ITPub博客
首页 博客 专栏 学院 社区 IT新媒体评选
发博

首页 > 数据库 > Oracle > 10GR2 RAC 更换存储

10GR2 RAC 更换存储

原创 Oracle 作者：chinaboysme 时间：2015-06-29 16:32:27 0 删除 编辑
 

1         做数据库RMAN完全备份

run

{

configure retention policy to recovery window of 25 days;

ALLOCATE CHANNEL c1 DEVICE disk;

ALLOCATE CHANNEL c2 DEVICE disk;

ALLOCATE CHANNEL c3 DEVICE disk;

ALLOCATE CHANNEL c4 DEVICE disk;

crosscheck backup;

crosscheck archivelog all;

backup database format 'ldsc_whole_db%U%T.bak';

sql 'ALTER SYSTEM switch logfile';

backup archivelog all format 'arch_bak%d_%s_%p_%T' not backed up 1 times;

backup current controlfile format 'ctrl_bak%d_%I_%s_%p_%T';

backup spfile format 'spfile_%T_%I.ora';

DELETE NOPROMPT OBSOLETE;

DELETE NOPROMPT EXPIRED BACKUP;

release channel c1;

release channel c2;

release channel c3;

release channel c4;

}

有条件一并做一次逻辑导出备份

2         基本结构

2.1     旧存储裸设备对应关系

查看磁盘ID的指令：

#ls –al /dev/disk/by-id

查看分区的UUID

#ls –al /dev/disk/by-uuid

编号

裸设备名称

磁盘名称

分区名称

磁盘ID

1

/dev/raw/raw1

/dev/sdb

/dev/sdb1

scsi-36000c294e3b550478be9c392c3aaefa6

2

/dev/raw/raw2

/dev/sdc

/dev/sdc1

scsi-36000c296c9f56c2fc88deceb5317d68b

3

/dev/raw/raw3

/dev/sdd

/dev/sdd1

scsi-36000c299c05893c4c2f923909f6ef94c

4

/dev/raw/raw4

/dev/sde

/dev/sde1

scsi-36000c29d65d6d3e89563e3dfc7f3faa8

2.2     旧裸设备与角色对应关系

编号

裸设备名称

属主

权限

裸设备角色

备注

1

/dev/raw/raw1

root:oinstall

640

OCR

2

/dev/raw/raw2

oracle:oinstall

644

VOTE DISK

3

/dev/raw/raw3

oracle:oinstall

660

Member of DATA

4

/dev/raw/raw4

oracle:oinstall

660

Member of DATA

2.3     新存储裸设备对应关系

编号

裸设备名称

磁盘名称

分区名称

磁盘ID

1

/dev/raw/raw5

/dev/sdf

/dev/sdf1

scsi-36000c29f7ca6341614886693a59aeec8

2

/dev/raw/raw6

/dev/sdg

/dev/sdg1

scsi-36000c292a9ebe7d24f2d94542e6c9824

3

/dev/raw/raw7

/dev/sdh

/dev/sdh1

scsi-36000c29322a954b4c677cf15e685b6b7

4

/dev/raw/raw8

/dev/sdi

/dev/sdi1

scsi-36000c2972866199de5aa51b136e90e5e

5

/dev/raw/raw9

/dev/sdj

/dev/sdj1

scsi-36000c29f2556d5adfaf30cfba157569b

6

/dev/raw/raw10

/dev/sdk

/dev/sdk1

scsi-36000c297fa7fbce9f5fa36422a089e4e

2.4     新裸设备与角色对应关系

编号

裸设备名称

属主

权限

裸设备角色

备注

1

/dev/raw/raw5

root:oinstall

640

ocr

2

/dev/raw/raw6

oracle:oinstall

644

Vote disk

3

/dev/raw/raw7

oracle:oinstall

660

DATA1

4

/dev/raw/raw8

oracle:oinstall

660

DATA1

5

/dev/raw/raw9

oracle:oinstall

660

ARC

6

/dev/raw/raw10

oracle:oinstall

660

ARC

3         集群结构迁移

3.1     迁移votedisk

      

3.1.1      查看当前votedisk情况

     查看当前集群环境使用的votedisk

     Oracle用户：

    [oracle@ora1 ~]$ crsctl query css votedisk

      0.     0    /dev/raw/raw2

located 1 votedisk(s).

3.1.2      停止集群的两个节点

     root用户执行：

     node1：

[root@ora1 ~]# /etc/init.d/init.crs stop

Shutting down Oracle Cluster Ready Services (CRS):

May 25 16:48:12.961 | INF | daemon shutting down

Stopping resources. This could take several minutes.

Successfully stopped CRS resources.

Stopping CSSD.

Shutting down CSS daemon.

Shutdown request successfully issued.

Shutdown has begun. The daemons should exit soon.    

node2:

[root@ora2 raw]# /etc/init.d/init.crs stop

Shutting down Oracle Cluster Ready Services (CRS):

May 25 16:49:20.379 | INF | daemon shutting down

Stopping resources. This could take several minutes.

Successfully stopped CRS resources.

Stopping CSSD.

Shutting down CSS daemon.

Shutdown request successfully issued.

Shutdown has begun. The daemons should exit soon.

   

3.1.3      备份vote盘

     这里备份vote盘到文件

[root@ora1 bin]# dd if=/dev/raw/raw2 of=/home/oracle/votediskbackup/vote.20150525 bs=1M

499+1 records in

499+1 records out

524271616 bytes (524 MB) copied, 13.0158 seconds, 40.3 MB/s

3.1.4      添加新存储中盘作为新的vote盘

     root用户执行：

     ORA_CRS_HOME目录bin目录下执行：

注意这里不加-force参数的时候，执行会报cluster没有准备

    [root@ora1 bin]# ./crsctl add css votedisk /dev/raw/raw6

Cluster is not in a ready state for online disk addition

正确的语句应该是这样的

[root@ora1 bin]# ./crsctl add css votedisk /dev/raw/raw6 -force

Now formatting voting disk: /dev/raw/raw6

successful addition of votedisk /dev/raw/raw6.

查看添加后的vote盘

      root用户执行：

    [root@ora1 bin]# ./crsctl query css votedisk

 0.     0    /dev/raw/raw2

 1.     0    /dev/raw/raw6

located 2 votedisk(s).

3.1.5      启动两个节点

     root用户执行

     node1：

    [root@ora1 bin]# /etc/init.d/init.crs start

Startup will be queued to init within 30 seconds.

     node2:

    [root@ora2 bin]# /etc/init.d/init.crs start

Startup will be queued to init within 30 seconds.

3.1.6      检查启动情况

     oracle用户执行

  [oracle@ora1 ~]$ crs_stat -t

Name           Type           Target    State     Host       

------------------------------------------------------------

ora....z1.inst application    ONLINE    ONLINE    ora1       

ora....z2.inst application    ONLINE    ONLINE    ora2       

ora.cbnz.db    application    ONLINE    ONLINE    ora2       

ora.etax.db    application    ONLINE    ONLINE    ora2       

ora....x1.inst application    ONLINE    ONLINE    ora1       

ora....x2.inst application    ONLINE    ONLINE    ora2       

ora....SM1.asm application    ONLINE    ONLINE    ora1       

ora....A1.lsnr application    ONLINE    ONLINE    ora1       

ora.ora1.gsd   application    ONLINE    ONLINE    ora1       

ora.ora1.ons   application    ONLINE    ONLINE    ora1       

ora.ora1.vip   application    ONLINE    ONLINE    ora1       

ora....SM2.asm application    ONLINE    ONLINE    ora2       

ora....A2.lsnr application    ONLINE    ONLINE    ora2       

ora.ora2.gsd   application    ONLINE    ONLINE    ora2       

ora.ora2.ons   application    ONLINE    ONLINE    ora2       

ora.ora2.vip   application    ONLINE    ONLINE    ora2    

3.1.7      停止两节点

     root用户执行

     node1：

     /etc/init.d/init.crs stop

     node2:

     /etc/init.d/init.crs stop

3.1.8      删除老的vote盘

     这里一定要注意不要删错盘

     root用户执行

    [root@ora1 bin]# ./crsctl delete css votedisk /dev/raw/raw2 -force

successful deletion of votedisk /dev/raw/raw2.

3.1.9      两节点启动crs

     root用户执行

     node1：

     /etc/init.d/init.crs start

     node2:

     /etc/init.d/init.crs start

3.1.10 检查启动情况

     oracle用户执行

    [oracle@ora1 ~]$ crs_stat -t

Name           Type           Target    State     Host       

------------------------------------------------------------

ora....z1.inst application    ONLINE    ONLINE    ora1       

ora....z2.inst application    ONLINE    ONLINE    ora2       

ora.cbnz.db    application    ONLINE    ONLINE    ora1       

ora.etax.db    application    ONLINE    ONLINE    ora2       

ora....x1.inst application    ONLINE    ONLINE    ora1       

ora....x2.inst application    ONLINE    ONLINE    ora2       

ora....SM1.asm application    ONLINE    ONLINE    ora1       

ora....A1.lsnr application    ONLINE    ONLINE    ora1       

ora.ora1.gsd   application    ONLINE    ONLINE    ora1       

ora.ora1.ons   application    ONLINE    ONLINE    ora1       

ora.ora1.vip   application    ONLINE    ONLINE    ora1       

ora....SM2.asm application    ONLINE    ONLINE    ora2       

ora....A2.lsnr application    ONLINE    ONLINE    ora2       

ora.ora2.gsd   application    ONLINE    ONLINE    ora2       

ora.ora2.ons   application    ONLINE    ONLINE    ora2       

ora.ora2.vip   application    ONLINE    ONLINE    ora2       

3.1.11 检查vote盘的情况

     root用户执行

    [oracle@ora1 ~]$ crsctl query css votedisk

 0.     0    /dev/raw/raw6

located 1 votedisk(s).

3.2     迁移OCR磁盘

3.2.1      查看OCR磁盘情况

oracle用户执行

[oracle@ora1 ~]$ ocrcheck

Status of Oracle Cluster Registry is as follows :

         Version                  :          2

         Total space (kbytes)     :     511744

         Used space (kbytes)      :       4636

         Available space (kbytes) :     507108

         ID                       :  350260548

         Device/File Name         : /dev/raw/raw1

                                    Device/File integrity check succeeded

                                    Device/File not configured

         Cluster registry integrity check succeeded

     root用户执行

    [root@ora1 bin]# ./ocrcheck

Status of Oracle Cluster Registry is as follows :

         Version                  :          2

         Total space (kbytes)     :     511744

         Used space (kbytes)      :       4636

         Available space (kbytes) :     507108

         ID                       :  350260548

         Device/File Name         : /dev/raw/raw1

                                    Device/File integrity check succeeded

                                    Device/File not configured

         Cluster registry integrity check succeeded

3.2.2      在线备份OCR

     root用户执行

     注意这里是在线备份

     CRS宿主目录下执行

   [root@ora1 bin]# ./ocrconfig -export /home/oracle/ocrdiskbackup/ocr.disk.20150526 -s online

3.2.3      迁移

 

3.2.3.1            迁移方法一

     改方法是编辑两个节点的/etc/oracle/ocr.loc文件，将ocrconfig_loc的位置/dev/raw/old_ocr_disk改为新存储的/dev/raw/new_ocr_disk

    

3.2.3.1.1    关闭集群

     root用户执行

     node1

     /etc/init.d/init.crs stop

     node2

     /etc/init.d/init.crs stop

    

3.2.3.1.2    导入配置并检查

     root用户执行

     ./ocrconfig -import /home/oracle/ocrback/ocr.bak20120315

     ./ocrcheck

3.2.3.1.3    启动集群并检查状态

     root用户执行

     node1

     /etc/init.d/init.crs start

     node2

     /etc/init.d/init.crs start

     Oracle用户执行

     crs_stat -t

3.2.3.2            迁移方法二

     该方法为给OCR所在磁盘做镜像，注意此时集群并没有停。

3.2.3.2.1    先做镜像

     [root@ora1 bin]# ./ocrconfig -replace ocrmirror /dev/raw/raw5

3.2.3.2.2    检查ocr

   [root@ora1 bin]# ./ocrcheck

Status of Oracle Cluster Registry is as follows :

         Version                  :          2

         Total space (kbytes)     :     511744

         Used space (kbytes)      :       4636

         Available space (kbytes) :     507108

         ID                       :  350260548

         Device/File Name         : /dev/raw/raw1

                                    Device/File integrity check succeeded

         Device/File Name         : /dev/raw/raw5

                                    Device/File integrity check succeeded

         Cluster registry integrity check succeeded

3.2.3.2.3    替换OCR

[root@ora1 bin]# ./ocrconfig -replace ocr   

//该指令是删除OCR设备，这是镜像OCR设备自动转换为OCR设备

[root@ora1 bin]# ./ocrcheck

Status of Oracle Cluster Registry is as follows :

         Version                  :          2

         Total space (kbytes)     :     511744

         Used space (kbytes)      :       4636

         Available space (kbytes) :     507108

         ID                       :  350260548

         Device/File Name         : /dev/raw/raw5

                                    Device/File integrity check succeeded

                                    Device/File not configured

         Cluster registry integrity check succeeded

此时重新启动下集群，检验集群的状态。

4         调整ASM磁盘组

4.1     创建新的ASM磁盘组

4.1.1      启动两个实例到mount状态

  先关闭数据库的两个实例

  oracle用户执行

  node1

  shutdown immediate

  startup mount

  node2

  shutdown immediate

  startup mount

4.1.2      添加asm磁盘发现位置

此步骤不用执行

  node1

  SQL> alter system set asm_diskstring='/dev/raw/raw*','/dev/raw/new_raw*';

4.1.3      创建磁盘组

 SQL> create diskgroup arc external redundancy disk '/dev/raw/raw9' name ARC_0000,'/dev/raw/raw10' name ARC_0001;

SQL> select name,state from v$asm_disk;

NAME                           STATE

------------------------------ ----------------

                               NORMAL

                               NORMAL

                               NORMAL

                               NORMAL

ARC_0001                       NORMAL

ARC_0000                       NORMAL

DATA_0001                      NORMAL

DATA_0000                      NORMAL

4.1.4      第二节点挂载该磁盘组

  node2

SQL> /

NAME                 STATE

-------------------- ----------------------

DATA                 MOUNTED

ARC                  MOUNTED

DATA                 MOUNTED

ARC                  DISMOUNTED

SQL> alter diskgroup arc mount;

Diskgroup altered.

SQL> select name,state from gv$asm_diskgroup;

NAME                 STATE

-------------------- ----------------------

DATA                 MOUNTED

ARC                  MOUNTED

DATA                 MOUNTED

ARC                  MOUNTED

  注意必须要在pfile中添加asm_diskgroups参数的更改值。

重新启动下crs，测试ARC能不能自动的挂载（mount）,经测试确实不能自动挂载

需要修改ASM的参数文件：

more init+ASM1.ora

asm_diskgroups='DATA','ARC'

这样在启动ASM实例时会自动挂载ARC卷组

4.1.5      在ASM中手动创建所需的目录

  注意这里要手动在ASM中创建如下需要的目录

  DATA1/数据库名称/DATAFILE

  DATA1/数据库名称/ARCHIVELOG

  DATA1/数据库名称/CONTROLFILE

  DATA1/数据库名称/ONLINELOG

  DATA1/数据库名称/PARAMETERFILE

  DATA1/数据库名称/TEMPFILE

ASMCMD> cd DATA1

ASMCMD> ls

ASMCMD> mkdir ETAX

ASMCMD> mkdir CBNZ

ASMCMD> ls

CBNZ/

ETAX/

ASMCMD> cd etax

ASMCMD> ls

ASMCMD> mkdir DATAFILE

ASMCMD> mkdir ARCHIVELOG

ASMCMD> mkdir CONTROLFILE

ASMCMD> mkdir ONLINELOG

ASMCMD> mkdir PARAMETERFILE

ASMCMD> mkdir TEMPFILE

5         数据库迁移

5.1     迁移数据文件

！！！！！！！！！！！做所有操作前做spfile的备份

5.1.1      查询数据文件的名称

SQL> select FILE#,NAME from v$datafile;

     FILE# NAME

---------- ---------------------------------------------

         1 +DATA/etax/datafile/system.256.880644513

         2 +DATA/etax/datafile/undotbs1.258.880644515

         3 +DATA/etax/datafile/sysaux.257.880644513

         4 +DATA/etax/datafile/users.259.880644515

         5 +DATA/etax/datafile/undotbs2.264.880644613

5.1.2      关闭数据库并启动数据到mount状态

srvctl stop database –d etax

srvctl start database –d etax –o mount

5.1.3      用RMAN复制数据文件

rman target / nocatalog

RMAN> copy datafile 1 to '+DATA1/etax/datafile/system01.dbf';

RMAN>copy datafile 2 to '+DATA1/etax/datafile/undotbs01.dbf';

RMAN>copy datafile 3 to '+DATA1/etax/datafile/sysaux01.dbf';

RMAN>copy datafile 4 to '+DATA1/etax/datafile/user01.dbf';

RMAN>copy datafile 5 to '+DATA1/etax/datafile/undotbs02.dbf';

5.1.4      更换数据文件的位置

登录数据库，更换数据文件的位置

alter database rename file '+DATA/etax/datafile/system.256.880644513' to '+DATA1/etax/datafile/system01.dbf';

alter database rename file '+DATA/etax/datafile/undotbs1.258.880644515' to '+DATA1/etax/datafile/undotbs01.dbf';

alter database rename file '+DATA/etax/datafile/sysaux.257.880644513' to '+DATA1/etax/datafile/sysaux01.dbf';

alter database rename file '+DATA/etax/datafile/users.259.880644515' to '+DATA1/etax/datafile/user01.dbf';

alter database rename file '+DATA/etax/datafile/undotbs2.264.880644613' to '+DATA1/etax/datafile/undotbs02.dbf';

5.1.5      两节点打开数据库并验证数据文件

alter database open;

     FILE# NAME

---------- ---------------------------------------------

         1 +DATA1/etax/datafile/system01.dbf

         2 +DATA1/etax/datafile/undotbs01.dbf

         3 +DATA1/etax/datafile/sysaux01.dbf

         4 +DATA1/etax/datafile/user01.dbf

         5 +DATA1/etax/datafile/undotbs02.dbf

已更换完成

5.2     迁移redolog文件

新添加重做日志组的成员到日志组中，在新磁盘位置，

SQL> alter database add logfile member '+DATA1' to group 1;

Database altered.

SQL> alter database add logfile member '+DATA1' to group 2;

Database altered.

SQL> alter database add logfile member '+DATA1' to group 3;

Database altered.

SQL> alter database add logfile member '+DATA1' to group 4;

Database altered.

每个现成新建一组日志组：

SQL> alter database add logfile thread 1 group 5 '+DATA1' size 50M;

Database altered.

SQL> alter database add logfile thread 2 group 6 '+DATA1' size 50M;

Database altered.

通过切换日志，将在旧存储上的重做日志成员删除

SQL> alter database drop logfile member '+DATA/cbnz/onlinelog/group_2.259.880881363';

SQL> alter database drop logfile member '+DATA/cbnz/onlinelog/group_3.280.880881425';

SQL> alter database drop logfile member '+DATA/cbnz/onlinelog/group_1.264.880881361';

SQL> alter database drop logfile member '+DATA/cbnz/onlinelog/group_4.281.880881427';

5.3     迁移控制文件

5.3.1      查看控制文件

SQL> select name from v$controlfile;

NAME

---------------------------------------------

+DATA/etax/controlfile/current.260.880644577

5.3.2      调整数据库的状态

srvctl stop database –d etax

SQL> conn / as sysdba

Connected to an idle instance.

SQL> startup nomount

ORACLE instance started.

Total System Global Area 1610612736 bytes

Fixed Size                  2096632 bytes

Variable Size             385876488 bytes

Database Buffers         1207959552 bytes

Redo Buffers               14680064 bytes

关闭一个实例，另一个实例启动到nomount状态

5.3.3      用现有控制文件恢复控制文件

在nomount实例节点上用rman连接实例

[oracle@ora1 ~]$ rman target / nocatalog

Recovery Manager: Release 10.2.0.5.0 - Production on Tue May 26 10:35:04 2015

Copyright (c) 1982, 2007, Oracle.  All rights reserved.

connected to target database: etax (not mounted)

using target database control file instead of recovery catalog

RMAN> restore controlfile to '+DATA1/etax/control01.ctl' from '+DATA/etax/controlfile/current.260.880644577';

Starting restore at 26-MAY-15

allocated channel: ORA_DISK_1

channel ORA_DISK_1: sid=149 instance=etax1 devtype=DISK

channel ORA_DISK_1: copied control file copy

Finished restore at 26-MAY-15

5.3.4      更改控制文件参数

在nomount状态的数据库中更改control_files参数

SQL> alter system set control_files='+DATA1/etax/control01.ctl' scope=spfile;

System altered.

关闭数据库，注意这里必须要先关闭数据库，然后才能启动数据库

不然会报如下的错误

ORA-01122: database file 1 failed verification check

ORA-01110: data file 1: '+DATA1/etax/datafile/system01.dbf'

ORA-01207: file is more recent than control file - old control file

正常启动数据库

5.4     迁移spfile文件

5.4.1      查看两个节点的spfile文件

oracle用户执行：

[oracle@ora2 ~]$ srvctl config database -d etax -a

ora1 etax1 /app/oracle/product/10.2.0/db

ora2 etax2 /app/oracle/product/10.2.0/db

DB_UNIQUE_NAME: etax

DB_NAME: etax

ORACLE_HOME: /app/oracle/product/10.2.0/db

SPFILE: +DATA/etax/spfileetax.ora

DOMAIN: null

DB_ROLE: null

START_OPTIONS: null

POLICY:  AUTOMATIC

ENABLE FLAG: DB ENABLED

5.4.2      用现有spfile生成pfile

create pfile='/home/oracle/pfilebackup/etax.pfile.20150526.00' from spfile;

5.4.3      用pfile文件启动数据库

在一个节点上用pfile文件将数据库启动到mount模式

startup pfile='/home/oracle/pfilebackup/etax.pfile.20150526.00' mount

5.4.4      在新位置创建pfile文件

SQL> create spfile='+DATA1/etax/spfileetax.ora' from pfile;

File created.

5.4.5      关闭数据库

SQL> shutdown immediate

5.4.6      设置数据库的参数文件位置

srvctl modify database -d etax -p '+DATA1/etax/spfileetax.ora'

检验两个节点$ORACLE_HOME/dbs下的initetax1.ora和initetax2.ora文件中的spfile参数指向位置

[oracle@ora1 ~]$ more /app/oracle/product/10.2.0/db/dbs/initetax1.ora

SPFILE='+DATA/etax/spfileetax.ora'

[oracle@ora2 ~]$ more /app/oracle/product/10.2.0/db/dbs/initetax2.ora

SPFILE='+DATA/etax/spfileetax.ora'

发现均没有变过来

查看数据库的配置：

[oracle@ora2 ~]$ srvctl config database -d etax -a

ora1 etax1 /app/oracle/product/10.2.0/db

ora2 etax2 /app/oracle/product/10.2.0/db

DB_UNIQUE_NAME: etax

DB_NAME: etax

ORACLE_HOME: /app/oracle/product/10.2.0/db

SPFILE: +DATA1/etax/spfileetax.ora

DOMAIN: null

DB_ROLE: null

START_OPTIONS: null

POLICY:  AUTOMATIC

ENABLE FLAG: DB ENABLED

已经改变

重新启动下crs，重启之后也没有改变

手动编辑两个pfile参数文件

[oracle@ora1 dbs]$ vi /app/oracle/product/10.2.0/db/dbs/initetax1.ora

SPFILE='+DATA1/etax/spfileetax.ora'

[oracle@ora2 dbs]$ vi /app/oracle/product/10.2.0/db/dbs/initetax2.ora

SPFILE='+DATA1/etax/spfileetax.ora'

然后再启动，仍然是不行的，参数文件仍然指向原位置.

来自官方的解决方法：

1.       是创建spfile时要制定pfile的路径到你刚才启动用的pfile路径：

比如我们生成的pfile文件在/tmp目录下，并且用该文件启动的数据库，则创建spfile的语句应该用如下的方式。

create spfile='+DATA/ora10g/spfileora10g.ora1' from pfile='/tmp/pfile.ora'

2.创建完成后要马上shutdown immediate

5.4.7      查看数据库的参数文件位置

srvctl config database -d 数据库名称 –a

DB_UNIQUE_NAME: etax

DB_NAME: etax

ORACLE_HOME: /app/oracle/product/10.2.0/db

SPFILE: +DATA1/etax/spfileetax.ora

DOMAIN: null

DB_ROLE: null

START_OPTIONS: null

POLICY:  AUTOMATIC

ENABLE FLAG: DB ENABLED

5.4.8      完整的创建过程

SQL> startup nomount

ORACLE instance started.

Total System Global Area 1610612736 bytes

Fixed Size                  2096632 bytes

Variable Size             385876488 bytes

Database Buffers         1207959552 bytes

Redo Buffers               14680064 bytes

SQL> create pfile='/home/oracle/pfilebackup/cbnz.spfile.backup.20150528' from spfile;

File created.

SQL> shutdown immediate

SQL> startup pfile='/home/oracle/pfilebackup/cbnz.spfile.backup.20150528' nomount

ORACLE instance started.

Total System Global Area 1610612736 bytes

Fixed Size                  2096632 bytes

Variable Size             385876488 bytes

Database Buffers         1207959552 bytes

Redo Buffers               14680064 bytes

SQL> create spfile='+DATA1/cbnz/spfilecbnz.ora' from pfile='/home/oracle/pfilebackup/cbnz.spfile.backup.20150528';

File created.

SQL> shutdown immediate

[oracle@ora1 dbs]$ srvctl config database -d cbnz -a

ora1 cbnz1 /app/oracle/product/10.2.0/db

ora2 cbnz2 /app/oracle/product/10.2.0/db

DB_UNIQUE_NAME: cbnz

DB_NAME: cbnz

ORACLE_HOME: /app/oracle/product/10.2.0/db

SPFILE: +DATA/cbnz/spfilecbnz.ora

DOMAIN: null

DB_ROLE: null

START_OPTIONS: null

POLICY:  AUTOMATIC

ENABLE FLAG: DB ENABLED

[oracle@ora1 dbs]$ srvctl modify database -d cbnz -p '+DATA1/cbnz/spfilecbnz.ora'

[oracle@ora1 dbs]$ srvctl config database -d cbnz -a

ora1 cbnz1 /app/oracle/product/10.2.0/db

ora2 cbnz2 /app/oracle/product/10.2.0/db

DB_UNIQUE_NAME: cbnz

DB_NAME: cbnz

ORACLE_HOME: /app/oracle/product/10.2.0/db

SPFILE: +DATA1/cbnz/spfilecbnz.ora

DOMAIN: null

DB_ROLE: null

START_OPTIONS: null

POLICY:  AUTOMATIC

ENABLE FLAG: DB ENABLED

[oracle@ora1 dbs]$ more initcbnz1.ora

SPFILE='+DATA/cbnz/spfilecbnz.ora'

[oracle@ora1 dbs]$ vi initcbnz1.ora

SPFILE='+DATA1/cbnz/spfilecbnz.ora'

[oracle@ora2 dbs]$ more initcbnz2.ora

SPFILE='+DATA/cbnz/spfilecbnz.ora'

[oracle@ora2 dbs]$ vi initcbnz2.ora

SPFILE='+DATA1/cbnz/spfilecbnz.ora'

SQL> conn / as sysdba

Connected to an idle instance.

SQL> startup

ORACLE instance started.

Total System Global Area 1610612736 bytes

Fixed Size                  2096632 bytes

Variable Size             385876488 bytes

Database Buffers         1207959552 bytes

Redo Buffers               14680064 bytes

Database mounted.

Database opened.

SQL> show parameter spfile

NAME                                 TYPE        VALUE

------------------------------------ ----------- ------------------------------

spfile                               string      +DATA1/cbnz/spfilecbnz.ora

5.5     临时表空间文件的处理

在新位置创建新的临时表空间作为数据库的默认临时表空间

SQL> create temporary tablespace temp01 tempfile '+DATA1' size 200M;

Tablespace created.

SQL> alter database default temporary tablespace temp01;

Database altered.

删除在旧存储上的临时表空间

SQL> alter database default temporary tablespace temp;

Database altered.

查看所有临时文件的存放地

SQL> select name from v$tempfile;

NAME

--------------------------------------------------------------------------------

+DATA1/cbnz/tempfile/temp.284.880989223

6         旧存储的ASM磁盘组处理

6.1     在两个节点卸载旧存储的磁盘组

SQL> alter diskgroup data dismount;

Diskgroup altered.

SQL> select name,state from gv$asm_diskgroup;

NAME       STATE

---------- ----------------------

ARC        MOUNTED

DATA1      MOUNTED

DATA       DISMOUNTED

DATA1      MOUNTED

ARC        MOUNTED

DATA       DISMOUNTED

rows selected.

6.2     关闭该卷组的自启动

vi /app/oracle/product/10.2.0/db/dbs/init+ASM1.ora

asm_diskgroups='ARC','DATA1'

vi /app/oracle/product/10.2.0/db/dbs/init+ASM2.ora

asm_diskgroups='ARC','DATA1'

6.3     重新启动集群和资源组

将集群和资源组全部重新启动，检查数据库状态

SQL> select name,state from gv$asm_diskgroup;

NAME                 STATE

-------------------- ----------------------

ARC                  MOUNTED

DATA1                MOUNTED

DATA                 DISMOUNTED

ARC                  MOUNTED

DATA1                MOUNTED

DATA                 DISMOUNTED

查看到该磁盘组没有自动启动，数据库的状态完全正常，说明存储迁移成功

[oracle@ora1 ~]$ crs_stat -t

Name           Type           Target    State     Host       

------------------------------------------------------------

ora....z1.inst application    ONLINE    ONLINE    ora1       

ora....z2.inst application    ONLINE    ONLINE    ora2       

ora.cbnz.db    application    ONLINE    ONLINE    ora2       

ora.etax.db    application    ONLINE    ONLINE    ora2       

ora....x1.inst application    ONLINE    ONLINE    ora1       

ora....x2.inst application    ONLINE    ONLINE    ora2       

ora....SM1.asm application    ONLINE    ONLINE    ora1       

ora....A1.lsnr application    ONLINE    ONLINE    ora1       

ora.ora1.gsd   application    ONLINE    ONLINE    ora1       

ora.ora1.ons   application    ONLINE    ONLINE    ora1       

ora.ora1.vip   application    ONLINE    ONLINE    ora1       

ora....SM2.asm application    ONLINE    ONLINE    ora2       

ora....A2.lsnr application    ONLINE    ONLINE    ora2       

ora.ora2.gsd   application    ONLINE    ONLINE    ora2       

ora.ora2.ons   application    ONLINE    ONLINE    ora2       

ora.ora2.vip   application    ONLINE    ONLINE    ora2  

6.4     删除磁盘组

删除时只能有一个节点挂载该卷组，在挂载该卷组的节点上执行移除操作。

SQL> alter diskgroup data mount;

Diskgroup altered.

SQL> drop diskgroup data including contents;

Diskgroup dropped.

6.5     查看现有磁盘组

SQL> select name from v$asm_diskgroup;

NAME

------------------------------------------------------------

ARC

DATA1

来自 “ ITPUB博客 ” ，链接：https://blog.itpub.net/20516214/viewspace-1716360/ ，如需转载，请注明出处，否则将追究法律责任。

1
1
上一篇： redo buffer大小的估量
下一篇： 10GR2 RAC恢复为单实例数据库
请登录后发表评论 登录
全部评论
chinaboysme

注册时间：2010-10-28

  • 博文量
    10
  • 访问量
    36854

最新文章

支持我们 作者招募 用户协议 FAQ Contact Us

北京盛拓优讯信息技术有限公司. 版权所有  京ICP备16024965号-8 北京市公安局海淀分局网监中心备案编号：11010802021510 niuxiaotong@pcpop.com 17352615567 未成年人举报专区

广播电视节目制作经营许可证(京) 字第1234号 中国互联网协会会员

违法和不良信息举报电话:18101376593