Hadoop 启动 NameNode 报错 ERROR: Cannot set priority of namenode process 2639

> 本文由 [简悦 SimpRead](http://ksria.com/simpread/) 转码， 原文地址 [blog.csdn.net](https://blog.csdn.net/goTsHgo/article/details/120352654)

项目场景：
========

Hadoop 启动 NameNode 报错：：ERROR: Cannot set priority of namenode process

问题描述：
========

Hadoop 启动 NameNode 报错 ERROR: Cannot set priority of namenode process 2639

[atguigu@localhost333 logs]$ tail -100  hadoop-localhost -namenode-hadoop333.log

```
2021-09-17 16:41:25,656 INFO org.apache.localhost.util.ExitUtil: 
Exiting with status 1: org.apache.localhost.hdfs.server.common.InconsistentFSStateException: 
Directory /opt/ha/hadoop-3.1.3/data/name is in an inconsistent state: 
storage directory does not exist or is not accessible.
2021-09-17 16:41:25,663 INFO org.apache.hadoop.hdfs.server.namenode.NameNode: 
SHUTDOWN_MSG: 
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at hadoop333/192.168.1.103
```

原因分析：
========

首先查看日志：hadoop-localhost -namenode-hadoop103.log

命令如下：  
[localhost @hadoop333 logs]$ tail -100  hadoop-localhost -namenode-hadoop103.log

```
rm -rf /opt/ha/hadoop-3.1.3/data /opt/ha/hadoop-3.1.3/logs
 
sudo rm -rf /tmp/*
 
hdfs namenode -format
```

解决方案：
========

尝试将 CentOS7 重启，之后重新将 [hadoop](https://so.csdn.net/so/search?q=hadoop&spm=1001.2101.3001.7020) 中的 data 和 logs 还有系统根目录的 tmp 中的文件都删除，对 Hadoop 重新进行初始化问题解决。

```
rm -rf /opt/ha/hadoop-3.1.3/data /opt/ha/hadoop-3.1.3/logs
 
sudo rm -rf /tmp/*
 
hdfs namenode -format
```