先弄个死锁的代码例子：

public class DeadLock implements Runnable {

	private int flagCurrent = -1;

	private static int flagThread1 = 0;

	private static int flagThread2 = 1;

	private static Object object1 = new Integer(1);

	private static Object object2 = new Double(2);

	@Override
	public void run() {

		System.out.println("CurrentThread" + flagCurrent);

		if (flagCurrent == flagThread1) {// 线程1进
			synchronized (object1) {// 线程1先锁定object1（期间要等500ms去锁定object2）
				try {
					Thread.sleep(500);
				} catch (final Exception e) {
					e.printStackTrace();
				}
				// 此时线程2已经锁定object2，需要【等待】线程2释放object2，然后线程1锁定object2释放object2，再释放object1
				// 死锁发生：线程1等待线程2释放object2，线程1才释放object1；线程2等待线程1释放object1，线程2才释放object2；
				synchronized (object2) {
					System.out.println("1");
				}
			}
		}
		if (flagCurrent == flagThread2) {// 线程2进
			synchronized (object2) {// 线程2先锁定object2（期间要等500ms去锁定object1）
				try {
					Thread.sleep(500);
				} catch (final Exception e) {
					e.printStackTrace();
				}
				// 此时线程1已经锁定object1，需要【等待】线程1释放object1，然后线程2锁定object1释放object1，再释放object2
				// 死锁发生：线程1等待线程2释放object2，线程1才释放object1；线程2等待线程1释放object1，线程2才释放object2；
				synchronized (object1) {
					System.out.println("0");
				}
			}
		}
	}

	public static void main(final String[] args) {
		final DeadLock runnableThread1 = new DeadLock();
		final DeadLock runnableThread2 = new DeadLock();

		runnableThread1.flagCurrent = flagThread1;
		runnableThread2.flagCurrent = flagThread2;

		new Thread(runnableThread1, "runnableThread1").start();
		new Thread(runnableThread2, "runnableThread2").start();

	}
}

查找死锁的2个方法：

1.用Jconsole

切换到线程那个选项卡，左下角有个死锁检查：

2.用Jstack

先jps 找出对应的代号，再Jstack -l  代号，就可看见了