ for (int i = size; i < max;i++){ createWorkTask(); } size = max; } synchronized (THREAD_QUEUE){ if(TASK_QUEUE.isEmpty() && size > active){ int release = size - active; for (Iterator<WorkerTask> it = THREAD_QUEUE.iterator();it.hasNext();){ if(release <=0){ break; } WorkerTask task = it.next(); task.close(); task.interrupt(); it.remove(); release--; } size = active; } } } catch (InterruptedException e) { break; } } } public void submit(Runnable runnable){ synchronized (TASK_QUEUE){ if(destory){ throw new DiscardException("线程池已经被摧毁了..."); } if(TASK_QUEUE.size() > queueSize){ discardPolicy.discard(); } TASK_QUEUE.addLast(runnable); TASK_QUEUE.notifyAll(); } } /** * 关闭 */ public void shutdown(){ while(!TASK_QUEUE.isEmpty()){ try { Thread.sleep(10); } catch (InterruptedException e) { e.printStackTrace(); } } synchronized (THREAD_QUEUE) { int initVal = THREAD_QUEUE.size(); while (initVal > 0) { for (WorkerTask workerTask : THREAD_QUEUE) { if (workerTask.getTaskState() == TaskState.BLOCKED) { workerTask.interrupt(); workerTask.close(); initVal--; } else { try { Thread.sleep(10); } catch (InterruptedException e) { e.printStackTrace(); } } } } this.destory = true; } } public int getSize() { return size; } public int getMin() { return min; } public int getMax() { return max; } public int getActive() { return active; } /** * 线程状态 */ private enum TaskState{ FREE , RUNNING , BLOCKED , DEAD } /** * 自定义异常类 */ public static class DiscardException extends RuntimeException{ public DiscardException(String message){ super(message); } } /** * 定义异常策略 */ @FunctionalInterface public interface DiscardPolicy{ void discard() throws DiscardException; } private static class WorkerTask extends Thread{ private volatile TaskState taskState = TaskState.FREE; public TaskState getTaskState(){ return this.taskState; } public WorkerTask(ThreadGroup group , String name){ super(group , name); } @Override public void run(){ OUTER: while(this.taskState != TaskState.DEAD){ Runnable runnable; synchronized (TASK_QUEUE){ while(TASK_QUEUE.isEmpty()){ try { taskState = TaskState.BLOCKED; TASK_QUEUE.wait(); } catch (InterruptedException e) { break OUTER; } } runnable = TASK_QUEUE.removeFirst(); } if(runnable != null){ taskState = TaskState.RUNNING; runnable.run(); taskState = TaskState.FREE; } } } public void close(){ this.taskState = TaskState.DEAD; } } /** * 测试 * @param args */ public static void main(String[] args) { SimpleThreadPool simpleThreadPool = new SimpleThreadPool(); // SimpleThreadPool simpleThreadPool = new SimpleThreadPool(6,15,SimpleThreadPool.DEFAULT_DISCARD_POLICY); IntStream.rangeClosed(0,40).forEach(i -> { simpleThreadPool.submit(() -> { try { Thread.sleep(1000); } catch (InterruptedException e) { e.printStackTrace(); } System.out.println("the runnable " + i + "be servered by " + Thread.currentThread()); }); }); // try { // Thread.sleep(15000); // } catch (InterruptedException e) { // e.printStackTrace(); // } simpleThreadPool.shutdown(); } }
以上为个人经验,希望能给大家一个参考,也希望大家多多支持编程宝库。
由于maven 使用上手很容易所以很多时候可以囫囵吞枣能够使用就可以了,由于作者最近在做的持续集成的代码扫描的时候,发现私有云里面大型工程maven依赖,如果没有弄清楚里面的配置复杂的项目很难在私有环境里面 ...