jstack(查看线程)、jmap(查看内存jvm)和jstat(性能分析)

jstack(查看线程)、jmap(查看内存)和jstat(性能分析)

1.使用top，查询出最耗内存的cpu对应的pid

/opt/java8/bin/jstack

Usage:
    jstack [-l] <pid>   举例：jstack -l 8748
        (to connect to running process) 连接活动线程
    jstack -F [-m] [-l] <pid>
        (to connect to a hung process) 连接阻塞线程
    jstack [-m] [-l] <executable> <core>
        (to connect to a core file) 连接dump的文件
    jstack [-m] [-l] [server_id@]<remote server IP or hostname>
        (to connect to a remote debug server) 连接远程服务器

Options:
    -F  to force a thread dump. Use when jstack <pid> does not respond (process is hung)
    -m  to print both java and native frames (mixed mode)
    -l  long listing. Prints additional information about locks
    -h or -help to print this help message

2.查询pid下有多少个线程：jstack -l pid | grep 'java.lang.Thread.State' | wc -l    显示 'java.lang.Thread.State'的行数

举例：jstack -l 8748  | grep 'java.lang.Thread.State' | wc -l   --->返回线程数 266

3.把jstack记录保存到文件里 ：

cd  /log/java/push-service

jstack -l pid >> 123.txt , rz 上传命令，sz 123.txt 下载命令

jstack 12652 | head -100

4.查看当前进程消耗资源的线程pid top -Hp pid

5 将线程Pid转为16进制，根据该16进制值去打印的堆栈日志内查询，查看该线程所驻留的方法位置。

printf "%x\n" 17880          
45d8

（基本上这步已经能够定位大多数问题了）

6.所在行的后面30行

[app@push-service42-N-1586391544-1-uc push-service]$ jstack 3608|grep f76 -A 30

"pool-13-thread-1" #234 prio=5 os_prio=0 tid=0x00007fc03c098f40 nid=0xf76 waiting on condition [0x00007fbff3089000]
   java.lang.Thread.State: WAITING (parking)
    at sun.misc.Unsafe.park(Native Method)
    - parking to wait for  <0x00000006c4509b00> (a java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
    at java.util.concurrent.locks.LockSupport.park(LockSupport.java:175)
    at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(AbstractQueuedSynchronizer.java:2039)
    at java.util.concurrent.LinkedBlockingQueue.take(LinkedBlockingQueue.java:442)
    at java.util.concurrent.ThreadPoolExecutor.getTask(ThreadPoolExecutor.java:1067)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1127)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
    at java.lang.Thread.run(Thread.java:745)

5 若依然无法定位问题，可以查看堆中对象数量和大小 jmap -histo pid

jmap -histo:live 12240 | head -30   !!!不要随意执行，会垃圾回收
jmap 12652 | head -30      前30行
 

下载dump文件：

另外可以通过如下命令生成堆dump文件拿到本地通过jdk/bin下自带的jvisualvm.exe打开图形化显示详细的内存信息。

jmap -dump:format=b,file=heapdump pid

jmap -dump:live,format=b,file=dump20191020.hprof 30623 

会下载到 /tmp/hsperfdata_app/
format=b指定为二进制格式文件

[app@push-service42-N-1586391544-1-uc push-service]$ jmap -histo 3608 |head -30

 num     #instances         #bytes  class name
----------------------------------------------
   1:       3881673      476282408  [C
   2:       1221885      275547624  [B
   3:         10497       78637560  [J
   4:        466593       73193440  [I
   5:       1052986       69120064  [Ljava.lang.Object;
   6:       2123613       50966712  java.lang.String
   7:        373472       23329856  [Ljava.util.HashMap$Node;
   8:        576545       18449440  java.util.LinkedHashMap$LinkedKeyIterator
   9:        418706       16748240  java.util.LinkedHashMap$Entry
  10:        338957       16269936  java.nio.HeapByteBuffer
  11:        271442       15200752  java.util.LinkedHashMap
  12:        632188       15172512  java.util.concurrent.CompletableFuture
  13:        190638       12200832  java.util.regex.Matcher
  14:        499751       11994024  java.util.ArrayList
  15:        361516       11568512  java.util.HashMap$Node
  16:        226204       10857792  java.nio.HeapCharBuffer

jstack命令解析 - 简书