标签归档：obet system

客户由于误操作直接在虚拟化平台点击电源键,强制关闭了正在运行的数据库服务器虚机,导致数据库无法正常启动,检查发现ntfs文件系统损坏

---

通过obet工具检测数据文件坏块(obet dbv功能完整说明),发现核心的system文件上面有一些坏块

File #1: D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF (225281 blocks) - Started: 2026-08-21 16:40:23
File #1: rfile=1 (0x00000001)  header_block_num=225280 (0x00037000)  filesize_status:OK
file 1, block 212238: tailchk error (expected 0x0106276F, got 0x01065E92), bad block
file 1, block 213507: tailchk error (expected 0x0106276F, got 0x01065E92), bad block
file 1, block 213515: tailchk error (expected 0x01065E92, got 0x0106276F), bad block
file 1, block 213547: tailchk error (expected 0x0106276F, got 0x01065E92), bad block
file 1, block 213555: tailchk error (expected 0x01065E92, got 0x0106276F), bad block
file 1, block 221387: tailchk error (expected 0x0106276F, got 0x01065E92), bad block
file 1, block 221403: tailchk error (expected 0x01065E92, got 0x0106B19C), bad block
file 1, block 221419: tailchk error (expected 0x01065E92, got 0x0106C7E3), bad block
file 1, block 222117: tailchk error (expected 0x01065E92, got 0x0106276F), bad block
file 1, block 222133: tailchk error (expected 0x0106276F, got 0x01065E92), bad block
file 1, block 222141: tailchk error (expected 0x01065E92, got 0x0106276F), bad block
file 1, block 222173: tailchk error (expected 0x0106C7E3, got 0x01065E92), bad block
  File #1 completed: 0 all zero, 0 soft corrupted, 12 tailchk error, 0 checksum error, 0 rdba error

这个12个坏块的数量和dbv检测结果一致

DBVERIFY: Release 19.0.0.0.0 - Production on 星期五 8月 21 17:37:02 2026

Copyright (c) 1982, 2019, Oracle and/or its affiliates.  All rights reserved.


DBVERIFY - 开始验证: FILE = D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF
页 212238 流入 - 很可能是介质损坏
Corrupt block relative dba: 0x00433d0e (file 1, block 212238)
Fractured block found during dbv: 
Data in bad block:
 type: 6 format: 2 rdba: 0x00433d0e
 last change scn: 0x0000.0002.d71a6f27 seq: 0x1 flg: 0x06
 spare3: 0x0
 consistency value in tail: 0x925e0601
 check value in block header: 0x93bb
 computed block checksum: 0xfd79

…………

页 222173 流入 - 很可能是介质损坏
Corrupt block relative dba: 0x004363dd (file 1, block 222173)
Fractured block found during dbv: 
Data in bad block:
 type: 6 format: 2 rdba: 0x004363dd
 last change scn: 0x0000.0002.d708e3c7 seq: 0x1 flg: 0x06
 spare3: 0x0
 consistency value in tail: 0x925e0601
 check value in block header: 0xe089
 computed block checksum: 0x7199


DBVERIFY - 验证完成

检查的页总数: 225280
处理的页总数 (数据): 125451
失败的页总数 (数据): 0
处理的页总数 (索引): 29984
失败的页总数 (索引): 0
处理的页总数 (其他): 48214
处理的总页数 (段)  : 1
失败的总页数 (段)  : 0
空的页总数: 21619
标记为损坏的总页数: 12
流入的页总数: 12
加密的总页数        : 0
最高块 SCN            : 12199365950 (2.3609431358)

使用obet的reair block功能进行修复(obet repair block使用说明)

OBET> set file 1
filename set to: D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF (file#1)

OBET> set block 212238
block set to: 212238

OBET> d

File: D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF
Block: 212238                Offsets:     0 to    31
--------------------------------------------------------------------------------
67A1C000 06A20000 0E3D4300 276F1AD7 02000106 BB930000 02000000 D0020000 E46E1AD7

<32 bytes read>

OBET> tailchk
Check tailchk for File D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF, Block 212238:
current = 0x925E0601, required = 0x6F270601

OBET> backup
Backing up file #1, block 212238
Successfully backed up block 212238 from D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF 
   to backup_blk\SYSTEM01.DBF.212238_20260821173825.blk

OBET> set mode edit
mode set to: edit

OBET> repair
Usage: repair [subcommand]
  repair block [file N] [X]  - Repair tailchk/checksum (optional: file N, block X)
  repair blkscn [file N] [X] - Repair block SCN (optional: file N, block X)

OBET> repair block
Warning: Missing value for 'block', using global setting: 212238

Repairing block 212238 in file D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF...

Repair analysis for block 212238:
1. seq_kcbh check: 0x01 -> OK
2. Tailchk check: 0x925E0601 -> needs repair (0x6F270601)
3. Checksum check: 0xBB93 -> OK

Confirm repair operations:
File: D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF
Block: 212238
Operations needed: fix tailchk
Confirm? (Y/YES to proceed): y

Verification after repair:
1. seq_kcbh: 0x01 OK
2. Tailchk: 0x6F270601 OK
3. Checksum: 0xBB93 OK

Block 212238 repair completed successfully.

所有坏块依次进行修复,然后再次使用dbv检测

DBVERIFY: Release 19.0.0.0.0 - Production on 星期五 8月 21 17:42:02 2026

Copyright (c) 1982, 2019, Oracle and/or its affiliates.  All rights reserved.


DBVERIFY - 开始验证: FILE = D:\APP\ADMINISTRATOR\ORADATA\NHIS\SYSTEM01.DBF
Block Checking: DBA = 4407811, Block Type = KTB-managed data block
**** actual rows locked by itl 2  = 0 != # in trans. header = 1
**** actual rows locked by itl 3  = 0 != # in trans. header = 1
---- end index block validation
页 213507 失败, 校验代码为 6401
Block Checking: DBA = 4407819, Block Type = KTB-managed data block
**** row 120: key out of order
**** actual rows locked by itl 2  = 0 != # in trans. header = 1
**** actual rows marked deleted = 1 != kdxlende = 0
---- end index block validation
页 213515 失败, 校验代码为 6401

DBVERIFY - 验证完成

检查的页总数: 225280
处理的页总数 (数据): 125451
失败的页总数 (数据): 0
处理的页总数 (索引): 29996
失败的页总数 (索引): 2
处理的页总数 (其他): 48214
处理的总页数 (段)  : 1
失败的总页数 (段)  : 0
空的页总数: 21619
标记为损坏的总页数: 0
流入的页总数: 0
加密的总页数        : 0
最高块 SCN            : 12199365950 (2.3609431358)

有两个block有少量逻辑错误,可以通过数据库级别设置进行跳过,基本上实现了这个12个坏块的自动修复.后续就是数据库的打开过程

SQL> startup mount pfile='d:/pfile.txt'
ORACLE 例程已经启动。

Total System Global Area 1.2885E+10 bytes
Fixed Size                 16120656 bytes
Variable Size            7751073792 bytes
Database Buffers         5100273664 bytes
Redo Buffers               17432576 bytes
数据库装载完毕。
SQL> recover database;
ORA-00283: 恢复会话因错误而取消
ORA-00742: 日志读取在线程 1 序列 19828 块 11511 中检测到写入丢失情况
ORA-00312: 联机日志 2 线程 1: 'D:\APP\ADMINISTRATOR\ORADATA\NHIS\REDO02.LOG'

SQL> RECOVER DATABASE UNTIL TIME '2026-08-21:13:28:56' USING BACKUP CONTROLFILE;
ORA-00279: change 12198905983 generated at 08/21/2026 12:43:18 needed for
thread 1
ORA-00289: suggestion :
D:\APP\ADMINISTRATOR\PRODUCT\19.0.0\DBHOME_1\RDBMS\ARC0000019827_1199133733.0001
ORA-00280: change 12198905983 for thread 1 is in sequence #19827


Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
D:\APP\ADMINISTRATOR\ORADATA\NHIS\REDO01.LOG
ORA-00279: change 12199301614 generated at 08/21/2026 13:27:06 needed for
thread 1
ORA-00289: suggestion :
D:\APP\ADMINISTRATOR\PRODUCT\19.0.0\DBHOME_1\RDBMS\ARC0000019828_1199133733.0001
ORA-00280: change 12199301614 for thread 1 is in sequence #19828
ORA-00278: log file 'D:\APP\ADMINISTRATOR\ORADATA\NHIS\REDO01.LOG' no longer
needed for this recovery


Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
D:\APP\ADMINISTRATOR\ORADATA\NHIS\REDO02.LOG
ORA-00756: 鎭㈠鎿嶄綔妫€娴嬪埌鏁版嵁鍧楀啓鍏ヤ涪澶?ORA-10567:
Redo is inconsistent with data block (file# 4, block# 31663, file offset is 259383296 bytes)
ORA-10564: tablespace UNDOTBS1
ORA-01110: 鏁版嵁鏂囦欢 4: 'D:\APP\ADMINISTRATOR\ORADATA\NHIS\UNDOTBS01.DBF'
ORA-10560: block type 'KTU UNDO BLOCK'

ORA-01112: media recovery not started

SQL> alter database open resetlogs;

Database altered.

使用expdp导出数据,完成本次恢复任务

---