某些时候我们需要到客户现场去做保障，偶尔会需要停下数据库集群，然后停机。通常我们会采取比较稳妥的方式，而不是野蛮暴力的方式。

1. 野蛮的方式
   

# /u01/app/oracle/product/10.2.0/db_1/bin/crsctl stop crs

日志

Shutting down instance (abort)

License high water mark = 5

Instance terminated by USER,pid = 9068

2.温柔的方式

[root@node1 root]# su - oracle

[oracle@node1 ~]$ crs_stat -t -v

[oracle@node1 ~]$ srvctl config database

racdb

[oracle@node1 ~]$ srvctl stop database -d racdb

[oracle@node1 ~]$ srvctl stop asm -n node1

[oracle@node1 ~]$ srvctl stop asm -n node2

[oracle@node1 ~]$ srvctl stop nodeapps -n node1

[oracle@node1 ~]$ srvctl stop nodeapps -n node2

[root@node1 root]# /u01/app/oracle/product/10.2.0/db_1/bin/crsctl stop crs

Stopping resources. This could take several minutes.

Successfully stopped CRS resources.

Stopping CSSD.

Shutting down CSS daemon.

Shutdown request successfully issued.

[root@node1 root]# shutdown -h 0

Broadcast message from root (pts/1) (Thu Jul 20 09:16:28 2017):

The system is going down for system halt NOW!

3.启动方法

[root@node1 ~]# su - oracle

[oracle@node1 ~]$ which crsctl

/u01/app/oracle/product/10.2.0/db_1/bin/crsctl

[oracle@node1 ~]$ exit

logout

[root@node1 ~]# /u01/app/oracle/product/10.2.0/db_1/bin/crsctl start crs

Attempting to start CRS stack

The CRS stack will be started shortly

最近有点忙，有些懈怠了，博客更新的也不及时了。还是要继续坚持才可以，少找借口，自我批评一下。