hadoop: unable to create new native thread

1.系统环境

CentOS6.4 Hadoop0.20.2  伪分布式

2.问题描述

在执行MapReduce程序时，在控制台出现unable to create new native thread。查看datanode日志其错误信息如下：

ERROR org.apache.hadoop.hdfs.server.datanode.DataNode: DatanodeRegistration(127.0.0.1:50010, storageID=DS-1204083579-127.0.0.1-50010-1395973569073, infoPort=50075, ipcPort=50020):DataXceiver
java.io.EOFException: while trying to read 65557 bytes

3.解决方法

In certain recent Linux distributions (like RHEL v6.x/CentOS v6.x), the default value of nproc is lower than the value required if you are deploying the HBase service. To change this value:

1. Using a text editor, open /etc/security/limits.d/90-nproc.confand change the nproc limit to approximately 32000.

2. Restart the Hadoop.

修改前nproc的值

修改后nproc的值