Oracle RAC 11.2（12C)正确关闭顺序

在11.2后的版本中crs被称为gi，通过crsctl stop cluster来关闭整个实例的资源非常方便，但是对于数据库实例需要单独关闭,stop cl

在11.2后的版本中crs被称为gi，通过crsctl stop cluster来关闭整个实例的资源非常方便，，但是对于数据库实例需要单独关闭,stop cluster对于实例直接以非正常的方式结束，环境是基于12C的。

正确的顺序:

先关闭数据库

[Oracle@rac1 ~]$ srvctl stop database -db tt

alert.log显示是通过immediate方式关闭.

Sat Apr 19 05:12:19 2014
 Shutting down instance (immediate)
 Stopping background process SMCO
 Shutting down instance: further logons disabled
 Stopping background process CJQ0
 Sat Apr 19 05:12:28 2014
 Killed process oracle@rac1 (QM05) with pid is 58, OS pid 26310
 Stopping background process MMNL
 Stopping background process MMON
 License high water mark = 8
 All dispatchers and shared servers shutdown
 ALTER DATABASE CLOSE NORMAL /* db agent *//* {1:28512:1593} */
 Sat Apr 19 05:12:33 2014
 SMON: disabling tx recovery
 Sat Apr 19 05:12:33 2014

确认一下配置oracle实例是不是immediate方式来关闭的。

[oracle@rac1 ~]$ srvctl config database -db tt
 Database unique name: tt
 Database name: tt
 Oracle home: /u01/app/oracle/product/12.1.0.0/db_1
 Oracle user: oracle
 Spfile: +DATA/tt/spfilett.ora
 Password file: +DATA/tt/orapwtt
 Domain:
Start options: open
 Stop options: immediate
 Database role: PRIMARY
 Management policy: AUTOMATIC
 Server pools: tt
 Database instances: tt1,tt2
 Disk Groups: DATA
 Mount point paths:
Services:
Type: RAC
 Start concurrency:
Stop concurrency:
Database is administrator managed
 [oracle@rac1 ~]$

错误的方法:

为图省事,直接关闭整个cluster

[root@rac1 ~]# crsctl stop cluster -all

CRS-2673: Attempting to stop 'ora.crsd' on 'rac1'
 CRS-2790: Starting shutdown of Cluster Ready Services-managed resources on 'rac1'
 CRS-2673: Attempting to stop 'ora.LISTENER.lsnr' on 'rac1'
 CRS-2673: Attempting to stop 'ora.OCR1.dg' on 'rac1'
 CRS-2673: Attempting to stop 'ora.tt.db' on 'rac1'
 CRS-2673: Attempting to stop 'ora.LISTENER_SCAN1.lsnr' on 'rac1'
 CRS-2673: Attempting to stop 'ora.cvu' on 'rac1'
 CRS-2677: Stop of 'ora.LISTENER.lsnr' on 'rac1' succeeded
 CRS-2673: Attempting to stop 'ora.rac1.vip' on 'rac1'
 CRS-2677: Stop of 'ora.LISTENER_SCAN1.lsnr' on 'rac1' succeeded
 CRS-2673: Attempting to stop 'ora.scan1.vip' on 'rac1'
 CRS-2677: Stop of 'ora.cvu' on 'rac1' succeeded
 CRS-2677: Stop of 'ora.tt.db' on 'rac1' succeeded
 CRS-2673: Attempting to stop 'ora.DATA.dg' on 'rac1'
 CRS-2677: Stop of 'ora.rac1.vip' on 'rac1' succeeded
 CRS-2677: Stop of 'ora.scan1.vip' on 'rac1' succeeded
 CRS-2673: Attempting to stop 'ora.crsd' on 'rac2'
 CRS-2790: Starting shutdown of Cluster Ready Services-managed resources on 'rac2'
 CRS-2673: Attempting to stop 'ora.tt.db' on 'rac2'
 CRS-2673: Attempting to stop 'ora.OCR1.dg' on 'rac2'
 CRS-2673: Attempting to stop 'ora.LISTENER.lsnr' on 'rac2'
 CRS-2677: Stop of 'ora.LISTENER.lsnr' on 'rac2' succeeded
 CRS-2673: Attempting to stop 'ora.rac2.vip' on 'rac2'
 CRS-2677: Stop of 'ora.rac2.vip' on 'rac2' succeeded
 CRS-2677: Stop of 'ora.DATA.dg' on 'rac1' succeeded
 CRS-2677: Stop of 'ora.tt.db' on 'rac2' succeeded
 CRS-2673: Attempting to stop 'ora.DATA.dg' on 'rac2'
 CRS-2677: Stop of 'ora.DATA.dg' on 'rac2' succeeded

..............................

alert.log显示实例是非常结束

License high water mark = 6
 Sat Apr 19 05:15:14 2014
 USER (ospid: 5792): terminating the instance
 Sat Apr 19 05:15:14 2014
 opiodr aborting process unknown ospid (3667) as a result of ORA-1092
 Sat Apr 19 05:15:15 2014
 Instance terminated by USER, pid = 5792

由此可以看出通过stop cluster来关闭实例会非常危险，直接是以非正常的方式来关闭实例的.
 

重启实例可以看到恢复操作
 
Sat Apr 19 05:31:20 2014
 Beginning crash recovery of 1 threads
  parallel recovery started with 2 processes
 Sat Apr 19 05:31:21 2014
 Started redo scan
 Sat Apr 19 05:31:21 2014
 Completed redo scan
  read 255 KB redo, 69 data blocks need recovery
 Sat Apr 19 05:31:21 2014
 Started redo application at
  Thread 2: logseq 2, block 3004
 Sat Apr 19 05:31:22 2014
 Recovery of Online Redo Log: Thread 2 Group 4 Seq 2 Reading mem 0
  Mem# 0: +DATA/tt/redo04.log
 Sat Apr 19 05:31:22 2014
 Completed redo application of 0.14MB
 Sat Apr 19 05:31:22 2014
 Completed crash recovery at
  Thread 2: logseq 2, block 3514, scn 1969283
  69 data blocks read, 69 data blocks written, 255 redo k-bytes read

推荐阅读：

CentOS 6.3(x32)下安装Oracle 10g R2

Linux-6-64下安装Oracle 12C笔记

本文永久更新链接地址：