开关机步骤：

1． 停止Oracle10g 环境

$ lsnrctl stop (停止监听)
$ emctl stop dbconsole (停止dbconsole)
$ shutdown immediate (停止实例)
$ srvctl stop nodeapps -n gnd-rac01 (停节点服务)
# /etc/init.d/init.crs stop (root 停cluster 软件)

2． 启动Oracle10g 环境
$ srvctl start nodeapps -n gnd-rac01 （启动一些服务）
$ startup （启动实例）
$ emctl start dbconsole (启动dbconsole)
$ lsnrctl start (启动监听)

平时的检查：

1. 平时查看instance 状态（所有节点） ：
gnd-rac01</home/oracle>$srvctl status database -d tpc [注意：tpc 是db name]
Instance tpc1 is running on node gnd-rac01
Instance tpc2 is running on node gnd-rac02

2. 检查 crs（所有节点） :

gnd-rac01</>$cd /u01/product/crs/bin
gnd-rac01</u01/product/crs/bin>$ crs_stat -t
Name Type Target State Host
┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅┅
ora....01.lsnr application ONLINE ONLINE gnd-rac01
ora....c01.gsd application ONLINE ONLINE gnd-rac01
ora....c01.ons application ONLINE ONLINE gnd-rac01
ora....c01.vip application ONLINE ONLINE gnd-rac01
ora....02.lsnr application OFFLINE OFFLINE
ora....c02.gsd application ONLINE ONLINE gnd-rac02
ora....c02.ons application ONLINE ONLINE gnd-rac02
ora....c02.vip application ONLINE ONLINE gnd-rac02
ora.tpc.db application ONLINE ONLINE gnd-rac01
ora....c1.inst application ONLINE ONLINE gnd-rac01
ora....c2.inst application ONLINE ONLINE gnd-rac02

3. 检查VIP, GSD,ONS, Listener 等（所有节点）。
gnd-rac01</home/oracle>$srvctl status nodeapps -n gnd-rac01
VIP is running on node: gnd-rac01
GSD is running on node: gnd-rac01
Listener is running on node: gnd-rac01
ONS daemon is running on node: gnd-rac01

gnd-rac01</home/oracle>$srvctl status nodeapps -n gnd-rac02
VIP is running on node: gnd-rac02
GSD is running on node: gnd-rac02
Listener is not running on node: gnd-rac02
ONS daemon is running on node: gnd-rac02

4．检查Oracle Cluster 状态： CSS, CRS, EVM .
gnd-rac01</home/oracle>$crsctl check crs
CSS appears healthy
CRS appears healthy
EVM appears healthy

gnd-rac02</home/oracle>$crsctl check crs
CSS appears healthy
CRS appears healthy
EVM appears healthy

5．Ocfs 方面的问题可以到/etc/init.d/o2cb 这里看看。一般不用管理。