ASM diskmetadata及ASM diskgroup metadata的备份及恢复 Voting Disk及OCR的备份及恢复

主要介绍ASM diskmetadata及ASM diskgroup metadata的备份及恢复，还介绍了Oracle RAC环境中比较重要的Voting Disk及OCR的备份及恢复。

 

一、备份ASM disk头文件

1、查询磁盘的路径（路径会因系统的不同而显示各异，以下是在Linux环境，如在Solaris环境会显示如’/dev/rdsk/c0t60080E5000367E98000004A25373CF5Cd0s6‘）

[grid@ractst01 ~]$ sqlplus / as sysasm

SQL>select g.group_number g_no, g.name,d.disk_number, d.mount_status, d.header_status, g.type,d.name, d.path fromv$asm_disk d,v$asm_diskgroup g where g.group_number=d.group_number;

 

     G_NO NAME       DISK_NUMBERMOUNT_S HEADER_STATU TYPE   NAME       PATH

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

        1 DATA                 0CACHED  MEMBER       EXTERN DISK1      ORCL:DISK1

        1 DATA                 1CACHED  MEMBER       EXTERN DISK2      ORCL:DISK2

        1 DATA                 2CACHED  MEMBER       EXTERN DISK3      ORCL:DISK3

2、备份ASM disk头文件（也可采用dd等其它工具，建议使用kfed）

[grid@ractst01 ~]$kfedread ORCL:DISK1 te=/home/grid/mdbackup/disk1.bak

[grid@ractst01 ~]$ more/home/grid/mdbackup/disk1.bak

kfbh.endian:                          1 ; 0x000: 0x01

kfbh.hard:                          130 ; 0x001: 0x82

kfbh.type:                            1 ; 0x002:KFBTYP_DISKHEAD

kfbh.datfmt:                          1 ; 0x003: 0x01

kfbh.block.blk:                       0 ; 0x004: T=0 NUMB=0x0

kfbh.block.obj:              2147483648 ; 0x008: TYPE=0x8NUMB=0x0

kfbh.check:                  2074855584 ; 0x00c:0x7babc8a0

kfbh.fcn.base:                        0 ; 0x010: 0x00000000

kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000

kfbh.spare1:                          0 ; 0x018: 0x00000000

kfbh.spare2:                          0 ; 0x01c: 0x00000000

kfdhdb.driver.provstr:    ORCLDISKDISK1 ; 0x000: length=13

kfdhdb.driver.reserved[0]:   1263749444 ; 0x008: 0x4b534944

kfdhdb.driver.reserved[1]:           49 ; 0x00c: 0x00000031

kfdhdb.driver.reserved[2]:            0 ; 0x010: 0x00000000

kfdhdb.driver.reserved[3]:            0 ; 0x014: 0x00000000

kfdhdb.driver.reserved[4]:            0 ; 0x018: 0x00000000

kfdhdb.driver.reserved[5]:            0 ; 0x01c: 0x00000000

kfdhdb.compat:                186646528 ; 0x020: 0x0b200000

kfdhdb.dsknum:                        0 ; 0x024: 0x0000

kfdhdb.grptyp:                        1 ; 0x026: KFDGTP_EXTERNAL

kfdhdb.hdrsts:                        3 ; 0x027:KFDHDR_MEMBER

kfdhdb.dskname:                   DISK1 ; 0x028: length=5

kfdhdb.grpname:                    DATA ; 0x048: length=4

kfdhdb.fgname:                    DISK1 ; 0x068: length=5

kfdhdb.capname:                         ; 0x088: length=0

kfdhdb.crestmp.hi:             33025220 ; 0x0a8: HOUR=0x4DAYS=0x6 MNTH=0xb YEAR=0x7df

kfdhdb.crestmp.lo:           2096627712 ; 0x0ac: USEC=0x0MSEC=0x200 SECS=0xf MINS=0x1f

kfdhdb.mntstmp.hi:             33025221 ; 0x0b0: HOUR=0x5 DAYS=0x6 MNTH=0xbYEAR=0x7df

kfdhdb.mntstmp.lo:           3831974912 ; 0x0b4: USEC=0x0MSEC=0x1d3 SECS=0x6 MINS=0x39

kfdhdb.secsize:                     512 ; 0x0b8: 0x0200

kfdhdb.blksize:                    4096 ; 0x0ba: 0x1000

kfdhdb.ausize:                  1048576 ; 0x0bc: 0x00100000

kfdhdb.mfact:                    113792 ; 0x0c0: 0x0001bc80

kfdhdb.dsksize:                    8189 ; 0x0c4: 0x00001ffd

kfdhdb.pmcnt:                         2 ; 0x0c8: 0x00000002

kfdhdb.fstlocn:                       1 ; 0x0cc: 0x00000001

kfdhdb.altlocn:                       2 ; 0x0d0: 0x00000002

kfdhdb.f1b1locn:                      2 ; 0x0d4: 0x00000002

kfdhdb.redomirrors[0]:                0 ; 0x0d8: 0x0000

kfdhdb.redomirrors[1]:                0 ; 0x0da: 0x0000

kfdhdb.redomirrors[2]:                0 ; 0x0dc: 0x0000

kfdhdb.redomirrors[3]:                0 ; 0x0de: 0x0000

kfdhdb.dbcompat:              168820736 ; 0x0e0: 0x0a100000

kfdhdb.grpstmp.hi:             33025220 ; 0x0e4: HOUR=0x4DAYS=0x6 MNTH=0xb YEAR=0x7df

kfdhdb.grpstmp.lo:           2096453632 ; 0x0e8: USEC=0x0MSEC=0x156 SECS=0xf MINS=0x1f

kfdhdb.vfstart:                     128 ; 0x0ec: 0x00000080

kfdhdb.vfend:                       160 ; 0x0f0: 0x000000a0

kfdhdb.spfile:                        0 ; 0x0f4: 0x00000000

kfdhdb.spfflg:                        0 ; 0x0f8: 0x00000000

kfdhdb.ub4spare[0]:                   0 ; 0x0fc: 0x00000000

kfdhdb.ub4spare[1]:                   0 ; 0x100: 0x00000000

kfdhdb.ub4spare[2]:                   0 ; 0x104: 0x00000000

kfdhdb.ub4spare[3]:                   0 ; 0x108: 0x00000000

kfdhdb.ub4spare[4]:                   0 ; 0x10c: 0x00000000

kfdhdb.ub4spare[5]:                   0 ; 0x110: 0x00000000

kfdhdb.ub4spare[6]:                   0 ; 0x114: 0x00000000

kfdhdb.ub4spare[7]:                   0 ; 0x118: 0x00000000

kfdhdb.ub4spare[8]:                   0 ; 0x11c: 0x00000000

kfdhdb.ub4spare[9]:                   0 ; 0x120: 0x00000000

kfdhdb.ub4spare[10]:                  0 ; 0x124: 0x00000000

kfdhdb.ub4spare[11]:                  0 ; 0x128: 0x00000000

kfdhdb.ub4spare[12]:                  0 ; 0x12c: 0x00000000

………………………………………………………………..

[grid@ractst01 ~]$ kfed read ORCL:DISK2te=/home/grid/mdbackup/disk2.bak

[grid@ractst01 ~]$ kfed read ORCL:DISK3te=/home/grid/mdbackup/disk3.bak

 

3、一般每磁盘备份一次即可，如有磁盘或磁盘组的属性调整需进行一次新的备份。

 

4、上传备份文件到统一备份归档目录。

 

二、ASM disk头文件恢复：

1、以下是模拟头文件损坏：

[grid@ractst01 ~]$dd if=/dev/zero of=ORCL:DISK2 bs=4096 count=2

 

SQL> select NAME,MOUNT_STATUS,HEADER_STATUS from v$asm_disk;

 

NAME                           MOUNT_S HEADER_STATU

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

                               CLOSED  PROVISIONED

                               CLOSED  PROVISIONED

                               CLOSED  PROVISIONED

DISK1                          CACHED  MEMBER

DISK2                          CACHED  CANDIDATE

DISK3                          CACHED  MEMBER

 

SQL> shutdown abort

ASM instance shutdown

SQL> startup nomount;

ORA-03113: end-of-file on communication channel

 

2、恢复损坏的ASM disk头文件

[grid@ractst01 ~]#kfed repair /dev/sdb1(有关这方面的相关知识将在另外文档中介绍)

[grid@ractst01 ~]#kfedwrite /dev/sdb1 te=/home/grid/mdbackup/disk2.bak

 

SQL> /

NAME                           MOUNT_S HEADER_STATU

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

                               CLOSED  PROVISIONED

                               CLOSED  PROVISIONED

                               CLOSED  PROVISIONED

DISK1                          CACHED  MEMBER

DISK2                          CACHED  MEMBER

DISK3                          CACHED  MEMBER

三、备份ASM diskgroup头文件

1、在本机建立备份目录

[grid@ractst01 ~]$ mkdir mdbackup

 

2、查询磁盘组的信息

[grid@ractst01 ~]$ asmcmd

ASMCMD>

ASMCMD> lsdg

State   Type    Rebal  Sector Block       AU  Total_MB Free_MB  Req_mir_free_MB  Usable_file_MB  Offline_disks Voting_files  Name

MOUNTED EXTERN  N         512  4096  1048576     24567   22432                0           22432              0             N DATA/ ----磁盘组名称为DATA，可能有多个磁盘组

 

3、备份ASM磁盘组头文件信息，建议每个磁盘组备份为一个文件

ASMCMD> md_backup/home/grid/mdbackup/DATA.bak -G DATA

Disk group metadata to be backed up:DATA

Current alias directory path:RACTST/ONLINELOG

Current alias directory path: RACTST

Current alias directory path: RACTST/TEMPFILE

Current alias directory path:ractst-cluster/OCRFILE

Current alias directory path:RACTST/DATAFILE

Current alias directory path:ractst-cluster/ASMPARAMETERFILE

Current alias directory path:RACTST/PARAMETERFILE

Current alias directory path: ractst-cluster

Current alias directory path:RACTST/CONTROLFILE

 

4、备份策略：一般不需要定期备份，如有磁盘组的修改比如磁盘的添加及删除、磁盘组属性的修改等，建议进行一次新的备份。

 

5、上传备份文件到统一备份归档目录。

 

四、恢复ASM diskgroup头文件

一旦ASM diskgroup metadata有问题，可以通过md_restore恢复：

ASMCMD> help md_restore

       md_restore  <backup_file>[--silent]

                   [--full|--nodg|--newdg] [-S<sql_script_file>]

                   [-G'<diskgroup_name>,<diskgroup_name>,...']

                   [-o'<old_diskgroup_name>:<new_diskgroup_name>,...']

 

       Perform ASM Metadata restore for disk groups.

       Read metadata information from <backup_file>.

       --silent Ignore errors. Normally if md_restore encounters an error, it

                 will stop. Specifying thisflag ignores that.

       --full  create disk group andrestore metadata.

       --nodg  restore metadata only.

       --newdg create disk group with a different name and restore

                   metadata; -o is required.

       -S Write SQL commands to <sql_script_file> instead of executingthem.

       -G Select the disk groups to be restored. If no disk groups defined,

           all of them will be restored.

       -o Rename disk group <old_diskgroup_name> to<new_diskgroup_name>.

 

 

五、以下内容是RAC环境中比较重要的VotingDisk及OCR的备份及恢复

 

1、Voting Disk

Voting Disk主要用于记录节点成员状态，在出现脑裂时，决定哪个成员获得控制权，其他的成员须从集群中剔除。

 

查看votedisk的位置：

[root@rac1 ~]# crsctl query cssvotedisk

0.   0   /dev/raw/raw2

located 1 votedisk(s).

 

备份votedisk盘：

[root@raw1 bin]# dd if=/dev/raw/raw2of=/home/grid/votebackup/voting_disk.bak

294912+0 records in

294912+0 records out

 

恢复votedisk盘：

[root@raw1 bin]# dd if=/home/grid/votebackup/voting_disk.bakof=/dev/raw/raw2

294912+0 records in

294912+0 records out

 

注：从11GR2开始不再需要手工备份votingdisk，votingdisk的改变会自动备份到ocr备份文件中，相关信息会自动还原到任何添加的表决磁盘文件中。

 

 

2、OCR

OCR记录节点成员的配置信息，如database，ASM，Instance，VIP等CRS资源的配置信息。OracleClusterware把集群的配置信息放在共享存储上，这些信息包括了集群节点的列表、集群数据库实例到节点的映射以及CRS应用程序资源信息。存放的位置就在OCR Disk上. 在整个集群中，只有一个节点能对OCR Disk 进行读写操作，这个节点叫作Master Node，所有节点都会在内存中保留一份OCR的拷贝，同时有一个OCR Process 从这个内存中读取内容。OCR 内容发生改变时，由Master Node的OCR Process负责同步到其他节点的OCR Process。

 

Oracle 每4个小时对其做一次备份，并且保留最后的3个备份，以及前一天，前一周的最后一个备份。这个备份由Master Node CRSD进程完成，备份的默认位置是$CRS_HOME/crs/cdata/<cluster_name>目录下，可以通过ocrconfig-backuploc <directory_name> 命令修改到新的目录。每次备份后，备份文件名自动更改，以反应备份时间顺序，最近一次的备份是backup00.ocr。

 

root@VPRAC01 # ocrconfig -showbackup

 

vprac02     2015/11/09 14:57:23    /oracle/app/11.2.0/grid/cdata/VPRAC-cluster/backup00.ocr

vprac02     2015/11/09 10:57:23    /oracle/app/11.2.0/grid/cdata/VPRAC-cluster/backup01.ocr

vprac02     2015/11/09 06:57:22    /oracle/app/11.2.0/grid/cdata/VPRAC-cluster/backup02.ocr

vprac02     2015/11/07 22:57:16    /oracle/app/11.2.0/grid/cdata/VPRAC-cluster/day.ocr

vprac02     2015/10/28 22:56:31    /oracle/app/11.2.0/grid/cdata/VPRAC-cluster/week.ocr

vprac01     2013/08/22 11:52:51    /oracle/app/11.2.0/grid/cdata/VPRAC-cluster/backup_20130822_115251.ocr

 

Oracle 推荐在对集群做调整时，比如增加，删除节点之前，修改RAC IP之前，对OCR做一个备份，可以使用export 备份到指定文件，如果做了replace或者restore 等操作，Oracle 建议使用 cluvfy comp ocr -n all 命令来做一次全面的检查。

 

对OCR的恢复，可使用ocrconfig 命令：

root@RACTSRT01 # ocrconfig

Name:

       ocrconfig - Configuration tool for Oracle Cluster/Local Registry.

 

Synopsis:

       ocrconfig [option]

       option:

                [-local] -export<filename>

                                                    - Export OCR/OLRcontents to a file

                [-local] -import<filename>         - Import OCR/OLRcontents from a file

                [-local] -upgrade [<user>[<group>]]

                                                   - Upgrade OCR from previousversion

                -downgrade [-version<version string>]

                                                   - Downgrade OCR to the specified version

                [-local] -backuploc<dirname>       - Configure OCR/OLRbackup location

                [-local] -showbackup[auto|manual]  - Show OCR/OLR backupinformation

                [-local] -manualbackup              - Perform OCR/OLR backup

                [-local] -restore<filename>        - Restore OCR/OLRfrom physical backup

                -replace <currentfilename> -replacement <new filename>

                                                   - Replace a OCR device/file <filename1> with <filename2>

                -add <filename>                     - Add a new OCRdevice/file

                -delete <filename>                  - Remove a OCR device/file

                -overwrite                          - Overwrite OCRconfiguration on disk

                -repair -add <filename> |-delete <filename> | -replace <current filename> -replacement<new filename>

                                                   - Repair OCR configuration on the local node

                -help                               - Print out thishelp information

 

Note:

       * A log file will be created in

       $ORACLE_HOME/log/<hostname>/client/ocrconfig_<pid>.log.Please ensure

       you have file creation privileges in the above directory before

       running this tool.

       * Only -local -showbackup [manual] is supported.

       * Use option '-local' to indicate that the operation is to be performedon the Oracle Local Registry.

 

OCR恢复步骤大致如下：

 

1）停止所有节点clusterware

# crsctl stop crs

# crsctl stop crs -f

2）以root用户在其中一个节点独占模式启动clusterware

# crsctl start crs -excl -nocrs

备注：如果发现crsd在运行，那么通过如下命令将之停止。

# crsctl stop resource ora.crsd -init

3）创建新的存放ocr和vote disk的磁盘组，磁盘组名和原有的一致(如果想改变位置，需修改/etc/oracle/ocr.loc文件)

备注：如发现无法创建等情况，可以采用如下删除磁盘组等排错思路

SQL> drop diskgroup disk_group_nameforce including contents;

4）还原ocr，并检查

# ocrconfig -restore file_name

# ocrcheck

5）恢复表决磁盘，并检查

# crsctl replace votedisk+asm_disk_group

# crsctl query css votedisk

6）停止独占模式运行的clusterware

# crsctl stop crs -f

7）所有节点正常启动clusterware

# crsctl start crs

8）CVU验证所有RAC节点OCR的完整性

$ cluvfy comp ocr -n all -verbose