标签归档：oracle异常恢复

Oracle异常恢复前备份保护现场建议—FileSystem环境

发表于 2015 年 5 月 11 日 由 惜分飞

无论是在各种会议上,还是在朋友/网友私下请教Oracle数据库恢复的问题之时,我都强调,如果你没有十足的把握,请你对您的现场进行备份,确保别对现场进行二次损坏。你不能恢复数据库,但绝对不能再次破坏数据库,给二次恢复增加难度.这里对恢复前备份提供一些指导思想和简单脚本,希望对大家有帮助.

哪些文件需要备份
熟悉数据库恢复的朋友可能都情况,Oracle在异常恢复的过程中主要修改的是system表空间里面数据,其他数据文件,redo数据,控制文件(当然由于redo,undo导致其他数据文件内部的block也可能发生改变)。在备份时间,备份空间允许的情况下,是对这些文件全部备份为好

完整备份文件

set lines 150
set pages 10000
select name from v$datafile
union all
select name from v$controlfile
union all
select member from v$logfile;

有些情况下:比如如果全部备份时间过长,备份空间不足等情况下,我们该如何备份,尽量减少因为异常恢复导致对原环境的损坏.备份最核心的system表空间,数据文件头,redo file,control file等数据,由于这个不是简单的拷贝操作,因此在生成备份语句同时,也生成还原语句,切不可生成了备份语句后,无恢复语句,导致后面还原故障现场难度增大.

无法全备情况下linux/unix数据库恢复前备份

set lines 150
set pages 10000
select 'dd if='||name||' of=&&back_dir/'||ts#||'_'||file#||'.dbf bs=1048576 count=10' 
from v$datafile where ts#<>0
union all
select 'dd if='||name||' of=&&back_dir/'||ts#||'_'||file#||'.dbf' from v$datafile where ts#=0
union all
select 'dd if='||name||' of=&&back_dir/control0'||rownum||'.ctl' from v$controlfile 
union all
select 'dd if='||member||' of=&&back_dir/'||thread#||'_'||a.group#||'_'||sequence#||'_'||substr(member,
instr(member,'/',-1)+1)  FROM v$log a, v$logfile b WHERE a.group# = B.GROUP#;

无法全备情况下linux/unix使用备份还原

set lines 150
set pages 1000
select 'dd of='||name||' if=&&back_dir/'||ts#||'_'||file#||'.dbf bs=1048576 count=10 conv=notrunc' 
from v$datafile where ts#<>0
union all
select 'dd if='||name||' if=&&back_dir/'||ts#||'_'||file#||'.dbf' from v$datafile where ts#=0
union all
select 'dd of='||name||' if=&&back_dir/control0'||rownum||'.ctl' from v$controlfile 
union all
select 'dd of='||member||' if=&&back_dir/'||thread#||'_'||a.group#||'_'||sequence#||'_'||substr(member,
instr(member,'/',-1)+1)    FROM v$log a, v$logfile b WHERE a.group# = B.GROUP#;

由于win路径斜杠不一样(/和\的区别),因此在无法全备情况下win备份语句

set lines 150
set pages 10000
select 'dd if='||name||' of=&&back_dir\'||ts#||'_'||file#||'.dbf bs=1048576 count=10' 
from v$datafile where ts#<>0
union all
select 'dd if='||name||' of=&&back_dir\'||ts#||'_'||file#||'.dbf' from v$datafile where ts#=0
union all
select 'dd if='||name||' of=&&back_dir\control0'||rownum||'.ctl' from v$controlfile 
union all
select 'dd if='||member||' of=&&back_dir\'||thread#||'_'||a.group#||'_'||sequence#||'_'||substr(member,
instr(member,'\',-1)+1)   FROM v$log a, v$logfile b WHERE a.group# = B.GROUP#;

在无法全备情况下win还原语句

set lines 150
set pages 1000
select 'dd of='||name||' if=&&back_dir\'||ts#||'_'||file#||'.dbf bs=1048576 count=10 conv=notrunc' 
from v$datafile where ts#<>0
union all
select 'dd if='||name||' if=&&back_dir\'||ts#||'_'||file#||'.dbf' from v$datafile where ts#=0
union all
select 'dd of='||name||' if=&&back_dir\control0'||rownum||'.ctl' from v$controlfile 
union all
select 'dd of='||member||' if=&&back_dir\'||thread#||'_'||a.group#||'_'||sequence#||'_'||substr(member,
instr(member,'\',-1)+1)    FROM v$log a, v$logfile b WHERE a.group# = B.GROUP#;

这里提供win环境下dd命令程序win环境dd命令工具

备注:对于asm情况异常情况恢复,备份情况请不要参考该文章,具体请见后续文章,具体见Oracle异常恢复前备份保护现场建议—ASM环境

发表在 Oracle备份恢复 | 标签为 , , | 评论关闭

Oracle 异常恢复案例汇总

发表于 2014 年 11 月 10 日 由 惜分飞

在2014年11月11日来临之际,我整理Blog中和异常恢复案例相关的部分文章,供大家参考:
dul处理分区表
误删除dual表恢复
dul恢复drop表测试
跳过obj$坏块方法
bbed解决ORA-01190
exp dmp文件损坏恢复
当前联机日志损坏恢复
ORA-01578坏块解决(1)
ORA-01578坏块解决(2)
undo异常处理步骤（9i）
DUL挖ORACLE 8.0数据库
undo异常处理步骤（10g）
ORA-600 2663 故障恢复
dul恢复truncate表测试
bbed处理ORA-01200故障
dul 10支持oracle 11g r2
使用 dul 挖数据文件初试
sysaux数据文件异常恢复
ORA-01244/ORA-01110解决
恢复被rm意外删除数据文件
ORA-00600[4194]故障解决
ORA-01207/ORA-00338恢复
DUL10直接支持ORACLE 8.0
bbed修改undo$(回滚段)状态
记录8.0.5数据库恢复过程
ORA-600[4194]/[4193]解决
使用rman找回被误删除表空间
记录一次系统回滚段坏块恢复
使用bbed解决ORA-00600[2662]
ORA-00600[kcfrbd_3]故障解决
数据库启动ORA-08103故障恢复
asm disk header 彻底损坏恢复
数据库恢复遭遇ORA-00600[3705]
Oracle 11g丢失access$恢复方法
undo segment header坏块异常恢复
ORA-600 kghstack_free2异常恢复
ORA-00600[kccpb_sanity_check_2]
重建控制文件引发ORA-00218故障
dul支持ORACLE 12C CDB数据库恢复
ORACLE 8.0.5 ORA-01207故障恢复
dul 10 export_mode=true功能增强
完美解决dul处理clob字段乱码问题
手工修复ASM DISK HEADER 异常
undo坏块导致数据库异常终止案
bbed 恢复 GLOBAL_NAME 为空故障
通过bbed解决ORA-00600[4000]案例
重建控制文件丢失数据文件导致悲剧
异常断电导致current redo损坏处理
创建控制文件遭遇ORA-600 kccscf_1
使用bbed修复损坏datafile header
通过修改控制文件scn推进数据库scn
bbed打开丢失部分system数据文件库
某集团ebs数据库redo undo丢失导致悲剧
obj$坏块exp/expdp导出不能正常执行
处理fast_recovery_area无剩余空间案例
obj$坏块情况下exp导出单个表解决方案
ORA-00600 [ktbdchk1: bad dscn] 解决
记录因磁盘头被重写,抢救redo恢复经历
rac redo log file被意外覆盖数据库恢复
使用bbed让rac中的sysaux数据文件online
通过bbed修改回滚段状态解决ORA-00704故障
处理smon清理临时段导致数据库异常案例
使用DUL挖数据文件恢复非数据外对象方法
一次侥幸的OSD-04016 O/S-Error异常恢复
记录一次ORA-600 4000数据库故障恢复
一起ORA-600 3020故障恢复的大体思路
redo异常 ORA-600 kclchkblk_4 故障恢复
Oracle 12C的第一次异常恢复—文件头坏块
数据库启动报ORA-00704 ORA-39714错误解决
ORACLE 8.1.7 数据库ORA-600 4000故障恢复
数据库报ORA-00607/ORA-00600[4194]错误
创建控制文件遭遇ORA-00600[3753]故障解决
ORA-00600[kcbshlc_1]导致数据库 down 案例
ORACLE 8.1.7 数据库ORA-600 4194故障恢复
ORA-00600[kcrf_resilver_log_1]异常恢复
控制文件异常导致ORA-00600[kccsbck_first]
分享一次ORA-01113 ORA-01110故障处理过程
记录一次ORA-600 3004 恢复过程和处理思路
Oracle安全警示录:加错裸设备导致redo异常
ORA-600 kcratr_nab_less_than_odr故障解决
双机mount数据库出现ORA-00600[kccsbck_first]
又一起存储故障导致ORA-00333 ORA-00312恢复
通过bbed模拟ORA-00607/ORA-00600 4194 故障
记录一次ORA-00316 ORA-00312 redo异常恢复
asmlib异常报ORA-00600[kfklLibFetchNext00]
某个pdb可以在root pdb open状态下进行恢复
使用bbed解决ORA-00607/ORA-00600[4194]故障
乱用_allow_resetlogs_corruption参数导致悲剧
遭遇ORA-07445[kkdliac()+346]使用odu抢救数据
ORA-27069: skgfdisp: 尝试在文件范围外执行 I/O
创建控制文件出现ORA-01565 ORA-27041 OSD-04002
记录一次system表空间坏块(ORA-01578)数据库恢复
模拟基表事务未提交数据库crash,undo丢失恢复异常恢复
重建控制文件丢失undo异常恢复—ORA-01173模拟与恢复
修改props$.NLS_CHARACTERSET导致ORA-00900异常恢复
ORA-600[2037]与ORA-07445[kcbs_dump_adv_state]错误
误drop tablespace后使用flashback database闪回异常处理
数据库恢复历史再次刷新到Oracle 7.3.2版本—redo异常恢复
ORA-27086: skgfglk: unable to lock file – already in use
ORACLE 12C ORA-07445[ktuHistRecUsegCrtMain()+1173]恢复
ORA-00600[kcratr1_lostwrt]/ORA-00600[3020]错误恢复
表空间online出现ORA-00600[kcbz_check_objd_typ]处理过程
_allow_resetlogs_corruption和adjust_scn解决ORA-01190
spfile被覆盖导致ORA-600[kmgs_parameter_update_timeout_1]
重建控制文件丢失undo异常恢复—ORA-600 25025模拟与恢复
使用_allow_resetlogs_corruption打开无归档日志rman备份库
使用_allow_resetlogs_corruption导致ORA-00704/ORA-01555故障
记录一次ORA-600 kccpb_sanity_check_2和ORA-600 kcbgtcr_13 错误恢复
ORA-00600[17182],ORA-00600[25027],ORA-00600[kghfrempty:ds]故障处理
因RAC的undo_management参数不一致导致数据库mount报ORA-01105 ORA-01606
使用bbed解决ORA-01178 file N created before last CREATE CONTROLFILE, cannot recreate
通过多次resetlogs规避类似ORA-01248: file N was created in the future of incomplete recovery错误
bootstrap$核心index(I_OBJ1,I_USER1,I_FILE#_BLOCK#,I_IND1,I_TS#,I_CDEF1等)异常恢复—ORA-00701错误解决
记录一次ORA-00600[kdxlin:psno out of range]/ORA-00600[3020]/ORA-00600[4000]/ORA-00600[4193]的数据库恢复

当你的数据库因为异常断电，强制关机，硬盘故障，drop表，truncate表，delete表，dmp文件异常，asm无法正常mount等故障无法解决导致数据丢失，且无法自行解决，请联系我们，提供专业ORACLE数据库恢复技术支持
Phone:17813235971    Q Q:107644445    E-Mail:dba@xifenfei.com

发表在 专题索引 | 标签为 | 评论关闭

通过多次resetlogs规避类似ORA-01248: file N was created in the future of incomplete recovery错误

发表于 2014 年 10 月 14 日 由 惜分飞

数据库现状
控制文件
recover_xifenfei0
控制文件中数据文件信息
recover_xifenfei1
数据文件头信息
recover_xifenfei2
redo信息
recover_xifenfei3
根据当前数据库恢复检查脚本(Oracle Database Recovery Check)收集的信息,数据库的是非归档状态,而且redo已经覆盖,数据库datafile 5 无法直接online.遇到这样情况,可以使用bbed修改文件头scn实现online(使用bbed让rac中的sysaux数据文件online),也可以通过使用_allow_resetlogs_corruption等隐含参数实现online.本恢复案例中有180个数据文件,160个offline,然后open数据库,所以大量数据文件无法正常online,bbed工作量太大.在恢复过程中不幸遇到ORA-01248

数据库resetlogs出现ORA-01248错误

SQL> alter database open resetlogs;
alter database open resetlogs
*
ERROR at line 1:
ORA-01248: file 5 was created in the future of incomplete recovery
ORA-01110: data file 5: 'F:\TTDATA\PUBRTS.DAT'

alert日志记录

Fri Oct 10 15:09:26 2014
alter database open resetlogs
Fri Oct 10 15:09:26 2014
RESETLOGS is being done without consistancy checks. This may result
in a corrupted database. The database should be recreated.
ORA-1248 signalled during: alter database open resetlogs...
Fri Oct 10 15:15:22 2014
alter database open
Fri Oct 10 15:15:22 2014
ORA-1589 signalled during: alter database open...
Fri Oct 10 15:15:30 2014
alter database  open resetlogs
RESETLOGS is being done without consistancy checks. This may result
in a corrupted database. The database should be recreated.
ORA-1248 signalled during: alter database  open resetlogs...

尝试offline文件然后resetlogs

SQL>ALTER DATABASE DATAFILE 5  OFFLINE;

Database altered.


sql>ALTER DATABASE OPEN RESETLOGS；

ERROR at line 1:
ORA-01245: ffline file 5 will be lost if resetlogs is done
ORA-01110: data file 5: 'F:\TTDATA\PUBRTS.DAT'

alert日志

Fri Oct 10 15:19:37 2014
ALTER DATABASE DATAFILE 5 offline
Fri Oct 10 15:19:37 2014
Completed: ALTER DATABASE DATAFILE 5 offline
Fri Oct 10 15:19:40 2014
alter database open resetlogs
RESETLOGS is being done without consistancy checks. This may result
in a corrupted database. The database should be recreated.
ORA-1245 signalled during: alter database open resetlogs...

出现该错误原因是由于数据库是非归档模式,offline数据文件需要使用offline drop

Fri Oct 10 15:22:16 2014
alter database datafile 5 offline drop
Fri Oct 10 15:22:17 2014
Completed: alter database datafile 5 offline drop
Fri Oct 10 15:23:13 2014
alter database open resetlogs
Fri Oct 10 15:23:14 2014
Fri Oct 10 15:23:49 2014
RESETLOGS after complete recovery through change 1422423346
Resetting resetlogs activation ID 3503292347 (0xd0cfffbb)
Fri Oct 10 15:24:01 2014
Setting recovery target incarnation to 3
Fri Oct 10 15:24:04 2014
Assigning activation ID 3649065262 (0xd980512e)
LGWR: STARTING ARCH PROCESSES
ARC0 started with pid=23, OS id=3772
Fri Oct 10 15:24:04 2014
ARC0: Archival started
ARC1: Archival started
LGWR: STARTING ARCH PROCESSES COMPLETE
ARC1 started with pid=24, OS id=3668
Fri Oct 10 15:24:05 2014
Thread 1 opened at log sequence 1
  Current log# 1 seq# 1 mem# 0: D:\ORACLE\PRODUCT\10.2.0\ORADATA\CLTTDB\REDO01.LOG
Successful open of redo thread 1
Fri Oct 10 15:24:05 2014
MTTR advisory is disabled because FAST_START_MTTR_TARGET is not set
Fri Oct 10 15:24:05 2014
ARC0: STARTING ARCH PROCESSES
ARC2: Archival started
ARC0: STARTING ARCH PROCESSES COMPLETE
ARC0: Becoming the 'no FAL' ARCH
ARC2 started with pid=25, OS id=636
Fri Oct 10 15:24:06 2014
ARC0: Becoming the 'no SRL' ARCH
Fri Oct 10 15:24:06 2014
ARC1: Becoming the heartbeat ARCH
Fri Oct 10 15:24:06 2014
SMON: enabling cache recovery
Fri Oct 10 15:24:07 2014
Successfully onlined Undo Tablespace 1.
Dictionary check beginning
File #5 is offline, but is part of an online tablespace.
data file 5: 'F:\TTDATA\PUBRTS.DAT'
Dictionary check complete
Fri Oct 10 15:24:19 2014
SMON: enabling tx recovery
Fri Oct 10 15:24:19 2014
Database Characterset is ZHS16GBK
replication_dependency_tracking turned off (no async multimaster replication found)
Starting background process QMNC
QMNC started with pid=26, OS id=868
Fri Oct 10 15:24:21 2014
LOGSTDBY: Validating controlfile with logical metadata
Fri Oct 10 15:24:21 2014
LOGSTDBY: Validation complete
Completed: alter database open resetlogs

open成功后,再次resetlogs库,实现数据文件online

Fri Oct 10 15:28:44 2014
ALTER DATABASE DATAFILE 5 online
Fri Oct 10 15:28:44 2014
Completed: ALTER DATABASE DATAFILE 5 online
Fri Oct 10 15:31:46 2014
alter database open resetlogs
Fri Oct 10 15:31:46 2014
RESETLOGS is being done without consistancy checks. This may result
in a corrupted database. The database should be recreated.
Setting recovery target incarnation to 4
Fri Oct 10 15:32:00 2014
Assigning activation ID 3649091231 (0xd980b69f)
LGWR: STARTING ARCH PROCESSES
ARC0 started with pid=23, OS id=700
Fri Oct 10 15:32:00 2014
ARC0: Archival started
ARC1: Archival started
LGWR: STARTING ARCH PROCESSES COMPLETE
ARC1 started with pid=24, OS id=3360
Fri Oct 10 15:32:01 2014
Thread 1 opened at log sequence 1
  Current log# 1 seq# 1 mem# 0: D:\ORACLE\PRODUCT\10.2.0\ORADATA\CLTTDB\REDO01.LOG
Successful open of redo thread 1
Fri Oct 10 15:32:01 2014
MTTR advisory is disabled because FAST_START_MTTR_TARGET is not set
Fri Oct 10 15:32:01 2014
ARC0: STARTING ARCH PROCESSES
ARC2: Archival started
ARC0: STARTING ARCH PROCESSES COMPLETE
ARC0: Becoming the 'no FAL' ARCH
ARC2 started with pid=25, OS id=2016
Fri Oct 10 15:32:02 2014
ARC0: Becoming the 'no SRL' ARCH
Fri Oct 10 15:32:02 2014
ARC1: Becoming the heartbeat ARCH
Fri Oct 10 15:32:02 2014
SMON: enabling cache recovery
Fri Oct 10 15:32:03 2014
Successfully onlined Undo Tablespace 1.
Dictionary check beginning
Fri Oct 10 15:32:15 2014
Dictionary check complete
Fri Oct 10 15:32:15 2014
SMON: enabling tx recovery
Fri Oct 10 15:32:15 2014
Database Characterset is ZHS16GBK
replication_dependency_tracking turned off (no async multimaster replication found)
Starting background process QMNC
QMNC started with pid=26, OS id=256
Fri Oct 10 15:32:17 2014
LOGSTDBY: Validating controlfile with logical metadata
Fri Oct 10 15:32:17 2014
LOGSTDBY: Validation complete
Completed: alter database open resetlogs
发表在 Oracle备份恢复 | 标签为 , , | 评论关闭