记录一次CDHserver 无法启动问题（server已死，但PID文件存在）

**

公司集群CDH没有正常关闭，导致server无法启动问题

1 国庆前为防止公司集群出现意外，特将公司CDH集群关闭，（可能当时心情较为激动没有按照正常顺序先关agent再关server）
2 假后上班时开始启动集群后，打开 CM 主页面时报以下错误

当时没有太当回事，只是以为网络的问题，结果弄了一圈，还是不行，有点慌了，开始考虑到是公司集群出现了问题，那么开始找吧（先看了下前辈们的经验）

```cpp
 [root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server start 
Starting cloudera-scm-server:                              [  确定  ]
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server (pid  10280) 正在运行...
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server (pid  10280) 正在运行...
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server (pid  10280) 正在运行...
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server 已死，但 pid 文件存在
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server 已死，但 pid 文件存在
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server 已死，但 pid 文件存在
 找到的server的日志文件
 [root@hadoop1 ~]# cd /opt/cm-5.15.0/log/cloudera-scm-server/
[root@hadoop1 cloudera-scm-server]# tail -200f cloudera-scm-server.log

也可查看cloudera-scm-server.out

发现报错如下
![在这里插入图片描述](https://img-blog.csdnimg.cn/20201012140738874.png#pic_center)

```cpp
 INFO main:com.cloudera.enterprise.CommonMain: Statistics not enabled, JMX will not be registered
2020-10-12 10:11:15,978 WARN com.mchange.v2.async.ThreadPoolAsynchronousRunner$PoolThread-#2:com.mchange.v2.resourcepool.BasicResourcePool: com.mchange.v2.resourcepool.BasicResourcePool$AcquireTask@7f328323 -- Acquisition Attempt Failed!!! Clearing pending acquires. While trying to acquire a needed new resource, we failed to succeed more than the maximum number of allowed acquisition attempts (5). Last acquisition attempt exception: 
java.sql.SQLException: Access denied for user 'scm'@'hadoop1' (using password: YES)

从这里大致可以看见是数据库无法连接的问题，即server节点无法连接MySQL库，无法得到登陆信息，本猿一开始查到的资料是删除CM库再重新建一个CM库，但又怕集群又会出现其他问题，最后找到了一个大佬说是权限问题，于是尝试了一下更新权限
进入MySQL库找到USER表

GRANT ALL PRIVILEGES ON *.* TO 'scm'@'hadoop1' IDENTIFIED BY 'scm';
FLUSH PRIVILEGES;

执行命令
再次执行

root@hadoop1 ~]# cd /opt/cm-5.15.0/run/
[root@hadoop1 run]# ls 
cloudera-scm-agent  cloudera-scm-server  cloudera-scm-server.pid
[root@hadoop1 run]# rm cloudera-scm-server.pid 

rm：是否删除普通文件 "cloudera-scm-server.pid"？y
[root@hadoop1 run]# ll
总用量 0
drwxr-xr-x 6 cloudera-scm cloudera-scm 129 6月  24 09:27 cloudera-scm-agent
drwxr-xr-x 2 cloudera-scm cloudera-scm   6 5月  24 2018 cloudera-scm-server
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
已停止 cloudera-scm-server
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server start 
Starting cloudera-scm-server:                              [  确定  ]
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server (pid  28386) 正在运行...
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
cloudera-scm-server (pid  28386) 正在运行...
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status  
cloudera-scm-server (pid  28386) 正在运行...
[root@hadoop1 run]# /opt/cm-5.15.0/etc/init.d/cloudera-scm-server status 
（因为一开始server无法启动，有个PID	文件，一定要找到并删该文件，不然还会报错）

好了 问题解决（ps:第一次写有不足之处请大家见谅）