jstack應用-查找CPU飈高的緣由 jstack應用-查找CPU飈高的緣由

時間 2020-08-21

原文原文鏈接

jstack應用-查找CPU飈高的緣由

場景
java

在系統上線後，常常會遇到運維的同窗跑過來講：「此次發版後，cpu線程使用率到一場，到100%了」。這時候不要慌，能夠使用堆轉儲來分析究竟是哪一個線程引發的。web

查找元兇

[root@java_mofei_01 test]# topMem: 16333644k total, 9472968k used, 6860676k free, 165616k buffers
Swap: 0k total, 0k used, 0k free, 6665292k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
17850 root      20   0 7588m 112m  11m S 100.7  0.7  47:53.80 java
1552 root      20   0  121m  13m 8524 S  0.7  0.1  14:37.75 AliYunDun
3581 root      20   0 9750m 2.0g  13m S  0.7 12.9 298:30.20 java
    1 root      20   0 19360 1612 1308 S  0.0  0.0   0:00.81 init
    2 root      20   0     0    0    0 S  0.0  0.0   0:00.00 kthreadd
    3 root      RT   0     0    0    0 S  0.0  0.0   0:00.14 migration/0

發現pid=17850的進程線程使用率100%，查詢進程中哪一個線程致使的問題app

[root@java_mofei_01 test]# top -H -p 17850top - 17:43:15 up 5 days, 7:31, 1 user, load average: 0.99, 0.97, 0.91Tasks: 32 total, 1 running, 31 sleeping, 0 stopped, 0 zombieCpu(s): 3.7%us, 8.9%sy, 0.0%ni, 87.4%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%stMem: 16333644k total, 9592504k used, 6741140k free, 165700k buffersSwap: 0k total, 0k used, 0k free, 6781620k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 17880 root 20 0 7588m 112m 11m R 99.9 0.7 50:47.43 java 17856 root 20 0 7588m 112m 11m S 0.3 0.7 0:02.08 java 17850 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17851 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.23 java 17852 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.09 java 17853 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.12 java 17854 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.07 java 17855 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.06 java 17857 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.07 java 17858 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.08 java 17859 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.04 java 17860 root 20 0 7588m 112m 11m S 0.0 0.7 0:02.05 java 17861 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17862 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17863 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17864 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.10 java 17865 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.12 java 17866 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.09 java 17867 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.12 java 17868 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17869 root 20 0 7588m 112m 11m S 0.0 0.7 0:01.04 java 17870 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17871 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17872 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17873 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17874 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17875 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17876 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17877 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17878 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17879 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java 17946 root 20 0 7588m 112m 11m S 0.0 0.7 0:00.00 java

發現17880的線程致使cpu飈高。運維

經過jstack查看線程信息

[root@java_mofei_01 test]# printf "%x\n" 17880          45d8
[root@java_mofei_01 test]# jstack 17850|grep 45d8 -A 30"pool-1-thread-11" #20 prio=5 os_prio=0 tid=0x00007fc860352800 nid=0x45d8 runnable [0x00007fc8417d2000]
   java.lang.Thread.State: RUNNABLE
        at java.io.FileOutputStream.writeBytes(Native Method)
        at java.io.FileOutputStream.write(FileOutputStream.java:326)
        at java.io.BufferedOutputStream.flushBuffer(BufferedOutputStream.java:82)
        at java.io.BufferedOutputStream.flush(BufferedOutputStream.java:140)
        - locked <0x00000006c6c2e708> (a java.io.BufferedOutputStream)
        at java.io.PrintStream.write(PrintStream.java:482)
        - locked <0x00000006c6c10178> (a java.io.PrintStream)
        at sun.nio.cs.StreamEncoder.writeBytes(StreamEncoder.java:221)
        at sun.nio.cs.StreamEncoder.implFlushBuffer(StreamEncoder.java:291)
        at sun.nio.cs.StreamEncoder.flushBuffer(StreamEncoder.java:104)
        - locked <0x00000006c6c26620> (a java.io.OutputStreamWriter)
        at java.io.OutputStreamWriter.flushBuffer(OutputStreamWriter.java:185)
        at java.io.PrintStream.write(PrintStream.java:527)
        - eliminated <0x00000006c6c10178> (a java.io.PrintStream)
        at java.io.PrintStream.print(PrintStream.java:597)
        at java.io.PrintStream.println(PrintStream.java:736)
        - locked <0x00000006c6c10178> (a java.io.PrintStream)
        at com.demo.guava.HardTask.call(HardTask.java:18)
        at com.demo.guava.HardTask.call(HardTask.java:9)
        at java.util.concurrent.FutureTask.run(FutureTask.java:266)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
        at java.lang.Thread.run(Thread.java:745)"pool-1-thread-10" #19 prio=5 os_prio=0 tid=0x00007fc860345000 nid=0x45d7 waiting on condition [0x00007fc8418d3000]
   java.lang.Thread.State: WAITING (parking)
        at sun.misc.Unsafe.park(Native Method)
        - parking to wait for  <0x00000006c6c14178> (a java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
        at java.util.concurrent.locks.LockSupport.park(LockSupport.java:175)

首先把pid 17880轉成16進制ide
查詢堆信息spa
發現HardTask（源碼在末尾）第18行可能有問題線程
源碼

public class HardTask implements Callable<String> {    @Override
    public String call() throws Exception {        while(true){            int a=100;            int b=100;            int rs=a+b/100-b;
            System.out.println(rs);
        }
    }
}public class SimpleTask implements Callable<String> {    @Override
    public String call() throws Exception {
        Thread.sleep(300);        //什麼都不作
        return null;
    }
}public class CpuTest {    public static void main(String[] args){
        ExecutorService executorService=Executors.newFixedThreadPool(11);        for (int i = 0; i < 10; i++) {
            executorService.submit(new SimpleTask());
        }
        executorService.submit(new HardTask());
    }
}

源碼很簡單，就是建立了2種線程HardTask和SimpleTask。而後經過CpuTest進行啓動。code

相關標籤/搜索

每日一句

每一个你不满意的现在，都有一个你没有努力的曾经。