月归档:三月 2020

又一例system大量坏块恢复

有朋友找到我们,说数据库服务可以启动,但是无法登陆,类似报错

C:\Users\XIFENFEI>D:\app\XIFENFEI\product\11.2.0.1\dbhome_2\bin\sqlplus / as sys
dba

SQL*Plus: Release 11.2.0.1.0 Production on 星期四 3月 12 15:04:32 2020

Copyright (c) 1982, 2010, Oracle.  All rights reserved.

ERROR:
ORA-01075: 您现在已登录


请输入用户名:
ERROR:
ORA-01017: 用户名/口令无效; 登录被拒绝


请输入用户名:
ERROR:
ORA-01017: 用户名/口令无效; 登录被拒绝


SP2-0157: 在 3 次尝试之后无法连接到 ORACLE, 退出 SQL*Plus

通过分析发现数据库启动报错(未正常open成功)

C:\Users\XIFENFEI>D:\app\XIFENFEI\product\11.2.0.1\dbhome_2\bin\sqlplus / as sysdba

SQL*Plus: Release 11.2.0.1.0 Production on 星期四 3月 12 14:58:28 2020

Copyright (c) 1982, 2010, Oracle.  All rights reserved.

已连接到空闲例程。

SQL> startup mount pfile='F:\pfile.txt';
ORACLE 例程已经启动。

Total System Global Area 3307048960 bytes
Fixed Size                  2180264 bytes
Variable Size            1811942232 bytes
Database Buffers         1476395008 bytes
Redo Buffers               16531456 bytes
数据库装载完毕。

SQL> alter database open;
alter database open
*
第 1 行出现错误:
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01578: ORACLE 数据块损坏 (文件号 1, 块号 48396)
ORA-01110: 数据文件 1: 'F:\ORADATA\SYSTEM01.DBF'

数据库没有正常启动成功的原因是由于system文件有坏块导致,通过dbv检查system文件发现有大量连续坏块

DBVERIFY: Release 11.2.0.1.0 - Production on 星期四 3月 12 19:12:34 2020

Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.


DBVERIFY - 开始验证: FILE = F:\ORADATA\SYSTEM01.DBF
页 48064 流入 - 很可能是介质损坏
Corrupt block relative dba: 0x0040bbc0 (file 1, block 48064)
Fractured block found during dbv: 
Data in bad block:
 type: 6 format: 2 rdba: 0x0040bbc0
 last change scn: 0x0000.0006f69c seq: 0x1 flg: 0x06
 spare1: 0x0 spare2: 0x0 spare3: 0x0
 consistency value in tail: 0x43455474
 check value in block header: 0x3893
 computed block checksum: 0xadfa

…………

页 48412 标记为损坏
Corrupt block relative dba: 0x0040bd1c (file 1, block 48412)
Bad header found during dbv: 
Data in bad block:
 type: 36 format: 2 rdba: 0x6dce856d
 last change scn: 0xfc44.d24c936f seq: 0xdc flg: 0x3b
 spare1: 0x9c spare2: 0x92 spare3: 0xcf67
 consistency value in tail: 0x43455474
 check value in block header: 0x2598
 block checksum disabled


DBVERIFY - 验证完成

检查的页总数: 249600
处理的页总数 (数据): 209467
失败的页总数 (数据): 0
处理的页总数 (索引): 13616
失败的页总数 (索引): 0
处理的页总数 (其他): 3369
处理的总页数 (段)  : 1
失败的总页数 (段)  : 0
空的页总数: 22799
标记为损坏的总页数: 349
流入的页总数: 1
加密的总页数        : 0
最高块 SCN            : 84123103 (0.84123103)

数据库alert日志信息

Thu Mar 12 14:52:20 2020
SMON: enabling cache recovery
Successfully onlined Undo Tablespace 2.
Verifying file header compatibility for 11g tablespace encryption..
Verifying 11g file header compatibility for tablespace encryption completed
SMON: enabling tx recovery
Database Characterset is ZHS16GBK
Hex dump of (file 1, block 48403) in trace file d:\app\xifenfei\diag\rdbms\qdbdc\o11201gbk\trace\o11201gbk_ora_9480.trc
Corrupt block relative dba: 0x0040bd13 (file 1, block 48403)
Bad header found during multiblock buffer read
Data in bad block:
 type: 36 format: 2 rdba: 0x6dce856d
 last change scn: 0xfc44.d24c936f seq: 0xdc flg: 0x3b
 spare1: 0x9c spare2: 0x92 spare3: 0xcf67
 consistency value in tail: 0x43455474
 check value in block header: 0x2598
 block checksum disabled
Reading datafile 'F:\ORADATA\SYSTEM01.DBF' for corruption at rdba: 0x0040bd13 (file 1, block 48403)
Reread (file 1, block 48403) found same corrupt data
Hex dump of (file 1, block 48404) in trace file d:\app\xifenfei\diag\rdbms\qdbdc\o11201gbk\trace\o11201gbk_ora_9480.trc
Corrupt block relative dba: 0x0040bd14 (file 1, block 48404)
Corrupt Block Found
Bad header found during multiblock buffer read
         TSN = 0, TSNAME = SYSTEM
Data in bad block:
         RFN = 1, BLK = 48403, RDBA = 4242707
 type: 36 format: 2 rdba: 0x6dce856d
 last change scn: 0xfc44.d24c936f seq: 0xdc flg: 0x3b
 spare1: 0x9c spare2: 0x92 spare3: 0xcf67
 consistency value in tail: 0x43455474
 check value in block header: 0x2598
 block checksum disabled
Reading datafile 'F:\ORADATA\SYSTEM01.DBF' for corruption at rdba: 0x0040bd14 (file 1, block 48404)
Reread (file 1, block 48404) found same corrupt data
Errors in file d:\app\xifenfei\diag\rdbms\qdbdc\o11201gbk\trace\o11201gbk_ora_9480.trc  (incident=3784):
ORA-01578: ORACLE 数据块损坏 (文件号 1, 块号 48404)
ORA-01110: 数据文件 1: 'F:\ORADATA\SYSTEM01.DBF'
Incident details in: d:\app\xifenfei\diag\rdbms\qdbdc\o11201gbk\incident\incdir_3784\o11201gbk_ora_9480_i3784.trc
         OBJN = 36, OBJD = 36, OBJECT = I_OBJ1, SUBOBJECT = 
         SEGMENT OWNER = SYS, SEGMENT TYPE = Index Segment
Corrupt Block Found
         TSN = 0, TSNAME = SYSTEM
         RFN = 1, BLK = 48404, RDBA = 4242708
         OBJN = 36, OBJD = 36, OBJECT = I_OBJ1, SUBOBJECT = 
         SEGMENT OWNER = SYS, SEGMENT TYPE = Index Segment
Errors in file d:\app\xifenfei\diag\rdbms\qdbdc\o11201gbk\trace\o11201gbk_ora_9480.trc:
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01578: ORACLE 数据块损坏 (文件号 1, 块号 48404)
ORA-01110: 数据文件 1: 'F:\ORADATA\SYSTEM01.DBF'
Errors in file d:\app\xifenfei\diag\rdbms\qdbdc\o11201gbk\trace\o11201gbk_ora_9480.trc:
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01578: ORACLE 数据块损坏 (文件号 1, 块号 48404)
ORA-01110: 数据文件 1: 'F:\ORADATA\SYSTEM01.DBF'
Error 604 happened during db open, shutting down database
USER (ospid: 9480): terminating the instance due to error 604

这里可以看出来数据库不能正常启动的原因,主要是由于I_OBJ1(obj$表的index)刚好被损坏,导致数据库无法,通过分析定位确定是如下sql导致启动失败

Dump continued from file: d:\app\xifenfei\diag\rdbms\qdbdc\o11201gbk\trace\o11201gbk_ora_9480.trc
ORA-01578: ORACLE 数据块损坏 (文件号 1, 块号 48404)
ORA-01110: 数据文件 1: 'F:\ORADATA\SYSTEM01.DBF'

========= Dump for incident 3784 (ORA 1578) ========

*** 2020-03-12 14:52:20.614
dbkedDefDump(): Starting incident default dumps (flags=0x2, level=3, mask=0x0)
----- Current SQL Statement for this session (sql_id=cq514nkrp38hv) -----
select distinct d.p_obj#,d.p_timestamp from sys.dependency$ d, obj$ o where d.p_obj#>=:1 and 
d.d_obj#=o.obj# and o.status not in (5,6)

通过对其i_obj1损坏block进行修复,数据库正启动成功

C:\Users\XIFENFEI>D:\app\XIFENFEI\product\11.2.0.1\dbhome_2\bin\sqlplus / as sysdba

SQL*Plus: Release 11.2.0.1.0 Production on 星期四 3月 12 15:05:48 2020

Copyright (c) 1982, 2010, Oracle.  All rights reserved.

已连接到空闲例程。

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

Total System Global Area 3307048960 bytes
Fixed Size                  2180264 bytes
Variable Size            1811942232 bytes
Database Buffers         1476395008 bytes
Redo Buffers               16531456 bytes
数据库装载完毕。
SQL> alter database open;

数据库已更改。

尝试导出数据

C:\Windows\system32>exp system/oracle owner=gis_sys file=f:/gis_sys.dmp FEEDBACK
=10000  COMPRESS=NO BUFFER=102400000 STATISTICS=none

Export: Release 11.2.0.1.0 - Production on 星期四 3月 12 15:46:19 2020

Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.


连接到: Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
已导出 ZHS16GBK 字符集和 AL16UTF16 NCHAR 字符集

即将导出指定的用户...
. 正在导出 pre-schema 过程对象和操作
. 正在导出用户 GIS_SYS 的外部函数库名
. 导出 PUBLIC 类型同义词
EXP-00008: 遇到 ORACLE 错误 604
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01410: 无效的 ROWID
EXP-00000: 导出终止失败

分析trace文件

*** SESSION ID:(192.3) 2020-03-12 15:05:36.132
OBJD MISMATCH typ=6, seg.obj=18, diskobj=224, dsflg=100100, dsobj=18, tid=18, cls=1
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01410: 无效的 ROWID
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01410: 无效的 ROWID

由于obj id为18(obj$)的对象和对应的数据库中实际block存储的表的block为224(aud$)不匹配,从而出现该错误,通过分析是由于i_obj1记录错误导致该问题,通过修复该记录之后,数据实现完美导出.
20200312202743


类似system文件坏块案例有:
通过拷贝block实现system文件大量坏块恢复
记录一次system表空间坏块(ORA-01578)数据库恢复
SYSTEM表空间坏块恢复—C_TS#对象坏块恢复(file 1 block 60)
file 1 block 128 corrupted/坏块恢复—system rollback坏块修复

发表在 非常规恢复 | 标签为 , , , | 评论关闭

Oracle 19c故障恢复

有客户找到我们,他们的oracle 19c数据库由于异常断电,导致启动异常,经过一系列恢复之后,依旧无法解决问题,请求我们给予支持.通过我们的Oracle数据库异常恢复检查脚本(Oracle Database Recovery Check),获取数据库当前信息如下:
数据库版本为19C并且安装了19.5.0.0.191015 (30125133)补丁
20200310220453
20200310220748


数据库使用pdb
20200310220610

数据库启动成功后,一会就crash掉

2020-03-10T01:44:41.018032+08:00
Pluggable database RACBAK opened read write
2020-03-10T01:44:41.018996+08:00
Pluggable database RAC opened read write
2020-03-10T01:44:51.244050+08:00
Completed: ALTER PLUGGABLE DATABASE ALL OPEN
Starting background process CJQ0
Completed: ALTER DATABASE OPEN
2020-03-10T01:44:51.317085+08:00
CJQ0 started with pid=224, OS id=32581 
2020-03-10T01:44:56.067043+08:00
Errors in file /opt/oracle/diag/rdbms/XFF/XFF/trace/XFF_j001_32588.trc  (incident=1095281) (PDBNAME=RAC):
ORA-00600: internal error code, arguments: [4193], [27733], [27754], [], [], [], [], [], [], [], [], []
RAC(4):Incident details in: /opt/oracle/diag/rdbms/XFF/XFF/incident/incdir_1095281/XFF_j001_32588_i1095281.trc
RAC(4):Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
2020-03-10T01:44:56.073112+08:00
RAC(4):*****************************************************************
RAC(4):An internal routine has requested a dump of selected redo.
RAC(4):This usually happens following a specific internal error, when
RAC(4):analysis of the redo logs will help Oracle Support with the
RAC(4):diagnosis.
RAC(4):It is recommended that you retain all the redo logs generated (by
RAC(4):all the instances) during the past 12 hours, in case additional
RAC(4):redo dumps are required to help with the diagnosis.
RAC(4):*****************************************************************
2020-03-10T01:44:56.079228+08:00
Errors in file /opt/oracle/diag/rdbms/XFF/XFF/trace/XFF_j002_32590.trc  (incident=1095289) (PDBNAME=RAC):
ORA-00600: internal error code, arguments: [4193], [2633], [2638], [], [], [], [], [], [], [], [], []
RAC(4):Incident details in: /opt/oracle/diag/rdbms/XFF/XFF/incident/incdir_1095289/XFF_j002_32590_i1095289.trc
RAC(4):Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
2020-03-10T01:44:56.085068+08:00
RAC(4):*****************************************************************
RAC(4):An internal routine has requested a dump of selected redo.
RAC(4):This usually happens following a specific internal error, when
RAC(4):analysis of the redo logs will help Oracle Support with the
RAC(4):diagnosis.
RAC(4):It is recommended that you retain all the redo logs generated (by
RAC(4):all the instances) during the past 12 hours, in case additional
RAC(4):redo dumps are required to help with the diagnosis.
RAC(4):*****************************************************************
2020-03-10T01:44:56.115765+08:00
Errors in file /opt/oracle/diag/rdbms/XFF/XFF/trace/XFF_j004_32594.trc  (incident=1095305) (PDBNAME=RAC):
ORA-00600: internal error code, arguments: [4193], [63532], [63537], [], [], [], [], [], [], [], [], []
RAC(4):Incident details in: /opt/oracle/diag/rdbms/XFF/XFF/incident/incdir_1095305/XFF_j004_32594_i1095305.trc
RAC(4):Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
2020-03-10T01:46:48.202213+08:00
RAC(4):Recovery of Online Redo Log: Thread 1 Group 2 Seq 2 Reading mem 0
RAC(4):  Mem# 0: /opt/oracle/oradata/XFF/redo02.log
RAC(4):Block recovery completed at rba 0.0.0, scn 0x0000000d3675e48e
RAC(4):DDE: Problem Key 'ORA 600 [4193]' was completely flood controlled (0x6)
Further messages for this problem key will be suppressed for up to 10 minutes
2020-03-10T01:46:48.384040+08:00
Errors in file /opt/oracle/diag/rdbms/XFF/XFF/trace/XFF_clmn_31741.trc:
ORA-00600: internal error code, arguments: [4193], [27733], [27754], [], [], [], [], [], [], [], [], []
Errors in file /opt/oracle/diag/rdbms/XFF/XFF/trace/XFF_clmn_31741.trc  (incident=1093505) (PDBNAME=CDB$ROOT):
ORA-501 [] [] [] [] [] [] [] [] [] [] [] []
Incident details in: /opt/oracle/diag/rdbms/XFF/XFF/incident/incdir_1093505/XFF_clmn_31741_i1093505.trc
2020-03-10T01:46:49.264624+08:00
USER (ospid: 31741): terminating the instance due to ORA error 501
2020-03-10T01:46:49.280664+08:00
System state dump requested by (instance=1, osid=31741 (CLMN)), summary=[abnormal instance termination].
System State dumped to trace file /opt/oracle/diag/rdbms/XFF/XFF/trace/XFF_diag_31759.trc
2020-03-10T01:46:53.156926+08:00
ORA-00501: CLMN process terminated with error
2020-03-10T01:46:53.157103+08:00
Errors in file /opt/oracle/diag/rdbms/XFF/XFF/trace/XFF_diag_31759.trc:
ORA-00501: CLMN process terminated with error
2020-03-10T01:46:53.157211+08:00
Dumping diagnostic data in directory=[cdmp_20200310014649], requested by (instance=1, osid=31741 (CLMN)), 
summary=[abnormal instance termination].

通过报错信息判断,数据库open之后(特别是pdb 4 open之后),开始报ORA-600 4193错误.然后由于CLMN进程异常,最后数据库crash.对于这类故障,因为使用的pdb,而且是由于pdb的undo异常导致数据库启动之后crash,可以通过对于pdb进行特殊处理,从而实现数据库启动之后不再crash.

发表在 ORACLE 19C, Oracle备份恢复 | 标签为 , , , | 评论关闭

oracle文件被删除且部分被覆盖恢复案例

有客户数据库异常,让我们对其进行分析,判断是否可以恢复,让客户通过Oracle数据库异常恢复检查脚本(Oracle Database Recovery Check)收集数据库信息,发现有三个数据文件异常
20200306223254


通过和客户确认,大概情况是这样的:由于/opt目录满了,客户把/opt/oracle整体迁移到/home目录中,然后通过link目录的方式实现迁移,但是在迁移之前把/home/oracle中的所有数据文件给rm掉了,然后把/opt中的数据拷贝到/home/中,在启动数据库的时候提示/home/oracle/JXWR.dbf文件丢失,然后又人工创建了一个该文件,从而出现了上述的三个文件异常(其实删除的数据库文件有十几个,涉及该库的有三个,客户只要恢复JXWR表空间数据即可).对于这种情况,有可能有覆盖的风险,让客户提供空间对现有/home 分区进行镜像,通过工具分析镜像文件
20200305142707
20200305142745

发现需要恢复文件大小/时间均不对,查看内容发现是oracle的审计trace,证明该文件对应的位置已经有覆盖,对于这样的情况,无法从os层面反删除进行恢复,考虑通过oracle碎片层面进行处理,对其分析发现大量block依旧存在(情况有点复杂,因为该目录涉及多个库,通过分析确认相关段为该数据库文件)
20200306230009

检查重组出来的数据文件效果
20200306224628


检查效果比较乐观,因为根据这样的情况,丢失15%左右的block算是非常理想的效果,然后通过oracle dul恢复客户需要的数据1

完成数据库恢复任务.
这次的恢复效果不是太好主要就是由于客户删除文件之后,对被删除文件所在分区进行了大量写操作导致不少数据库block被覆盖,最终只能抱着试试看的态度最大限度恢复,属于比较侥幸的恢复成功,再次提醒各位:在数据被误删除之后,应该先保护现场(不要对其分区进行写操作),覆盖的越少,恢复的效果越好.

发表在 非常规恢复 | 评论关闭