image
服务CPU飙高了
项目运行时，JVM占用的CPU飙高，跟我们预期的不符，CPU在干什么呢？是什么让CPU如此忙碌呢？让我们来一次来追踪一下。
1、查找JVM进程ID：ps aux |grep java
2、根据pid查找占用CPU较高的线程：ps -mp pid -o THREAD,tid,time
[xxxxxxxxx-22027 ~]$ ps -mp 10605 -o THREAD,tid,time
USER %CPU PRI SCNT WCHAN USER SYSTEM TID TIME
root 697 - - - - - - 41-23:49:14
root 0.0 19 - futex_ - - 10605 00:00:00
root 0.0 19 - futex_ - - 10606 00:05:15
root 0.0 19 - poll_s - - 10607 00:00:03
root 99.6 33 - ? - - 10608 5-23:54:02
root 99.6 33 - ? - - 10609 6-00:01:38
root 99.6 33 - ? - - 10610 6-00:00:17
root 99.6 33 - ? - - 10611 6-00:01:01
root 99.6 33 - ? - - 10612 5-23:58:03
root 99.6 33 - ? - - 10613 5-23:54:01
root 99.6 33 - ? - - 10614 5-23:54:51
3、利用tid转换成16进制的数字：printf “%x\n” tid
[xxxxxxxxx-22027 ~]$ printf "%x\n" 10608
2970
4、使用jstack命令，查询线程信息，从而定位到具体线程和代码：jstack pid | grep 16进制数字 -A 30
checking pid(10605)
******************************************************************* ThreadId=0 **************************************************************************
2017-11-08 11:54:54
Full thread dump Java HotSpot(TM) 64-Bit Server VM (24.55-b03 mixed mode):
"Attach Listener" daemon prio=10 tid=0x00007ffa98002000 nid=0x28e29 waiting on condition [0x0000000000000000]
java.lang.Thread.State: RUNNABLE
"scheduler-10" prio=10 tid=0x00007ffa7000e000 nid=0x28c99 waiting on condition [0x00007ff9bbfbf000]
java.lang.Thread.State: WAITING (parking)
at sun.misc.Unsafe.park(Native Method)
- parking to wait for <0x00000000cc91c8c0> (a java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
at java.util.concurrent.locks.LockSupport.park(LockSupport.java:186)
at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(AbstractQueuedSynchronizer.java:2043)
at java.util.concurrent.ScheduledThreadPoolExecutor$DelayedWorkQueue.take(ScheduledThreadPoolExecutor.java:1085)
at java.util.concurrent.ScheduledThreadPoolExecutor$DelayedWorkQueue.take(ScheduledThreadPoolExecutor.java:807)
at java.util.concurrent.ThreadPoolExecutor.getTask(ThreadPoolExecutor.java:1068)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1130)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:745)
可直接查出原因。
觉得以上操作麻烦的话，可直接执行此脚本。
#!/bin/bash
#
# 当JVM占用CPU特别高时，查看CPU正在做什么
# 可输入两个参数：1、pid Java进程ID，必须参数 2、打印线程ID上下文行数，可选参数，默认打印10行
#
pid=$1
if test -z $pid
then
echo "pid can not be null!"
exit
else
echo "checking pid($pid)"
fi
if test -z "$(jps -l | cut -d '' -f 1 | grep $pid)"
then
echo "process of $pid is not exists"
exit
fi
lineNum=$2
if test -z $lineNum
then
lineNum=10
fi
jstack $pid >> "$pid".bak
ps -mp $pid -o THREAD,tid,time | sort -k2r | awk '{if ($1 !="USER" && $2 != "0.0" && $8 !="-") print $8;}' | xargs printf "%x\n" >> "$pid".tmp
tidArray="$( cat $pid.tmp)"
for tid in $tidArray
do
echo "******************************************************************* ThreadId=$tid **************************************************************************"
cat "$pid".bak | grep $tid -A $lineNum
done
rm -rf $pid.bak
rm -rf $pid.tmp
