一：CRS自启动

二：实例自启动

一：CRS自启动

默认情况下，crs是随着系统启动自动启动的。

查看版本

[root@rac01 bin]# ./crsctl query crs softwareversionOracle Clusterware version on node [rac01] is [11.2.0.4.0]

启动自启动(默认)

[root@rac01 bin]# ./crsctl enable crsCRS-4622: Oracle High Availability Services autostart is enabled.

查看

[root@rac01 bin]# cat /etc/oracle/scls_scr/rac01/root/ohasdstr enable

禁用自启动

[root@rac01 ./crsctl disable crsCRS-4621: Oracle High Availability Services autostart is disabled.

查看

[root@rac01 bin]# cat /etc/oracle/scls_scr/rac01/root/ohasdstr disable

二：实例自启动 

默认情况下,AUTO_START值为restore。

2.1 AUTO_START=restore

其中restore表示还原、恢复的含义，即将实例状态恢复到crs停止前的状态。

(1)如果crs停止前，手动通过sqlplus或srvctl停止了实例，那么在crs重启后，实例不会跟着自动启动。(2)如果crs停止前，实例处于open状态，那么重启crs(或操作系统重启)，实例会跟着crs启动而启动。注意：在没有停止实例前直接停止crs,实例是以abort方式停止的[Shutting down instance (abort)]

2.2 AUTO_START=always

将AUTO_START由restore改成always后，无论crs停止前实例是open还是close状态，实例都会跟着crs启动而启动。

2.1 AUTO_START=restore

测试AUTO_START=restore下，实例启动情况。

查看数据库信息

[root@rac01 bin]# ./crsctl start crs[root@rac01 bin]# ./srvctl config databasecjcdb[root@rac01 bin]# ./srvctl status database -d cjcdbInstance cjcdb1 is running on node rac01Instance cjcdb2 is running on node rac02

默认AUTO_START=restore

[root@rac01 bin]# ./crsctl stat res ora.cjcdb.db -p|grep AUTO_STARTAUTO_START=restore

测试：启动crs，实例没有自动启动(srvctl停库)

[root@rac01 bin]# ./crsctl stat res ora.cjcdb.db -p|grep AUTO_STARTAUTO_START=restore[root@rac02 bin]# ./crsctl stat res ora.cjcdb.db -p|grep AUTO_STARTAUTO_START=restore手动停止实例[root@rac01 bin]# ./srvctl stop database -d cjcdb###[root@rac01 bin]# ./srvctl stop instance -d cjcdb -i cjcdb1[root@rac01 bin]# ./crsctl stop crs[root@rac02 bin]# ./crsctl stop crs[root@rac01 bin]# ./crsctl start crs[root@rac02 bin]# ./crsctl start crs

查看实例状态

[root@rac01 bin]# ./crsctl stat res ora.cjcdb.db -t--------------------------------------------------------------------------------NAME           TARGET  STATE        SERVER                   STATE_DETAILS       --------------------------------------------------------------------------------Cluster Resources--------------------------------------------------------------------------------ora.cjcdb.db      1        OFFLINE OFFLINE                               Instance Shutdown         2        OFFLINE OFFLINE                               Instance Shutdown  

测试：启动crs,实例没有自动启动(SQLPLUS停库)

停止实例

SQL> shutdown immediate[root@rac01 bin]# ./crsctl stop crsSQL> shutdown immediate[root@rac02 bin]# ./crsctl stop crs

重启crs

[root@rac01 bin]# ./crsctl start crs[root@rac02 bin]# ./crsctl start crs

查看实例状态

[root@rac01 bin]# ./srvctl status database -d cjcdbInstance cjcdb1 is not running on node rac01Instance cjcdb2 is not running on node rac02

测试：启动crs，实例自动启动(srvctl)

启动实例

[root@rac01 bin]# ./srvctl start database -d cjcdb[root@rac01 bin]# ./srvctl status database -d cjcdbInstance cjcdb1 is running on node rac01Instance cjcdb2 is running on node rac02

重启crs

[root@rac01 bin]# ./crsctl stop crs[root@rac01 bin]# ./crsctl start crs[root@rac02 bin]# ./crsctl stop crs[root@rac02 bin]# ./crsctl start crs

查看实例状态

[root@rac01 bin]# ./crsctl stat res ora.cjcdb.db -t--------------------------------------------------------------------------------NAME           TARGET  STATE        SERVER                   STATE_DETAILS       --------------------------------------------------------------------------------Cluster Resources--------------------------------------------------------------------------------ora.cjcdb.db      1        ONLINE  ONLINE       rac01                    Open                      2        ONLINE  ONLINE       rac02                    Open 

查看实例状态

[root@rac01 bin]# ./srvctl status database -d cjcdbInstance cjcdb1 is running on node rac01Instance cjcdb2 is running on node rac02

启动crs，实例自动启动(sqlplus)

###节点1###SQL> startup###节点2###SQL> startup

重启crs

[root@rac01 bin]# ./crsctl stop crs[root@rac01 bin]# ./crsctl start crs[root@rac02 bin]# ./crsctl stop crs[root@rac02 bin]# ./crsctl start crs

查看实例状态

[root@rac01 bin]# ./srvctl status database -d cjcdbInstance cjcdb1 is running on node rac01Instance cjcdb2 is running on node rac02

2.2 AUTO_START=always

实例自启动

[root@rac01 bin]# ./crsctl stat res ora.cjcdb.db -p|grep AUTO_STARTAUTO_START=restore

实例1修改AUTO_START=always

[root@rac01 bin]# ./crsctl modify resource ora.cjcdb.db -attr "AUTO_START=always"[root@rac01 bin]# ./crsctl stat res ora.cjcdb.db -p|grep AUTO_STARTAUTO_START=always

查看实例状态

[root@rac01 bin]# ./srvctl status database -d cjcdbInstance cjcdb1 is not running on node rac01Instance cjcdb2 is not running on node rac02

重启crs

[root@rac01 bin]# ./crsctl stop crs[root@rac01 bin]# ./crsctl start crs

查看实例状态

[root@rac01 bin]# ./srvctl status instance -d cjcdb -i cjcdb1,cjcdb2 -f -vInstance cjcdb1 is running on node rac01. Instance status: Open.Instance cjcdb2 is not running on node rac02[root@rac01 bin]# ./srvctl status database -d cjcdbInstance cjcdb1 is running on node rac01Instance cjcdb2 is not running on node rac02

参考：

SRVCTL : Disabled DB Instance Status Changed to Online when ASM Restarted (Doc ID 2299664.1)Instance Does not Start After 11gR2 Grid Infrastructure Restart or Node Reboot (Doc ID 1149003.1)

更多数据库相关学习资料，可以查看我的ITPUB博客，网名chenoracle：

http://blog.itpub.net/29785807/