分类目录归档:非常规恢复

Oracle Recovery Tools实战批量坏块修复

有客户数据库无法正常启动ORA-600 6711错误

SQL> startup mount pfile='d:/pfile.txt'
ORACLE instance started.

Total System Global Area 4294964032 bytes
Fixed Size                  9036608 bytes
Variable Size             889192448 bytes
Database Buffers         3388997632 bytes
Redo Buffers                7737344 bytes
Database mounted.
SQL> alter database open ;
alter database open 
*
ERROR at line 1:
ORA-00603: ORACLE server session terminated by fatal error
ORA-01092: ORACLE instance terminated. Disconnection forced
ORA-00600: internal error code, arguments: [6711], [4313028], [1], [4309898],
[0], [], [], [], [], [], [], []
Process ID: 22708
Session ID: 978 Serial number: 56675

alert日志报错

2022-06-26T12:34:41.855326+08:00
alter database open
2022-06-26T12:34:41.984974+08:00
Ping without log force is disabled:
  instance mounted in exclusive mode.
Endian type of dictionary set to little
Undo initialization finished serial:0 start:313418906 end:313418906 diff:0 ms (0.0 seconds)
Database Characterset is ZHS16GBK
No Resource Manager plan active
2022-06-26T12:34:43.302315+08:00
Errors in file C:\APP\XFF\diag\rdbms\orcl\ora19c\trace\ora19c_ora_22708.trc  (incident=38629):
ORA-00600: internal error code, arguments: [6711], [4313028], [1], [4309898], [0], [], [], [], [], [], [], []
Incident details in: C:\APP\XFF\diag\rdbms\orcl\ora19c\incident\incdir_38629\ora19c_ora_22708_i38629.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
2022-06-26T12:34:44.232115+08:00
*****************************************************************
An internal routine has requested a dump of selected redo.
This usually happens following a specific internal error, when
analysis of the redo logs will help Oracle Support with the
diagnosis.
It is recommended that you retain all the redo logs generated (by
all the instances) during the past 12 hours, in case additional
redo dumps are required to help with the diagnosis.
*****************************************************************
2022-06-26T12:34:44.315431+08:00
Errors in file C:\APP\XFF\diag\rdbms\orcl\ora19c\trace\ora19c_ora_22708.trc:
ORA-00600: internal error code, arguments: [6711], [4313028], [1], [4309898], [0], [], [], [], [], [], [], []
2022-06-26T12:34:44.315431+08:00
Errors in file C:\APP\XFF\diag\rdbms\orcl\ora19c\trace\ora19c_ora_22708.trc:
ORA-00600: internal error code, arguments: [6711], [4313028], [1], [4309898], [0], [], [], [], [], [], [], []
Error 600 happened during db open, shutting down database
Errors in file C:\APP\XFF\diag\rdbms\orcl\ora19c\trace\ora19c_ora_22708.trc  (incident=38630):
ORA-00603: ORACLE server session terminated by fatal error
ORA-01092: ORACLE instance terminated. Disconnection forced
ORA-00600: internal error code, arguments: [6711], [4313028], [1], [4309898], [0], [], [], [], [], [], [], []
Incident details in: C:\APP\XFF\diag\rdbms\orcl\ora19c\incident\incdir_38630\ora19c_ora_22708_i38630.trc
2022-06-26T12:34:45.266678+08:00
opiodr aborting process unknown ospid (22708) as a result of ORA-603
2022-06-26T12:34:45.274688+08:00
ORA-603 : opitsk aborting process
License high water mark = 1
USER (ospid: (prelim)): terminating the instance due to ORA error 

通过分析trace文件进行分析,确认是由于histgrm$表异常导致,通过一些特殊处理,绕过该表相关sql,open数据库,并且尝试导出数据

SQL> startup mount pfile='d:/pfile.txt';
ORACLE instance started.

Total System Global Area 4294964032 bytes
Fixed Size                  9036608 bytes
Variable Size             889192448 bytes
Database Buffers         3388997632 bytes
Redo Buffers                7737344 bytes
Database mounted.
SQL>
SQL>
SQL> alter database open;

Database altered.

使用expdp导出数据报ORA-01578错
expdp-ora-1578


通过分析是由于system有坏块导致,dbv检查文件
dbv-huikuai

通过Oracle Recovery Tools工具批量坏块修复功能修复
20220626123245
20220626123343

通过工具修复大量主要坏块被修复,还有一些内部逻辑错误(后续工具继续完善),再次尝试逻辑导出数据,无任何报错,数据比较完美恢复
20220626160209

发表在 小工具, 非常规恢复 | 标签为 , , , , | 留下评论

分享运气超级好的一次drop tablespace 数据恢复

分享一次运气超级好的恢复,本身是一个测试库,应用厂商今天准备把应用正式上线,操作流程是:先删除用户,然后删除表空间,在创建表空间导入数据正式上线,不知何种原因最终客户在测试业务中做了一些正式数据,结果是无情的被删除了,通过alert日志找到应用厂商的一些操作记录
2021年8月份创建了业务表空间

Wed Aug 18 09:49:03 2021
create tablespace xifenfei datafile 'D:\app\Administrator\oradata\xifenfei\xifenfei.dbf' size 10g
Wed Aug 18 09:52:28 2021
Completed: create tablespace xifenfei datafile 'D:\app\Administrator\oradata\xifenfei\xifenfei.dbf' size 10g

今天删除表空间

Tue Apr 12 11:15:02 2022
drop tablespace xifenfei including contents and datafiles
WARNING: Cannot delete file D:\APP\ADMINISTRATOR\ORADATA\xifenfei\xifenfei.DBF
Errors in file d:\app\administrator\diag\rdbms\xifenfei\xifenfei\trace\xifenfei_ora_4296.trc:
ORA-01265: 鏃犳硶鍒犻櫎 DATA D:\APP\ADMINISTRATOR\ORADATA\xifenfei\xifenfei.DBF
ORA-27056: 鏃犳硶鍒犻櫎鏂囦欢
OSD-04024: 无法删除文件。
O/S-Error: (OS 32) 另一个程序正在使用此文件,进程无法访问。
Completed: drop tablespace xifenfei including contents and datafiles

然后客户创建新表空间提示ORA-01119,然后人工删除掉该数据文件

Tue Apr 12 11:49:02 2022
create tablespace xifenfei datafile'D:\oracle\oradata\xifenfei\xifenfei.dbf'size 20480m
ORA-1119 signalled during: create tablespace xifenfei datafile'D:\oracle\oradata\xifenfei\xifenfei.dbf'size 20480m...
Tue Apr 12 11:49:16 2022
create tablespace xifenfei datafile'D:\oracle\oradata\xifenfei\xifenfei.dbf'size 20480m
ORA-1119 signalled during: create tablespace xifenfei datafile'D:\oracle\oradata\xifenfei\xifenfei.dbf'size 20480m...

创建新表空间成功,并增加数据文件

Tue Apr 12 12:08:43 2022
create tablespace xifenfei datafile'D:\app\Administrator\oradata\xifenfei\xifenfei.dbf'size 5120m
Tue Apr 12 12:10:25 2022
Completed: create tablespace xifenfei datafile'D:\app\Administrator\oradata\xifenfei\xifenfei.dbf'size 5120m
Tue Apr 12 12:11:19 2022
alter tablespace xifenfei add datafile'D:\app\Administrator\oradata\xifenfei\xifenfei1.dbf'size 5120m
Tue Apr 12 12:13:02 2022
Completed: alter tablespace xifenfei add datafile'D:\app\Administrator\oradata\xifenfei\xifenfei1.dbf'size 5120m
alter tablespace xifenfei add datafile'D:\app\Administrator\oradata\xifenfei\xifenfei2.dbf'size 5120m
Tue Apr 12 12:14:52 2022
Completed: alter tablespace xifenfei add datafile'D:\app\Administrator\oradata\xifenfei\xifenfei2.dbf'size 5120m

基本情况就是客户删除了一个10G的业务数据文件,然后创建了3个5G的业务数据文件,现在要恢复被以前的两个表的核心数据,需要做的就是把以前的10G的数据文件找出来,但是由于删除10G文件之后又写入了15G的数据文件(而且这里面有文件的file#和删除的文件一致),理论上无法直接做block层面扫描恢复,对于此类情况,尝试文件系统层面直接反删除恢复,不过没有任何记录,文件目录被覆盖,这条路走不通.通过block扫描,发现2个file# 5文件的起始位置(分别是block 2和block 0),而且结束位置文件大小分别是10G和5G,根据经验这两个连续的磁盘分配空间很可能就是这两个file# 5的文件
20220415221719


通过winhex把数据拷贝出来,使用工具检测
20220415210709

除损坏的block 1之外(block 0 不统计在内),其他block都正常,也就是说这个10G的被删除的数据文件,只是丢失一个文件头,业务数据全部再,后续通过dul恢复客户需要数据,完成这次数据恢复,类似这种文件丢失,文件系统损坏,文件大小为0kb等类似恢复,参见以前类似blog:
win文件系统损坏oracle恢复
dbca删除库和rm删库恢复
文件系统重新分区oracle恢复
restore database误操作恢复
文件系统损坏导致数据文件异常恢复
Oracle 数据文件大小为0kb或者文件丢失恢复
rm -rf 删除数据文件恢复方法—文件系统反删除+oracle碎片重组

发表在 非常规恢复 | 标签为 , , , | 评论关闭

asm disk 磁盘部分被清空恢复

由于未知原因导致磁盘组的一个磁盘前面1G被置空(data磁盘组一共5个1T磁盘,第一个盘未知原因置空了1G数据)

[root@oradb1 ~]#kfed read /dev/mapper/asm-disk1
kfbh.endian:                          0 ; 0x000: 0x00
kfbh.hard:                            0 ; 0x001: 0x00
kfbh.type:                            0 ; 0x002: KFBTYP_INVALID
kfbh.datfmt:                          0 ; 0x003: 0x00
kfbh.block.blk:                       0 ; 0x004: blk=0
kfbh.block.obj:                       0 ; 0x008: file=0
kfbh.check:                           0 ; 0x00c: 0x00000000
kfbh.fcn.base:                        0 ; 0x010: 0x00000000
kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000
kfbh.spare1:                          0 ; 0x018: 0x00000000
kfbh.spare2:                          0 ; 0x01c: 0x00000000
7F539088C400 00000000 00000000 00000000 00000000  [................]
  Repeat 255 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][0]
…………
[root@oradb1 ~]#kfed read /dev/mapper/asm-disk1 aun=999
kfbh.endian:                          0 ; 0x000: 0x00
kfbh.hard:                            0 ; 0x001: 0x00
kfbh.type:                            0 ; 0x002: KFBTYP_INVALID
kfbh.datfmt:                          0 ; 0x003: 0x00
kfbh.block.blk:                       0 ; 0x004: blk=0
kfbh.block.obj:                       0 ; 0x008: file=0
kfbh.check:                           0 ; 0x00c: 0x00000000
kfbh.fcn.base:                        0 ; 0x010: 0x00000000
kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000
kfbh.spare1:                          0 ; 0x018: 0x00000000
kfbh.spare2:                          0 ; 0x01c: 0x00000000
7F8FEB795400 00000000 00000000 00000000 00000000  [................]
  Repeat 255 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][0]

通过分析disk 1和disk 2的前面1001MB数据,以及asm的数据分布特性,可以确认丢失的1000MB数据为39号文件的数据(和客户当时这个库是通过rman还原过来的操作有关系)
20220330224604
20220330225211


对于这样的情况,通过底层扫描,可以很好的恢复数据,参考以前类似文章
asm disk被加入vg恢复
asm disk header 彻底损坏恢复
asm磁盘dd破坏恢复
通过底层恢复,恢复结果如下
20220331133634

dbv检查数据文件
20220331134249

然后把数据库恢复出来数据文件中的数据恢复到新库中,至此完成该case恢复

发表在 非常规恢复 | 标签为 , , , | 评论关闭