安大
分类: IT业界
2013-10-15 15:37:42
原文地址:一次核心盘阵故障数据恢复过程 作者:lzp_0004
vcs双机
vxvm卷管理、vxfs文件系统
两个数据库
一库是NBU带库备份2周一全备每天一増备
总数据量约为4T
新借一相同容量的阵列做完恢复点
将新阵列连接到光交,使得两台服务器能够识别到新阵列,将阵列按照原规划划卷映射到主机组。
将新阵列接入光交使得主机能够识别到新阵列
参考原来规划划分lun
增加一个4T左右空间的磁盘做为备份盘
新建存储池bims_10_20 存储配置文件选择第一步创建存储池
创建volume,并制定名称和volume大小
创建一个新的虚拟磁盘
指定虚拟磁盘中的所属磁盘
此时可将volume映射到主机,也可在volume创建完毕后映射。选择Map Volume:
将预处理修改为归档模式进行rman备份到指定磁盘空间
中心库备份数据库文件(不包括数据文件)
拷贝数据库软件:
其中/orahome和/archivelog跨了坏卷盘bims_raid1_20new,需要备份这两个目录的数据。
cp /export/home/jyc/oraclesoft /M8000/* 到新阵列备份磁盘
其中/orahome10g跨了坏卷盘bims_raid1_20new,需要备份该目录下的数据。
通过停库拷贝方式恢复:
将预处理库通过拷贝的方式恢复到新盘阵。通过后台运行。
cp –rp /orahome10g/oracle/oradata/* /orahome10gnew/oracle/oradata/
将新阵列划出的目录卷挂载到M8000相应目录,通过nbu恢复中心库
通过拷贝方式(cp -rp)恢复预处理库
tar备份拷贝数据库软件(除了数据文件)。
注意tnsnames.ora,vcsspfile.ora,listener.ora
$ /usr/openv/netbackup/bin/bplist -C bimsdNBU -t 4 -R /|more
/cntrl_117639_1_826613979
/bk_117635_1_826521020
/cntrl_117632_1_826454206
/al_117631_1_826453892
/al_117630_1_826453866
用oracle用户登录:
$ rman target /
恢复管理器: 版本9.2.0.6.0 - 64bit Production
Copyright (c) 1995, 2002, Oracle Corporation. All rights reserved.
连接到目标数据库: BIMS (DBID=1957152904)
然后执行下面操作
set DBID=1957152904;
RUN {
ALLOCATE CHANNEL ch00 TYPE 'SBT_TAPE';
ALLOCATE CHANNEL ch01 TYPE 'SBT_TAPE';
SEND 'NB_ORA_SERV=bims-nbu, NB_ORA_CLIENT=bimsdNBU';
restore controlfile from autobackup;
#restore controlfile from '/cntrl_117639_1_826613979';
RELEASE CHANNEL ch00;
RELEASE CHANNEL ch01;
}
alter database mount;
RUN {
ALLOCATE CHANNEL ch00 TYPE 'SBT_TAPE';
ALLOCATE CHANNEL ch01 TYPE 'SBT_TAPE';
SEND 'NB_ORA_SERV=bims-nbu, NB_ORA_CLIENT=bimsdNBU';
#restore archivelog from logseq 63112 until logseq 631127 thread 1;
restore archivelog all;
RELEASE CHANNEL ch00;
RELEASE CHANNEL ch01;
}
另外从原主机/archivelog备份的归档文件拷贝到统一目录进行恢复。
RUN {
ALLOCATE CHANNEL ch00 TYPE 'SBT_TAPE';
ALLOCATE CHANNEL ch01 TYPE 'SBT_TAPE';
SEND 'NB_ORA_SERV=bims-nbu, NB_ORA_CLIENT=bimsdNBU';
RESTORE DATABASE;
RECOVER DATABASE;
RELEASE CHANNEL ch00;
RELEASE CHANNEL ch01;
}
数据库恢复完成后,启动数据库。
Alter database open;
如果原来控制文件丢失,或者日志文件有丢失,那么需要下面的命令打开数据库:
Alter database open resetlogs;
添加临时表空间文件:
ALTER TABLESPACE TEMP ADD TEMPFILE '/orahome/oracle/oradata/bims/temp01.dbf' size 10240M REUSE;
ALTER TABLESPACE TEMP ADD TEMPFILE '/orahome/oracle/oradata/bims/temp02.dbf' size 10240M REUSE;
ALTER TABLESPACE TEMP ADD TEMPFILE '/orahome/oracle/oradata/bims/temp03.dbf' size 10240M REUSE;
最后检查临时表空间文件是否正常:select name from v$tempfile;
到此数据库恢复完成。
停机、拷贝恢复/orahome10g下的所有文件,过程和恢复中心库相同。
当前新阵列数据恢复完毕,可以启动中心库和预处理库测试正常则停库,
停止6540阵列上的资源组下线:
# haconf -makerw
hagrp -unfreeze beagrp -persistent
hagrp -offline beagrp -sys bimsc
hagrp -freeze beagrp -persistent
hagrp -unfreeze pre_oragrp -persistent
hagrp -offline pre_oragrp -sys bimsc
hagrp -freeze pre_oragrp –persistent
hagrp -unfreeze oragrp -persistent
hagrp -offline oragrp -sys bimsc
hagrp -freeze oragrp -persistent
在M8000上启动新阵列上的数据库:
备份main.cf文件
修改main.cf文件中磁盘路径,通过vcs启动资源组:
# haconf -makerw
hagrp -unfreeze beagrp -persistent
hagrp -online beagrp -sys bimsd
hagrp -freeze beagrp -persistent
hagrp -unfreeze pre_oragrp -persistent
hagrp -online pre_oragrp -sys bimsd
hagrp -freeze pre_oragrp –persistent
hagrp -unfreeze oragrp -persistent
hagrp -online oragrp -sys bimsd
hagrp -freeze oragrp -persistent
将6540阵列故障raid重做、划卷映射到主机组
按照原规划,重做6540raid,步骤和新阵列相同。
划卷映射,
重做raid和卷
重做raid且使用vxvm和vxfs
中心库:
bash-2.05# vxdctl enable
bash-2.05# vxdisk list
bash-2.05# vxdg list
NAME STATE ID
rootvg enabled 1211793701.16.bimsc
beadg enabled 1212085201.21.bimsb
oradg enabled 1212085219.24.bimsb
bash-2.05# vxdctl enable
将磁盘添加入VXVM
bash-2.05# /usr/lib/vxvm/bin/vxdisksetup -i Disk_24 format=sliced
bash-2.05# /usr/lib/vxvm/bin/vxdisksetup -i Disk_25 format=sliced
bash-2.05# /usr/lib/vxvm/bin/vxdisksetup -i Disk_26 format=sliced
bash-2.05# /usr/lib/vxvm/bin/vxdisksetup -i Disk_27 format=sliced
bash-2.05# /usr/lib/vxvm/bin/vxdisksetup -i Disk_28 format=sliced
bash-2.05# /usr/lib/vxvm/bin/vxdisksetup -i Disk_29 format=sliced
bash-2.05# vxdisk list
DEVICE TYPE DISK GROUP STATUS
Disk_5 auto:sliced oradg02 oradg online
Disk_6 auto:sliced beadg01 beadg online
Disk_7 auto:sliced oradg04 oradg online
Disk_9 auto - - error
Disk_10 auto:sliced oradg05 oradg online
Disk_11 auto - - error
Disk_12 auto:sliced oradg01 oradg online
Disk_13 auto:sliced - - online
Disk_17 auto:sliced - - online
Disk_19 auto:sliced oradg06 oradg online
Disk_20 auto - - error
Disk_21 auto:sliced oradg08 oradg online
Disk_22 auto:sliced oradg09 oradg online
Disk_23 auto:sliced oradg07 oradg online
Disk_24 auto:sliced - - online
Disk_25 auto:sliced - - online
Disk_26 auto:sliced - - online
Disk_27 auto:sliced - - online
Disk_28 auto:sliced - - online
将磁盘加入磁盘组oradg
bash-2.05# vxdg -g oradg adddisk oradg24=Disk_24
bash-2.05# vxdg -g oradg adddisk oradg25=Disk_25
bash-2.05# vxdg -g oradg adddisk oradg26=Disk_26
bash-2.05# vxdg -g oradg adddisk oradg28=Disk_28
bash-2.05# vxdg -g oradg adddisk oradg29=Disk_29
bash-2.05# vxdisk list
DEVICE TYPE DISK GROUP STATUS
Disk_5 auto:sliced oradg02 oradg online
Disk_6 auto:sliced beadg01 beadg online
Disk_7 auto:sliced oradg04 oradg online
Disk_9 auto - - error
Disk_10 auto:sliced oradg05 oradg online
Disk_11 auto - - error
Disk_12 auto:sliced oradg01 oradg online
Disk_13 auto:sliced - - online
Disk_17 auto:sliced - - online
Disk_19 auto:sliced oradg06 oradg online
Disk_20 auto - - error
Disk_21 auto:sliced oradg08 oradg online
Disk_22 auto:sliced oradg09 oradg online
Disk_23 auto:sliced oradg07 oradg online
Disk_24 auto:sliced oradg24 oradg online
Disk_25 auto:sliced oradg25 oradg online
Disk_26 auto:sliced oradg26 oradg online
Disk_27 auto:sliced - - online
Disk_28 auto:sliced oradg28 oradg online
bash-2.05# vxprint -htv
Disk group: rootvg
V NAME RVG/VSET/CO KSTATE STATE LENGTH READPOL PREFPLEX UTYPE
PL NAME VOLUME KSTATE STATE LENGTH LAYOUT NCOL/WID MODE
SD NAME PLEX DISK DISKOFFS LENGTH [COL/]OFF DEVICE MODE
SV NAME PLEX VOLNAME NVOLLAYR LENGTH [COL/]OFF AM/NM MODE
SC NAME PLEX CACHE DISKOFFS LENGTH [COL/]OFF DEVICE MODE
DC NAME PARENTVOL LOGVOL
SP NAME SNAPVOL DCO
EX NAME ASSOC VC PERMS MODE STATE
v home - ENABLED ACTIVE 176126208 ROUND - fsgen
pl home-01 home ENABLED ACTIVE 176126208 CONCAT - RW
sd rootvg01-04 home-01 rootvg01 110470656 176126208 0 c0t0d0 ENA
pl home-02 home ENABLED ACTIVE 176126208 CONCAT - RW
sd rootvg02-04 home-02 rootvg02 102329856 176126208 0 c0t1d0 ENA
v rootvol - ENABLED ACTIVE 20494464 ROUND - root
pl rootvol-01 rootvol ENABLED ACTIVE 20494464 CONCAT - RW
sd rootvg01-02 rootvol-01 rootvg01 89976192 20494464 0 c0t0d0 ENA
pl rootvol-02 rootvol ENABLED ACTIVE 20494464 CONCAT - RW
sd rootvg02-03 rootvol-02 rootvg02 81835392 20494464 0 c0t1d0 ENA
v swapvol - ENABLED ACTIVE 65452032 ROUND - swap
pl swapvol-01 swapvol ENABLED ACTIVE 65452032 CONCAT - RW
sd rootvg01-01 swapvol-01 rootvg01 0 65452032 0 c0t0d0 ENA
pl swapvol-02 swapvol ENABLED ACTIVE 65452032 CONCAT - RW
sd rootvg02-01 swapvol-02 rootvg02 0 65452032 0 c0t1d0 ENA
v usr - ENABLED ACTIVE 16383360 ROUND - fsgen
pl usr-01 usr ENABLED ACTIVE 16383360 CONCAT - RW
sd rootvg01-03 usr-01 rootvg01 73592832 16383360 0 c0t0d0 ENA
pl usr-02 usr ENABLED ACTIVE 16383360 CONCAT - RW
sd rootvg02-02 usr-02 rootvg02 65452032 16383360 0 c0t1d0 ENA
Disk group: beadg
V NAME RVG/VSET/CO KSTATE STATE LENGTH READPOL PREFPLEX UTYPE
PL NAME VOLUME KSTATE STATE LENGTH LAYOUT NCOL/WID MODE
SD NAME PLEX DISK DISKOFFS LENGTH [COL/]OFF DEVICE MODE
SV NAME PLEX VOLNAME NVOLLAYR LENGTH [COL/]OFF AM/NM MODE
SC NAME PLEX CACHE DISKOFFS LENGTH [COL/]OFF DEVICE MODE
DC NAME PARENTVOL LOGVOL
SP NAME SNAPVOL DCO
EX NAME ASSOC VC PERMS MODE STATE
v beahomevol - ENABLED ACTIVE 104599552 ROUND - fsgen
pl uservol-01 beahomevol ENABLED ACTIVE 104599552 CONCAT - RW
sd beadg01-01 uservol-01 beadg01 126975 104599552 0 Disk_6 ENA
Disk group: oradg
V NAME RVG/VSET/CO KSTATE STATE LENGTH READPOL PREFPLEX UTYPE
PL NAME VOLUME KSTATE STATE LENGTH LAYOUT NCOL/WID MODE
SD NAME PLEX DISK DISKOFFS LENGTH [COL/]OFF DEVICE MODE
SV NAME PLEX VOLNAME NVOLLAYR LENGTH [COL/]OFF AM/NM MODE
SC NAME PLEX CACHE DISKOFFS LENGTH [COL/]OFF DEVICE MODE
DC NAME PARENTVOL LOGVOL
SP NAME SNAPVOL DCO
EX NAME ASSOC VC PERMS MODE STATE
v archivevol - DISABLED ACTIVE 1675624448 ROUND - fsgen
pl uservol3-01 archivevol DISABLED NODEVICE 1675624448 CONCAT - RW
sd oradg03-01 uservol3-01 oradg03 425984 838303744 0 - NDEV
sd oradg07-01 uservol3-01 oradg07 0 837320704 838303744 Disk_23 ENA
v orabakvol - ENABLED ACTIVE 1463812096 ROUND - fsgen
pl uservol4-01 orabakvol ENABLED ACTIVE 1465319424 CONCAT - RW
sd oradg04-01 uservol4-01 oradg04 393216 1257701376 0 Disk_7 ENA
sd oradg06-01 uservol4-01 oradg06 0 207618048 1257701376 Disk_19 ENA
v orahomevol - ENABLED ACTIVE 7529233711 ROUND - fsgen
pl uservol2-01 orahomevol ENABLED ACTIVE 7529234432 CONCAT - RW
sd oradg01-01 uservol2-01 oradg01 131071 1886257152 0 Disk_12 ENA
sd oradg05-01 uservol2-01 oradg05 0 1868496896 1886257152 Disk_10 ENA
sd oradg08-01 uservol2-01 oradg08 0 1887240192 3754754048 Disk_21 ENA
sd oradg09-01 uservol2-01 oradg09 0 1887240192 5641994240 Disk_22 ENA
v oraredovol - ENABLED ACTIVE 6807552 ROUND - fsgen
pl uservol-01 oraredovol ENABLED ACTIVE 6807552 CONCAT - RW
sd oradg02-01 uservol-01 oradg02 450560 6807552 0 Disk_5 ENA
bash-2.05# vxvol -g oradg stop orahomevol
移除卷
bash-2.05# vxassist -g oradg remove volume archivevol
bash-2.05# vxassist -g oradg remove volume orahomevol
bash-2.05# vxprint -htv
bash-2.05# vxassist -b -g oradg make orahomevol 3599g layout=stripe oradg25 oradg26 oradg28 oradg29
bash-2.05# vxprint -htv
v orahomevol - ENABLED ACTIVE 7547650048 SELECT orahomevol-01 fsgen
pl orahomevol-01 orahomevol ENABLED ACTIVE 7547650048 STRIPE 4/128 RW
sd oradg25-01 orahomevol-01 oradg25 0 1886912512 0/0 Disk_25 ENA
sd oradg26-01 orahomevol-01 oradg26 0 1886912512 1/0 Disk_26 ENA
sd oradg28-01 orahomevol-01 oradg28 0 1886912512 2/0 Disk_28 ENA
sd oradg29-01 orahomevol-01 oradg29 0 1886912512 3/0 Disk_29 ENA
bash-2.05# vxassist -b -g oradg make archivevol 799g oradg24
bash-2.05# vxprint -htv
v archivevol - ENABLED ACTIVE 1675624448 SELECT - fsgen
pl archivevol-01 archivevol ENABLED ACTIVE 1675624448 CONCAT - RW
sd oradg24-01 archivevol-01 oradg24 0 1675624448 0 Disk_24 ENA
bash-2.05# mkfs -F vxfs /dev/vx/rdsk/oradg/archivevol
version 7 layout
1675624448 sectors, 837812224 blocks of size 1024, log size 262144 blocks
largefiles supported
bash-2.05# mkfs -F vxfs /dev/vx/rdsk/oradg/orahomevol
version 7 layout
7547650048 sectors, 471728128 blocks of size 8192, log size 32768 blocks
largefiles supported
bash-2.05# mount -F vxfs /dev/vx/dsk/oradg/archivevol /archivelog
bash-2.05# mount -F vxfs /dev/vx/dsk/oradg/orahomevol /orahome
bash-2.05# df -h
Filesystem size used avail capacity Mounted on
/dev/vx/dsk/bootdg/rootvol
9.6G 7.3G 2.3G 77% /
/dev/vx/dsk/bootdg/usr
7.7G 2.9G 4.7G 39% /usr
/proc 0K 0K 0K 0% /proc
mnttab 0K 0K 0K 0% /etc/mnttab
fd 0K 0K 0K 0% /dev/fd
swap 81G 88K 81G 1% /var/run
dmpfs 81G 0K 81G 0% /dev/vx/dmp
dmpfs 81G 0K 81G 0% /dev/vx/rdmp
swap 81G 688K 81G 1% /tmp
/dev/vx/dsk/bootdg/home
83G 30G 52G 37% /export/home
/dev/vx/dsk/beadg/beahomevol
50G 13G 35G 27% /beahome
/dev/vx/dsk/oradg/archivevol
799G 458M 749G 1% /archivelog
/dev/vx/dsk/oradg/orahomevol
3.5T 370M 3.5T 1% /orahome
预处理库:
bash-2.05# /usr/lib/vxvm/bin/vxdisksetup -i Disk_29 format=sliced
将磁盘加入磁盘组pre_oradg
bash-2.05# vxdg -g pre_oradg adddisk oradg24=Disk_27
bash-2.05# vxprint -htv -g pre_oradg
bash-2.05# vxvol -g pre_oradg stop orahomevol
移除卷
bash-2.05# vxassist -g pre_oradg remove volume orahomevol
bash-2.05# vxassist -b -g pre_oradg make orahomevol 899g oradg27
bash-2.05# vxprint -htv
启动磁盘组上的卷:
bash-2.05# vxvol -g pre_oradg start orahomevol 或者 vxrecover -g pre_oradg -sb
bash-2.05# vxprint -htv -g pre_oradg
V NAME RVG/VSET/CO KSTATE STATE LENGTH READPOL PREFPLEX UTYPE
PL NAME VOLUME KSTATE STATE LENGTH LAYOUT NCOL/WID MODE
SD NAME PLEX DISK DISKOFFS LENGTH [COL/]OFF DEVICE MODE
SV NAME PLEX VOLNAME NVOLLAYR LENGTH [COL/]OFF AM/NM MODE
SC NAME PLEX CACHE DISKOFFS LENGTH [COL/]OFF DEVICE MODE
DC NAME PARENTVOL LOGVOL
SP NAME SNAPVOL DCO
EX NAME ASSOC VC PERMS MODE STATE
v orabakvol - ENABLED ACTIVE 417333248 SELECT - fsgen
pl orabakvol-01 orabakvol ENABLED ACTIVE 417333248 CONCAT - RW
sd pre_oradg01-01 orabakvol-01 pre_oradg01 0 417333248 0 Disk_10 ENA
v orahomevol - ENABLED ACTIVE 1885339648 SELECT - fsgen
pl orahomevol-01 orahomevol ENABLED ACTIVE 1885339648 CONCAT - RW
sd oradg27-01 orahomevol-01 oradg27 0 1885339648 0 Disk_22 ENA
v oraredovol - ENABLED ACTIVE 7049216 ROUND - fsgen
pl uservol3-01 oraredovol ENABLED ACTIVE 7049216 CONCAT - RW
sd pre_oradg03-01 uservol3-01 pre_oradg03 163839 7049216 0 Disk_9 ENA
bash-2.05# mkfs -F vxfs /dev/vx/rdsk/pre_oradg/orahomevol
bash-2.05# mount -F vxfs /dev/vx/dsk/pre_oradg/orahomevol /orahome10g
全部完成后在两台服务器上做磁盘组的导入导出测试
bimsc:
# vxvol -g pre_oradg stopall
vxdg deport oradg
vxdg import oradg
vxdg deport pre_oradg
vxdg import pre_oradg
bimsd:
vxdg deport oradg
vxdg import oradg
vxdg deport pre_oradg
vxdg import pre_oradg
修复和重启动目标磁盘组中的卷:
# vxrecover -g oradg -m [volume ...]
# vxvol -g oradg startall
将中心库和预处理数据库重新恢复到6540阵列
首先重新备份新阵列中的中心库,预处理通过拷贝的方式恢复到原6540阵列。
如果新阵列有空间,那么就备份到磁盘。(将备份磁盘挂接到M8000,通过RMAN直接备份指定磁盘路径进行备份)
划好的6540阵列 挂接到m8000:将oracle软件等(除了数据文件)先备份到 6540.
中心库:
#!/bin/sh
su - oracle -c "rman target / <
RUN {
allocate channel c1 device type disk;
allocate channel c2 device type disk;
allocate channel c3 device type disk;
allocate channel c4 device type disk;
backup database format='/rmanbak/bims/%d_%s.bak';
sql 'alter system archive log current';
}
EOF" >> /rmanbak/bims/bak.out
预处理:
cp –rp /orahome10g/oracle/oradata/* /orahome10gnew/oracle/oradata/
如果有空间:
把备份磁盘挂接到6900,做rman数据库恢复。
tar备份拷贝数据库软件(除了数据文件)。
注意tnsnames.ora,vcsspfile.ora,listener.ora
Rman target /
List backup;
用oracle用户登录:
$ rman target /
恢复管理器: 版本9.2.0.6.0 - 64bit Production
Copyright (c) 1995, 2002, Oracle Corporation. All rights reserved.
连接到目标数据库: BIMS (DBID=1957152904)
然后执行下面操作
set DBID=1957152904;
RUN {
restore controlfile from autobackup;
#restore controlfile from '/cntrl_117639_1_826613979';
}
alter database mount;
将在新阵列运行是产生的归档文件拷贝到6540阵列相应归档位置
#!/bin/sh
su - oracle -c
"rman target / <
RUN {
allocate channel c1 device type disk;
allocate channel c2 device type disk;
allocate channel c3 device type disk;
allocate channel c4 device type disk;
RESTORE DATABASE;
RECOVER DATABASE;
}
EOF" >> /rmanbak/bimsc/restore.out写到shell脚本放到后台执行。
恢复完成后停止业务
停机恢复
cp
–rp /orahome10g/oracle/oradata/* /orahome10gnew/oracle/oradata/
当恢复到最后一个归档日志的时候,停掉新阵列的数据库和应用。
停库之前做以下动作
中心库和预处理相同处理方法:
做一次alter system archive log current;
把归档拷贝到6540阵列挂载到6900和M8000的相应的归档目录,进行恢复。
Recover database using backup controlfile until cancel;
Alter database open resetlogs;
添加临时表空间文件tempfile;
ALTER TABLESPACE TEMP ADD TEMPFILE '/orahome/oracle/oradata/bims/temp01.dbf' size 10240M REUSE;
ALTER TABLESPACE TEMP ADD TEMPFILE '/orahome/oracle/oradata/bims/temp02.dbf' size 10240M REUSE;
ALTER TABLESPACE TEMP ADD TEMPFILE '/orahome/oracle/oradata/bims/temp03.dbf' size 10240M REUSE;
最后检查临时表空间文件是否正常:select name from v$tempfile;
关掉数据库重新启动vcs
停止M8000上的资源组:
hagrp -unfreeze beagrp -persistent
hagrp -offline beagrp -sys bimsd
hagrp -freeze beagrp -persistent
hagrp -unfreeze pre_oragrp -persistent
hagrp -offline pre_oragrp -sys bimsd
hagrp -freeze pre_oragrp –persistent
hagrp -unfreeze oragrp -persistent
hagrp -offline oragrp -sys bimsd
hagrp -freeze oragrp –persistent
在6900上启动资源组beagrp、pre_oragrp:
将main.cf备份,将6900上的main.cf拷贝到8000相应位置
hagrp -unfreeze beagrp -persistent
hagrp -online beagrp -sys bimsc
hagrp -freeze beagrp -persistent
hagrp -unfreeze pre_oragrp -persistent
hagrp -online pre_oragrp -sys bimsc
hagrp -freeze pre_oragrp –persistent
在bimsd上启动资源组oragrp:
hagrp -unfreeze oragrp -persistent
hagrp -online oragrp -sys bimsd
hagrp -freeze oragrp -persistent
到此数据库恢复完成。