Linux笔记

vmstat  可以查看上下文切换的次数 http://www.cnblogs.com/ggjucheng/archive/2012/01/05/2312625.html

root@ubuntu:~# vmstat 2 1
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
 1  0      0 3498472 315836 3819540    0    0     0     1    2    0  0  0 100  0

jstack PID > /tmp/dump17  查看PID中线程的状态

2017-11-10 17:39:05
Full thread dump Java HotSpot(TM) 64-Bit Server VM (25.45-b02 mixed mode):

"Attach Listener" #3381 daemon prio=9 os_prio=0 tid=0x00007f3c08075800 nid=0x1e58 waiting on condition [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"Worker-408" #3380 prio=5 os_prio=0 tid=0x0000000001bd7000 nid=0x8ea in Object.wait() [0x00007f3bc43f7000]
   java.lang.Thread.State: TIMED_WAITING (on object monitor)
        at java.lang.Object.wait(Native Method)
        at org.eclipse.core.internal.jobs.WorkerPool.sleep(WorkerPool.java:188)
        - locked <0x00000000c149edb0> (a org.eclipse.core.internal.jobs.WorkerPool)
        at org.eclipse.core.internal.jobs.WorkerPool.startJob(WorkerPool.java:220)
        at org.eclipse.core.internal.jobs.Worker.run(Worker.java:52)

 grep java.lang.Thread.State /tmp/dump17 | awk '{print $2$3$4$5}' | sort | uniq -c

     11 RUNNABLE
     12 TIMED_WAITING(onobjectmonitor)
      1 TIMED_WAITING(parking)
      2 TIMED_WAITING(sleeping)
     29 WAITING(onobjectmonitor)
     20 WAITING(parking)

sort 按照ASCII码排序， sort -n 按照数字排序， sort -r 倒序排序

uniq 去除重复行， uniq -d 只显示重复行， uniq -u只显示出现一次的行 uniq -c prefix lines by the number of occurrences

执行 uniq --help可查看使用情况

这里记录一下线程的几种状态

死锁，Deadlock（重点关注） 
执行中，Runnable   
等待资源，Waiting on condition（重点关注） 
等待获取监视器，Waiting on monitor entry（重点关注）(entry set)
暂停，Suspended
对象等待中，Object.wait() 或 TIMED_WAITING （wait set）
阻塞，Blocked（重点关注）  
停止，Parked

参考文章 http://www.cnblogs.com/zhengyun_ustc/archive/2013/01/06/dumpanalysis.html

实例分析一段代码，找到死锁的根源

public class DeadLock {
	
	private static String A = "A";
	private static String B = "B";

	public static void main(String[] args) {
		new DeadLock().deadLock();
	}

	private void deadLock() {
		Thread t1 = new Thread(new Runnable() {

			@Override
			public void run() {
				synchronized (A) {
					System.out.println("[Thread1] Got A");
					synchronized (B) { //blocked
						System.out.println("[Thread1] Got B");
						try {
							Thread.sleep(1000);
						} catch (InterruptedException e) {
							System.out.println("The thread is interruped " + e.getMessage());
						}
					}
				}

			}

		});
		
		Thread t2 = new Thread() {
			public void run() {
				synchronized(B){
					System.out.println("[Thread2] Got B");
					synchronized(A){ // blocked
						System.out.println("[Thread2] Got A");
						try{
							Thread.sleep(1000);
							}
						catch(InterruptedException e) {
							System.out.println("The thread is interruped " + e.getMessage());
						}
					}
				}
			}
		};
		
		t1.start();
		t2.start();
	}

}

我们执行下列命令

ps -aux | grep DeadLock  可以获得进程PID
jstack 31165 > /tmp/dump31165 
vi /tmp/dump3116

"Thread-1" #9 prio=5 os_prio=0 tid=0x00007f093c0d2800 nid=0x79d2 waiting for monitor entry [0x00007f0925783000]
   java.lang.Thread.State: BLOCKED (on object monitor)
        at JavaConcurrency.Test.DeadLock$2.run(DeadLock.java:37)
        - waiting to lock <0x0000000782b5aba8> (a java.lang.String)
        - locked <0x0000000782b5abd8> (a java.lang.String)


"Thread-0" #8 prio=5 os_prio=0 tid=0x00007f093c0d1000 nid=0x79d1 waiting for monitor entry [0x00007f0925884000]
   java.lang.Thread.State: BLOCKED (on object monitor)
        at JavaConcurrency.Test.DeadLock$1.run(DeadLock.java:19)
        - waiting to lock <0x0000000782b5abd8> (a java.lang.String)
        - locked <0x0000000782b5aba8> (a java.lang.String)
        at java.lang.Thread.run(Thread.java:745)

可以看到这两个线程互相等待对方的锁， 可见  jstack 是根据线程状态来分析死锁是很有用的，虽然上面的代码非常简单，但是在复杂的业务场景中，这是个浓缩版的死锁

避免死锁的方式：

避免一个线程同时获取多个锁

避免一个线程在锁内同时占用多个资源，尽量保证每个锁只占用一个资源

尝试使用定时锁，使用lock.tryLock(timeout)来替代使用内部锁机制（在以后的文章中需要详细了解下定时锁）

对于数据库锁，加锁和解锁必须在同一个数据库连接里，否则会出现解锁失败的情况 （后续需要看一下数据库的内容）