模拟数据库丢失undo表空间

数据库无事务情况下丢失undo表空间数据文件

1. 查看当前undo表空间,并删除物理undo文件

SYS@userdata>show parameter undo_tablespace;

NAME                     TYPE                   VALUE
------------------------------------ --------------------------------- ------------------------------
undo_tablespace              string                   UNDOTBS1
SYS@userdata>select file_name,tablespace_name,status from dba_data_files where tablespace_name='UNDOTBS1';

FILE_NAME                             TABLESPACE_NAME                                        STATUS
------------------------------------------------------------ ------------------------------------------------------------------------------------------ ---------------------------
/u01/app/oracle/oradata/userdata/undotbs01.dbf             UNDOTBS1                                            AVAILABLE

SYS@userdata>host rm -rf /u01/app/oracle/oradata/userdata/undotbs01.dbf

SYS@userdata>shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.

2.将数据库正常启动,可以看到缺少undo表空间数据文件的报错

SYS@userdata>startup;
ORACLE instance started.

Total System Global Area  314572800 bytes
Fixed Size            2083592 bytes
Variable Size          255853816 bytes
Database Buffers       50331648 bytes
Redo Buffers            6303744 bytes
Database mounted.
ORA-01157: cannot identify/lock data file 2 - see DBWR trace file
ORA-01110: data file 2: '/u01/app/oracle/oradata/userdata/undotbs01.dbf'

SYS@userdata>show parameter back;

NAME                     TYPE                   VALUE
------------------------------------ --------------------------------- ------------------------------
background_core_dump             string                   partial
background_dump_dest             string                   /u01/app/oracle/admin/userdata
                                       /bdump
backup_tape_io_slaves             boolean                   FALSE
db_flashback_retention_target         integer                   1440
fast_start_parallel_rollback         string                   LOW
rollback_segments             string
transactions_per_rollback_segment    integer                   5
SYS@userdata>host tail -n 10 /u01/app/oracle/admin/userdata/bdump/alert_userdata.log
Mon Oct  2 10:53:58 2017
ALTER DATABASE OPEN
Mon Oct  2 10:53:58 2017
Errors in file /u01/app/oracle/admin/userdata/bdump/userdata_dbw0_8121.trc:
ORA-01157: cannot identify/lock data file 2 - see DBWR trace file
ORA-01110: data file 2: '/u01/app/oracle/oradata/userdata/undotbs01.dbf'
ORA-27037: unable to obtain file status
Linux-x86_64 Error: 2: No such file or directory
Additional information: 3
ORA-1157 signalled during: ALTER DATABASE OPEN...

3.因为undo表空间此时不包含数据,drop undo表空间

SYS@userdata>alter database datafile 2 offline drop;

Database altered.

4. 将undo管理改为manual

SYS@userdata>alter system set undo_management='MANUAL' scope=spfile;

System altered.

5. 重启数据库

SYS@userdata>shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SYS@userdata>startup;
ORACLE instance started.

Total System Global Area  314572800 bytes
Fixed Size            2083592 bytes
Variable Size          255853816 bytes
Database Buffers       50331648 bytes
Redo Buffers            6303744 bytes
Database mounted.
Database opened.
SYS@userdata>show parameter undo;

NAME                     TYPE                   VALUE
------------------------------------ --------------------------------- ------------------------------
undo_management              string                   MANUAL
undo_retention                 integer                   900
undo_tablespace              string                   UNDOTBS1

6. 将丢失文件的undo表空间drop后重建

SYS@userdata>drop tablespace UNDOTBS1;

Tablespace dropped.

SYS@userdata>create undo tablespace undotbs1 datafile '/u01/app/oracle/oradata/userdata/undotbs01.dbf' size 100m autoextend on;

Tablespace created.

7. 查看目前的数据库undo设置,需要将undo_management 改为auto

SYS@userdata>alter system set undo_management='AUTO' scope=spfile;

System altered.

SYS@userdata>shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SYS@userdata>startup;
ORACLE instance started.

Total System Global Area  314572800 bytes
Fixed Size            2083592 bytes
Variable Size          255853816 bytes
Database Buffers       50331648 bytes
Redo Buffers            6303744 bytes
Database mounted.
Database opened.
SYS@userdata>select name,status,enabled from v$datafile;

NAME                           STATUS         ENABLED
-------------------------------------------------- --------------------- ------------------------------
/u01/app/oracle/oradata/userdata/system01.dbf       SYSTEM         READ WRITE
/u01/app/oracle/oradata/userdata/undotbs01.dbf       ONLINE         READ WRITE
/u01/app/oracle/oradata/userdata/sysaux01.dbf       ONLINE         READ WRITE
/u01/app/oracle/oradata/userdata/users01.dbf       ONLINE         READ WRITE
/u01/app/oracle/oradata/userdata/example01.dbf       ONLINE         READ WRITE

数据库有事务情况下丢失undo表空间数据文件

1. undo表空间数据文件丢失后启动数据库

SYS@userdata>startup
ORACLE instance started.

Total System Global Area  314572800 bytes
Fixed Size            2083592 bytes
Variable Size          255853816 bytes
Database Buffers       50331648 bytes
Redo Buffers            6303744 bytes
Database mounted.
ORA-01157: cannot identify/lock data file 2 - see DBWR trace file
ORA-01110: data file 2: '/u01/app/oracle/oradata/userdata/undotbs01.dbf'

SYS@userdata>alter database datafile 2 offline drop;

Database altered.

SYS@userdata>alter system set undo_management='MANUAL' scope=spfile;

System altered.

SYS@userdata>shutdown immediate;
ORA-01109: database not open

Database dismounted.
ORACLE instance shut down.
SYS@userdata>startup;
ORACLE instance started.

Total System Global Area  314572800 bytes
Fixed Size            2083592 bytes
Variable Size          255853816 bytes
Database Buffers       50331648 bytes
Redo Buffers            6303744 bytes
Database mounted.
ORA-01092: ORACLE instance terminated. Disconnection forced

2. 查看跟踪日志的详细报错

SYS@userdata>host tail -n 10 /u01/app/oracle/admin/userdata/bdump/alert_userdata.log
Database Characterset is AL32UTF8
Mon Oct  2 11:20:10 2017
Errors in file /u01/app/oracle/admin/userdata/udump/userdata_ora_8444.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-00376: file 2 cannot be read at this time
ORA-01110: data file 2: '/u01/app/oracle/oradata/userdata/undotbs01.dbf'
Error 604 happened during db open, shutting down database
USER: terminating instance due to error 604
Instance terminated by USER, pid = 8444
ORA-1092 signalled during: ALTER DATABASE OPEN...

SYS@userdata>host tail -n 10 /u01/app/oracle/admin/userdata/udump/userdata_ora_8444.trc
Longest hash chain = 1
Average hash chain = 20/20 = 1.0
Max compares per lookup = 1
Avg compares per lookup = 95/195 = 0.5
----------------------------------------------
tkcrrsarc: (WARN) Failed to find ARCH for message (message:0x1)
tkcrrpa: (WARN) Failed initial attempt to send ARCH message (message:0x1)
ORA-00604: error occurred at recursive SQL level 1
ORA-00376: file 2 cannot be read at this time
ORA-01110: data file 2: '/u01/app/oracle/oradata/userdata/undotbs01.dbf'

3. 加undo隐藏参数重新启动数据库

SYS@userdata>select ksppinm from x$ksppi where ksppinm like '%roll%';

KSPPINM
--------------------------------------------------
transactions_per_rollback_segment
rollback_segments
_rollback_segment_initial
_rollback_segment_count
_offline_rollback_segments
_corrupted_rollback_segments
_cleanup_rollback_entries
_rollback_stopat
fast_start_parallel_rollback
_mv_rolling_inv

10 rows selected.

SYS@userdata>alter system set "_offline_rollback_segments"=true scope=spfile;

System altered.

SYS@userdata>alter system set "_corrupted_rollback_segments"='_SYSSMU1$','_SYSSMU2$','_SYSSMU3$','_SYSSMU4$','_SYSSMU5$','_SYSSMU6$','_SYSSMU7$','_SYSSMU8$','_SYSSMU9$','_SYSSMU10$' scope=spfile;

System altered.

SYS@userdata>shutdown abort;
ORACLE instance shut down.
SYS@userdata>startup;
ORACLE instance started.

Total System Global Area  314572800 bytes
Fixed Size            2083592 bytes
Variable Size          255853816 bytes
Database Buffers       50331648 bytes
Redo Buffers            6303744 bytes
Database mounted.
Database opened.

4.将丢失文件的undo表空间drop后重建

SYS@userdata>drop tablespace UNDOTBS1;

Tablespace dropped.

SYS@userdata>create undo tablespace undotbs1 datafile '/u01/app/oracle/oradata/userdata/undotbs01.dbf' size 100m autoextend on;

Tablespace created.

5.将undo管理方式修改为auto

SYS@userdata>alter system set undo_management='AUTO' scope=spfile;

System altered.

6.将隐藏参数恢复为默认值

SYS@userdata>alter system reset "_offline_rollback_segments" scope=spfile sid='*';

System altered.

SYS@userdata>alter system reset "_corrupted_rollback_segments" scope=spfile sid='*';

System altered.

7.重启数据库

SYS@userdata>shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SYS@userdata>startup;
ORACLE instance started.

Total System Global Area  314572800 bytes
Fixed Size            2083592 bytes
Variable Size          255853816 bytes
Database Buffers       50331648 bytes
Redo Buffers            6303744 bytes
Database mounted.
Database opened.

SYS@userdata>select name,status,enabled from v$datafile;

NAME                           STATUS         ENABLED
-------------------------------------------------- --------------------- ------------------------------
/u01/app/oracle/oradata/userdata/system01.dbf       SYSTEM         READ WRITE
/u01/app/oracle/oradata/userdata/undotbs01.dbf       ONLINE         READ WRITE
/u01/app/oracle/oradata/userdata/sysaux01.dbf       ONLINE         READ WRITE
/u01/app/oracle/oradata/userdata/users01.dbf       ONLINE         READ WRITE
/u01/app/oracle/oradata/userdata/example01.dbf       ONLINE         READ WRITE