[JAVA]记一次内存溢出

  1. 异常发现
    1. 账号登录不上解决方法
    2. 堆栈分析
    3. 代码追踪
    4. 解决

项目部署到服务器后,第二天发现异常,记录下排查过程。

异常发现

带有权限的账号,突然ssh连接不上服务器了,项目也出现访问不了的情况。后面尝试用root账号,意外的可以登录上。

top
jps

发现CPU没啥明显的异常,项目进程也没有挂。于是进入项目,查看日志,发现日志中存在异常。

2019-12-15 11:36:58.200 default [nioEventLoopGroup-4-2] ERROR o.r.client.handler.CommandsQueue - Exception occured. Channel: [id: 0x2f80895c, L:/172.16.149.12:60435 - R:172.16.149.13/172.16.149.13:6379]
java.lang.OutOfMemoryError: unable to create new native thread
    at java.lang.Thread.start0(Native Method)
    at java.lang.Thread.start(Thread.java:717)
    at io.netty.util.concurrent.ThreadPerTaskExecutor.execute(ThreadPerTaskExecutor.java:33)
    at io.netty.util.internal.ThreadExecutorMap$1.execute(ThreadExecutorMap.java:57)
    at io.netty.util.concurrent.SingleThreadEventExecutor.doStartThread(SingleThreadEventExecutor.java:895)
    at io.netty.util.concurrent.SingleThreadEventExecutor.startThread(SingleThreadEventExecutor.java:866)
    at io.netty.util.concurrent.SingleThreadEventExecutor.execute(SingleThreadEventExecutor.java:759)
    at io.netty.util.concurrent.AbstractScheduledEventExecutor.schedule(AbstractScheduledEventExecutor.java:232)
    at io.netty.util.concurrent.AbstractScheduledEventExecutor.schedule(AbstractScheduledEventExecutor.java:155)
    at io.netty.util.concurrent.AbstractEventExecutorGroup.schedule(AbstractEventExecutorGroup.java:50)
    at org.redisson.client.RedisConnection.async(RedisConnection.java:205)
    at org.redisson.client.RedisConnection.async(RedisConnection.java:191)
    at org.redisson.client.RedisConnection.async(RedisConnection.java:183)
    at org.redisson.client.handler.BaseConnectionHandler.channelActive(BaseConnectionHandler.java:68)

竟然内存溢出了,开始了一段分析过程。


报错为无法创建线程unable to create new native thread。使用命令查看下是否有问题。

pstree -a    //查看所有进程相关的线程
pstree -p PID|wc -l    //查看指定进程的线程数

发现一个项目竟然创建了接近3000线程。能大致推出这是代码有问题,线程未释放导致溢出,但是为什么服务器有权限的账号在发生异常时登录不上了,而root可以呢?

账号登录不上解决方法

把服务kill之后,发现用权限的账户可以登录了,原因在于最初时用该账户执行的项目,那么也就是说跟权限配置有关了。用ulimit -a查看下,发现max user processes只有4096,这样一算的话启动那些java服务加起来创建的线程数正好就超过这个数了,提高该配置之后,未发现之前出现的权限用户登录不上的情况。

ulimit -a

---
core file size          (blocks, -c) unlimited
data seg size           (kbytes, -d) unlimited
scheduling priority             (-e) 0
file size               (blocks, -f) unlimited
pending signals                 (-i) 63459
max locked memory       (kbytes, -l) 32000
max memory size         (kbytes, -m) unlimited
open files                      (-n) 1000000
pipe size            (512 bytes, -p) 8
POSIX message queues     (bytes, -q) 8192000
real-time priority              (-r) 0
stack size              (kbytes, -s) 8192
cpu time               (seconds, -t) unlimited
max user processes              (-u) 4096
virtual memory          (kbytes, -v) unlimited
file locks                      (-x) unlimited

堆栈分析

为了排查错误,项目开着,发现线程一直在慢慢累加,最终到了10000多,不再继续,但也没有释放。linux中jstack分析下线程堆栈情况。

jstack -l PID > jstack.log

日志

Full thread dump Java HotSpot(TM) 64-Bit Server VM (25.212-b10 mixed mode):

"nioEventLoopGroup-1240-13" #13017 prio=10 os_prio=0 tid=0x00007f1278f46000 nid=0x374 runnable [0x00007f1264b4a000]
   java.lang.Thread.State: RUNNABLE
    at sun.nio.ch.EPollArrayWrapper.epollWait(Native Method)
    at sun.nio.ch.EPollArrayWrapper.poll(EPollArrayWrapper.java:269)
    at sun.nio.ch.EPollSelectorImpl.doSelect(EPollSelectorImpl.java:93)
    at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:86)
    - locked <0x0000000768f9f6d8> (a io.netty.channel.nio.SelectedSelectionKeySet)
    - locked <0x0000000768f9f6c8> (a java.util.Collections$UnmodifiableSet)
    - locked <0x0000000768f9f6f0> (a sun.nio.ch.EPollSelectorImpl)
    at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:97)
    at io.netty.channel.nio.SelectedSelectionKeySetSelector.select(SelectedSelectionKeySetSelector.java:62)
    at io.netty.channel.nio.NioEventLoop.select(NioEventLoop.java:791)
    at io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:439)
    at io.netty.util.concurrent.SingleThreadEventExecutor$5.run(SingleThreadEventExecutor.java:906)
    at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
    at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
    at java.lang.Thread.run(Thread.java:748)

   Locked ownable synchronizers:
    - None

"nioEventLoopGroup-1240-16" #13016 prio=10 os_prio=0 tid=0x00007f15700d4800 nid=0x373 runnable [0x00007f1264c4b000]
   java.lang.Thread.State: RUNNABLE
    at sun.nio.ch.EPollArrayWrapper.epollWait(Native Method)
    at sun.nio.ch.EPollArrayWrapper.poll(EPollArrayWrapper.java:269)
    at sun.nio.ch.EPollSelectorImpl.doSelect(EPollSelectorImpl.java:93)
    at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:86)
    - locked <0x0000000768fa15e8> (a io.netty.channel.nio.SelectedSelectionKeySet)
    - locked <0x0000000768fa15d8> (a java.util.Collections$UnmodifiableSet)
    - locked <0x0000000768fa1600> (a sun.nio.ch.EPollSelectorImpl)
    at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:97)
    at io.netty.channel.nio.SelectedSelectionKeySetSelector.select(SelectedSelectionKeySetSelector.java:62)
    at io.netty.channel.nio.NioEventLoop.select(NioEventLoop.java:791)
    at io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:439)
    at io.netty.util.concurrent.SingleThreadEventExecutor$5.run(SingleThreadEventExecutor.java:906)
    at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
    at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
    at java.lang.Thread.run(Thread.java:748)

   Locked ownable synchronizers:
    - None

"nioEventLoopGroup-1240-14" #13015 prio=10 os_prio=0 tid=0x00007f1578006800 nid=0x362 runnable [0x00007f1264d4c000]
   java.lang.Thread.State: RUNNABLE
    at sun.nio.ch.EPollArrayWrapper.epollWait(Native Method)
    at sun.nio.ch.EPollArrayWrapper.poll(EPollArrayWrapper.java:269)
    at sun.nio.ch.EPollSelectorImpl.doSelect(EPollSelectorImpl.java:93)
    at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:86)
    - locked <0x0000000768f9d840> (a io.netty.channel.nio.SelectedSelectionKeySet)
    - locked <0x0000000768f9d830> (a java.util.Collections$UnmodifiableSet)
    - locked <0x0000000768f9d858> (a sun.nio.ch.EPollSelectorImpl)
    at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:97)
    at io.netty.channel.nio.SelectedSelectionKeySetSelector.select(SelectedSelectionKeySetSelector.java:62)
    at io.netty.channel.nio.NioEventLoop.select(NioEventLoop.java:791)
    at io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:439)
    at io.netty.util.concurrent.SingleThreadEventExecutor$5.run(SingleThreadEventExecutor.java:906)
    at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
    at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
    at java.lang.Thread.run(Thread.java:748)

   Locked ownable synchronizers:
    - None

以上,能看到nioEventLoopGroup一直处于运行状态,为了更加直观的查看,我本地运行,使用jdk自带的jvm分析工具(在jdk安装目录,bin目录下的jvisualvm.exe),,发现随着时间的推移,nioEventLoopGroup创建数量越来越多。

代码追踪

查询代码后,发现有个定时任务中循环创建了redisson,而每次创建琐时都会new NioEventLoopGroup();正是因为这个每次创建后一直处于执行中,JVM垃圾回收并不会处理,所以溢出了。

解决

我的处理方法
1、初始化之创建一个线程池。

private static NioEventLoopGroup nioEventLoopGroup;

private static NioEventLoopGroup getNioEventLoopGroup() {
        if (nioEventLoopGroup == null) {
            nioEventLoopGroup = new NioEventLoopGroup();
        }
        return nioEventLoopGroup;
    }

2、或者释放

nioEventLoopGroup.shutdownGracefully();

转载请注明来源,欢迎对文章中的引用来源进行考证,欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论,也可以邮件至 xiaoxin1218@qq.com

文章标题:[JAVA]记一次内存溢出

文章字数:1.4k

本文作者:周信

发布时间:2019-12-21, 15:03:40

最后更新:2023-05-03, 10:25:35

原始链接:http://zx21.xyz/2019/12/21/JAVA-记一次内存溢出/

版权声明: "署名-非商用-相同方式共享 4.0" 转载请保留原文链接及作者。

目录
×

可为我打赏鼓励~