RMAN備份與恢復系列之redo日誌檔案恢復

阿新 • • 發佈：2019-01-28

實驗環境

作業系統 Redhat5.4 x86
資料庫版本 oracle 11gR2 (11.2.0.1.0)
實驗前已經做了RMAN全量備份包括controlfile、spfile

實驗模擬

1. INACTIVE日誌組部分member成員損壞或丟失

案例模擬

模擬group 4 中member disk4b.log檔案丟失：

SQL> select group#, sequence#,bytes/1024/1024M, members, status from v$log;

    GROUP#  SEQUENCE#          M    MEMBERS STATUS
---------- ---------- ---------- ---------- ----------------
     1 
     11        200      2 INACTIVE
     2     12        200      2 INACTIVE
     3     13        200      2 INACTIVE
     4     14        200      2 INACTIVE
     5     15        200      2 CURRENT

SQL> select * from v$logfile;

    GROUP# STATUS  TYPE    MEMBER                         IS_
---------- ------- ------- -------------------------------------------------- ---
     1 
     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk1a.log      NO
     1     ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk1b.log      NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk2a.log      NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk2b.log      NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk3a.log 
      NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk3b.log      NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk4a.log      NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk4b.log      NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk5a.log      NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk5b.log      NO

10 rows selected.

SQL> !rm /u01/app/oracle/oradata/PROD/disk5/disk4b.log

SQL> !ls -la /u01/app/oracle/oradata/PROD/disk5/disk4b.log
ls: /u01/app/oracle/oradata/PROD/disk5/disk4b.log: No such file or directory

案例恢復

首先，在不重啟資料庫情況下，直接利用rman進行恢復：

[[email protected] dbs]$ rman target /

Recovery Manager: Release 11.2.0.1.0 - Production on Sun May 7 11:57:10 2017

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

connected to target database: PROD (DBID=352761597)

RMAN> list failure;

using target database control file instead of recovery catalog
no failures found that match specification
RMAN> exit

Recovery Manager complete.

在不重啟資料庫的情況下，暫時檢查不到檔案丟失，
重啟資料庫以後，再次嘗試：

SQL> shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup;
ORACLE instance started.

Total System Global Area  836976640 bytes
Fixed Size          1339740 bytes
Variable Size         679480996 bytes
Database Buffers      150994944 bytes
Redo Buffers            5160960 bytes
Database mounted.
Database opened.
SQL> exit
Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
--資料庫可以正常啟動，alert日誌中會有報錯提示資訊。

[[email protected] dbs]$ rman target / 

Recovery Manager: Release 11.2.0.1.0 - Production on Sun May 7 11:58:39 2017

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

connected to target database: PROD (DBID=352761597)

RMAN> list failure;

using target database control file instead of recovery catalog
List of Database Failures
=========================

Failure ID Priority Status    Time Detected Summary
---------- -------- --------- ------------- -------
2176       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk5/disk4b.log is missing

RMAN> advise failure;

List of Database Failures
=========================

Failure ID Priority Status    Time Detected Summary
---------- -------- --------- ------------- -------
2176       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk5/disk4b.log is missing

analyzing automatic repair options; this may take some time
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=35 device type=DISK
analyzing automatic repair options complete

Mandatory Manual Actions
========================
no manual actions available

Optional Manual Actions
=======================
1. If file /u01/app/oracle/oradata/PROD/disk5/disk4b.log was unintentionally renamed or moved, restore it

Automated Repair Options
========================
Option Repair Description
------ ------------------
1      Drop and re-create redo log group member /u01/app/oracle/oradata/PROD/disk5/disk4b.log  
  Strategy: The repair includes complete media recovery with no data loss
  Repair script: /u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_3272967096.hm

RMAN> repair failure;

Strategy: The repair includes complete media recovery with no data loss
Repair script: /u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_3272967096.hm

contents of repair script:
   # drop/recreate redo log group member
   sql "begin sys.dbms_ir.execsqlscript(filename => ''/u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_162888668.hm'' ); end;";

Do you really want to execute the above repair (enter YES or NO)? YES
executing repair script

sql statement: begin sys.dbms_ir.execsqlscript(filename => ''/u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_162888668.hm'' ); end;
repair failure complete

RMAN> exit


Recovery Manager complete.
[[email protected] dbs]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.1.0 Production on Sun May 7 17:28:16 2017

Copyright (c) 1982, 2009, Oracle.  All rights reserved.


Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options

SQL> !ls -la /u01/app/oracle/oradata/PROD/disk5/disk4b.log
-rw-r----- 1 oracle oinstall 209715712 May  7 17:28 /u01/app/oracle/oradata/PROD/disk5/disk4b.log

SQL> 

--通過重啟資料庫後，可以正常使用rman進行恢復。

RMAN中使用advise failure 生成的指令碼可以進行檢視具體的操作內容：

[[email protected] ~]$ cat /u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_162888668.hm
begin
/*Drop the Log Member*/
execute immediate 'ALTER DATABASE DROP LOGFILE MEMBER ''/u01/app/oracle/oradata/PROD/disk5/disk4b.log''';
/*Add the Log Member (This will create a new file)*/
execute immediate 'ALTER DATABASE ADD LOGFILE MEMBER ''/u01/app/oracle/oradata/PROD/disk5/disk4b.log'' REUSE TO GROUP 4';
end;

因此，我們同樣可以按照指令碼內容手動進行恢復，inactive 日誌組中redo日誌檔案的恢復(這種方法不需要重啟資料庫)：

SQL> alter database drop logfile member '/u01/app/oracle/oradata/PROD/disk5/disk4b.log';

Database altered.

SQL> alter database add logfile member '/u01/app/oracle/oradata/PROD/disk5/disk4b.log' reuse to group 4;

Database altered.

SQL> !ls -ls /u01/app/oracle/oradata/PROD/disk5/disk4b.log
205008 -rw-r----- 1 oracle oinstall 209715712 May  7 17:49 /u01/app/oracle/oradata/PROD/disk5/disk4b.log

這種情況下的資料庫可以正常啟動，資料恢復不會產生資料丟失。

2. INACTIVE日誌組全部的member成員都損壞或丟失

案例模擬

[[email protected] dbs]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.1.0 Production on Sun May 7 18:15:24 2017

Copyright (c) 1982, 2009, Oracle.  All rights reserved.


Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options

SQL> select * from v$logfile;
    GROUP# STATUS  TYPE    MEMBER                           IS_
---------- ------- ------- ------------------------------------------------------------ ---
     1     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk1a.log        NO
     1 INVALID ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk1b.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk2a.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk2b.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk3a.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk3b.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk4b.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk4a.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk5a.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk5b.log        NO

10 rows selected.
SQL> select group#, sequence#, members, archived, status from v$log;

    GROUP#  SEQUENCE#    MEMBERS ARC STATUS
---------- ---------- ---------- --- ----------------
     1     16          2 YES INACTIVE
     2     17          2 YES INACTIVE
     3     18          2 YES INACTIVE
     4     19          2 YES INACTIVE
     5     20          2 NO  CURRENT

SQL> !rm /u01/app/oracle/oradata/PROD/disk2/disk2a.log

SQL> !rm /u01/app/oracle/oradata/PROD/disk3/disk2b.log

SQL> exit

案例恢復

[[email protected] dbs]$ rman target /

Recovery Manager: Release 11.2.0.1.0 - Production on Sun May 7 18:17:40 2017

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

connected to target database: PROD (DBID=352761597)

RMAN> list failure;

using target database control file instead of recovery catalog
no failures found that match specification
--同樣需要重啟資料庫以後才能檢測到。
RMAN> exit


Recovery Manager complete.
[[email protected] dbs]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.1.0 Production on Sun May 7 18:17:55 2017

Copyright (c) 1982, 2009, Oracle.  All rights reserved.


Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options

SQL> shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup;
ORACLE instance started.

Total System Global Area  836976640 bytes
Fixed Size          1339740 bytes
Variable Size         683675300 bytes
Database Buffers      146800640 bytes
Redo Buffers            5160960 bytes
Database mounted.
ORA-03113: end-of-file on communication channel
Process ID: 27465
Session ID: 1 Serial number: 5


SQL> exit
--重啟資料庫以後發現數據庫無法啟動，報錯。

檢視alert日誌發現以下報錯資訊：

ALTER DATABASE OPEN
LGWR: STARTING ARCH PROCESSES
Sun May 07 18:18:12 2017
ARC0 started with pid=18, OS id=27467 
ARC0: Archival started
LGWR: STARTING ARCH PROCESSES COMPLETE
ARC0: STARTING ARCH PROCESSES
Sun May 07 18:18:13 2017
ARC1 started with pid=19, OS id=27469 
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_lgwr_27413.trc:
ORA-00313: open failed for members of log group 1 of thread 1
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_lgwr_27413.trc:
ORA-00313: open failed for members of log group 2 of thread 1
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk3/disk2b.log'
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk2/disk2a.log'
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_lgwr_27413.trc:
ORA-00313: open failed for members of log group 2 of thread 1
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk3/disk2b.log'
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk2/disk2a.log'
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_ora_27465.trc:
ORA-00313: open failed for members of log group 1 of thread 
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk2/disk2a.log'
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk3/disk2b.log'
USER (ospid: 27465): terminating the instance due to error 313
Sun May 07 18:18:13 2017
ARC2 started with pid=20, OS id=27471 
Instance terminated by USER, pid = 27465

再次登入RMAN進行恢復：

[[email protected] trace]$ rman target /

Recovery Manager: Release 11.2.0.1.0 - Production on Sun May 7 18:19:34 2017

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

connected to target database (not started)

RMAN> startup mount;

Oracle instance started
database mounted

Total System Global Area     836976640 bytes

Fixed Size                     1339740 bytes
Variable Size                683675300 bytes
Database Buffers             146800640 bytes
Redo Buffers                   5160960 bytes

RMAN> list failure;

using target database control file instead of recovery catalog
List of Database Failures
=========================

Failure ID Priority Status    Time Detected Summary
---------- -------- --------- ------------- -------
2351       CRITICAL OPEN      07-MAY-17     Redo log group 2 is unavailable
2357       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk3/disk2b.log is missing
2354       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk2/disk2a.log is missing

RMAN> advise failure;

List of Database Failures
=========================

Failure ID Priority Status    Time Detected Summary
---------- -------- --------- ------------- -------
2351       CRITICAL OPEN      07-MAY-17     Redo log group 2 is unavailable
2357       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk3/disk2b.log is missing
2354       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk2/disk2a.log is missing

analyzing automatic repair options; this may take some time
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=18 device type=DISK
analyzing automatic repair options complete

Mandatory Manual Actions
========================
no manual actions available

Optional Manual Actions
=======================
1. If file /u01/app/oracle/oradata/PROD/disk3/disk2b.log was unintentionally renamed or moved, restore it
2. If file /u01/app/oracle/oradata/PROD/disk2/disk2a.log was unintentionally renamed or moved, restore it

Automated Repair Options
========================
Option Repair Description
------ ------------------
1      Clear redo log group 2  
  Strategy: The repair includes complete media recovery with no data loss
  Repair script: /u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_3329076409.hm
RMAN> repair failure;

Strategy: The repair includes complete media recovery with no data loss
Repair script: /u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_3329076409.hm

contents of repair script:
   # clear redo log group
   sql "begin sys.dbms_ir.execsqlscript(filename => ''/u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_4284357731.hm'' ); end;";

Do you really want to execute the above repair (enter YES or NO)? yes
executing repair script

sql statement: begin sys.dbms_ir.execsqlscript(filename => ''/u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_4284357731.hm'' ); end;
repair failure complete

Do you want to open the database (enter YES or NO)? yes
database opened

RMAN>

恢復完成後，登入資料庫進行檢查：

[[email protected] trace]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.1.0 Production on Sun May 7 18:30:29 2017

Copyright (c) 1982, 2009, Oracle.  All rights reserved.


Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options

SQL> set line 180
SQL> col member for a60
SQL> select group#, sequence#, members, archived, status from v$log;

    GROUP#  SEQUENCE#    MEMBERS ARC STATUS
---------- ---------- ---------- --- ----------------
     1     16          2 YES INACTIVE
     2      0          2 YES UNUSED
     3     18          2 YES INACTIVE
     4     19          2 YES INACTIVE
     5     20          2 NO  CURRENT

SQL> select * from v$logfile;

    GROUP# STATUS  TYPE    MEMBER                           IS_
---------- ------- ------- ------------------------------------------------------------ ---
     1     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk1a.log        NO
     1 INVALID ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk1b.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk2a.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk2b.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk3a.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk3b.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk4b.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk4a.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk5a.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk5b.log        NO

10 rows selected.

SQL> !ls -la /u01/app/oracle/oradata/PROD/disk2/disk2a.log
-rw-r----- 1 oracle oinstall 209715712 May  7 18:25 /u01/app/oracle/oradata/PROD/disk2/disk2a.log

SQL> !ls -la /u01/app/oracle/oradata/PROD/disk3/disk2b.log
-rw-r----- 1 oracle oinstall 209715712 May  7 18:25 /u01/app/oracle/oradata/PROD/disk3/disk2b.log

SQL> 
--之前手動刪除的兩個檔案已經恢復成功。

advise failure 生成的指令碼內容如下：

[[email protected] ~]$ cat /u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_3329076409.hm
   # clear redo log group
   sql "begin sys.dbms_ir.execsqlscript(filename => ''/u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_4284357731.hm'' ); end;";
[[email protected] ~]$ cat /u01/app/oracle/diag/rdbms/prod/PROD/hm/reco_4284357731.hm
begin
/*Clear the Log Group*/
execute immediate 'ALTER DATABASE CLEAR LOGFILE GROUP 2';
end;

因此，對於invactive redo group中所有member都丟失的情況下，我們還可以通過手動執行以下命令進行修復(由於inactive redo group中的日誌已經完成了checkpoint了，因此檔案中的內容已經不再需要，可以直接clear了)：

ALTER DATABASE CLEAR LOGFILE GROUP 2

具體步驟如下：

--模擬檔案丟失或損壞：
SQL> select * from v$logfile;
SQL> /

    GROUP# STATUS  TYPE    MEMBER                           IS_
---------- ------- ------- ------------------------------------------------------------ ---
     1     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk1a.log        NO
     1 INVALID ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk1b.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk2a.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk2b.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk3a.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk3b.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk4b.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk4a.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk5a.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk5b.log        NO

10 rows selected.

SQL> select * from v$log;

    GROUP#    THREAD#  SEQUENCE#      BYTES  BLOCKSIZE    MEMBERS ARC STATUS           FIRST_CHANGE# FIRST_TIM NEXT_CHANGE# NEXT_TIME
---------- ---------- ---------- ---------- ---------- ---------- --- ---------------- ------------- --------- ------------ ---------
     1      1         16  209715200    512      2 YES INACTIVE            640479 07-MAY-17       640496 07-MAY-17
     2      1         21  209715200    512      2 NO  CURRENT             682181 07-MAY-17   2.8147E+14
     3      1         18  209715200    512      2 YES INACTIVE            640502 07-MAY-17       640514 07-MAY-17
     4      1         19  209715200    512      2 YES INACTIVE            640514 07-MAY-17       661074 07-MAY-17
     5      1         20  209715200    512      2 YES INACTIVE            661074 07-MAY-17       682181 07-MAY-17

SQL> 
SQL> !rm /u01/app/oracle/oradata/PROD/disk1/disk1a.log /u01/app/oracle/oradata/PROD/disk2/disk1b.log

SQL> startup force;
ORACLE instance started.

Total System Global Area  836976640 bytes
Fixed Size          1339740 bytes
Variable Size         683675300 bytes
Database Buffers      146800640 bytes
Redo Buffers            5160960 bytes
Database mounted.
ORA-03113: end-of-file on communication channel
Process ID: 29227
Session ID: 1 Serial number: 5


SQL> exit
Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options

--alert日誌報錯資訊如下：
ARC0: STARTING ARCH PROCESSES
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_lgwr_29174.trc:
ORA-00313: open failed for members of log group 1 of thread 1
ORA-00312: online log 1 thread 1: '/u01/app/oracle/oradata/PROD/disk1/disk1a.log'
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_lgwr_29174.trc:
ORA-00313: open failed for members of log group 1 of thread 1
ORA-00312: online log 1 thread 1: '/u01/app/oracle/oradata/PROD/disk1/disk1a.log'
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
Sun May 07 18:45:46 2017
ARC1 started with pid=19, OS id=29232 
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_ora_29227.trc:
ORA-00313: open failed for members of log group 1 of thread 
ORA-00312: online log 1 thread 1: '/u01/app/oracle/oradata/PROD/disk1/disk1a.log'
ORA-00312: online log 1 thread 1: '/u01/app/oracle/oradata/PROD/disk2/disk1b.log'
USER (ospid: 29227): terminating the instance due to error 313
Instance terminated by USER, pid = 29227


--手動恢復方法：
[[email protected] ~]$ sqlplus / as sysdba

SQL*Plus: Release 11.2.0.1.0 Production on Sun May 7 18:46:05 2017

Copyright (c) 1982, 2009, Oracle.  All rights reserved.

Connected to an idle instance.

SQL> startup mount;
ORACLE instance started.

Total System Global Area  836976640 bytes
Fixed Size          1339740 bytes
Variable Size         683675300 bytes
Database Buffers      146800640 bytes
Redo Buffers            5160960 bytes
Database mounted.
SQL> alter database clear logfile group 1;

Database altered.

SQL> !ls -la /u01/app/oracle/oradata/PROD/disk2/disk1b.log
-rw-r----- 1 oracle oinstall 209715712 May  7 18:47 /u01/app/oracle/oradata/PROD/disk2/disk1b.log

SQL> !ls -la /u01/app/oracle/oradata/PROD/disk1/disk1a.log
-rw-r----- 1 oracle oinstall 209715712 May  7 18:47 /u01/app/oracle/oradata/PROD/disk1/disk1a.log

--執行完clear logfile命令以後檔案已經自動生成，下面可以直接將資料庫啟動到open狀態。

SQL> alter database open;

Database altered.

SQL>

這種情況下，資料庫無法正常啟動，資料恢復不會造成資料丟失。

3. ACTIVE非CURRENT日誌組的部分member成員都損壞或丟失

案例模擬

SQL> select group#, sequence#, status, archived from v$log;

    GROUP#  SEQUENCE# STATUS           ARC
---------- ---------- ---------------- ---
     1     16 INACTIVE         YES
     2     17 ACTIVE           YES
     3     18 CURRENT          NO
     4     14 INACTIVE         YES
     5     15 INACTIVE         YES

SQL> select * from v$logfile;

    GROUP# STATUS  TYPE    MEMBER                           IS_
---------- ------- ------- ------------------------------------------------------------ ---
     1     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk1a.log        NO
     1     ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk1b.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk2/disk2a.log        NO
     2     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk2b.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk3b.log        NO
     3     ONLINE  /u01/app/oracle/oradata/PROD/disk3/disk3a.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk4/disk4a.log        NO
     4     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk4b.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk5/disk5a.log        NO
     5     ONLINE  /u01/app/oracle/oradata/PROD/disk1/disk5b.log        NO

10 rows selected.

SQL> !rm /u01/app/oracle/oradata/PROD/disk3/disk2b.log
SQL> shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup;
ORACLE instance started.

Total System Global Area  836976640 bytes
Fixed Size          1339740 bytes
Variable Size         683675300 bytes
Database Buffers      146800640 bytes
Redo Buffers            5160960 bytes
Database mounted.
Database opened.
SQL> exit
Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options

--資料庫可以正常啟動，檢視alert日誌存在以下報錯資訊：

Sun May 07 20:57:11 2017
ARC0 started with pid=18, OS id=3495 
ARC0: Archival started
LGWR: STARTING ARCH PROCESSES COMPLETE
ARC0: STARTING ARCH PROCESSES
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_lgwr_3433.trc:
ORA-00313: open failed for members of log group 2 of thread 1
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk3/disk2b.log'
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory
Additional information: 3
Errors in file /u01/app/oracle/diag/rdbms/prod/PROD/trace/PROD_lgwr_3433.trc:
ORA-00321: log 2 of thread 1, cannot update log file header
ORA-00312: online log 2 thread 1: '/u01/app/oracle/oradata/PROD/disk3/disk2b.log'
Sun May 07 20:57:12 2017
ARC1 started with pid=19, OS id=3497

案例恢復

[[email protected] ~]$ rman target /

Recovery Manager: Release 11.2.0.1.0 - Production on Sun May 7 20:57:49 2017

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

connected to target database: PROD (DBID=352761597)

RMAN> list failure;

using target database control file instead of recovery catalog
List of Database Failures
=========================

Failure ID Priority Status    Time Detected Summary
---------- -------- --------- ------------- -------
2945       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk3/disk2b.log is missing

RMAN> advise failure;

List of Database Failures
=========================

Failure ID Priority Status    Time Detected Summary
---------- -------- --------- ------------- -------
2945       HIGH     OPEN      07-MAY-17     Redo log file /u01/app/oracle/oradata/PROD/disk3/disk2b.log is missing

analyzing automatic repair options; this may take some time
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=34 device type=DISK
analyzing automatic repair options complete

Mandatory Manual Actions
========================
no manual actions available

Optional Manual Actions
====

 
 
              
           
              
              
            
            相關推薦
			   
            
            
            
 

    

    
    RMAN備份與恢復系列之redo日誌檔案恢復
      
							
							
							實驗環境


作業系統 Redhat5.4 x86
資料庫版本 oracle 11gR2 (11.2.0.1.0)
實驗前已經做了RMAN全量備份包括controlfile、spfile




實驗模擬



1. INACTIVE日誌組部分member成員損 

  
 

    

    
    RMAN備份與恢復之基於時間點的不完全恢復
      
                
一 不完全恢復講解

        在上一篇文章（RMAN備份與恢復之UNDO表空間丟失）中，我們講到UNDO表空間丟失怎麼處理。UNDO表空間是不可離線的，同樣，SYSTEM表空間也是不可離線的，所以SYSTEM表空間丟失的處理辦法和UNDO表空間丟失的處理辦法類似，在 

  
 

    

    
    數據庫的備份與還原系列——單表備份和恢復詳細完整實現
      單表備份   單表還原   表定義備份還原   表數據備份還原   參考實現：https://www.percona.com/doc/percona-xtrabackup/LATEST/innobackupex/innobackupex_script.htmlRestoring Individual Tabl 

  
 

    

    
    catalog方式的rman備份與恢復示例
      
C:\>rman catalog rman/rman target sys/orcl
RMAN>create script fullbk{      #建立備份指令碼
backup full tag 'sys' database
include current controlfile
forma 

  
 

    

    
    程序設計基石與實踐系列之按值傳遞還是按引用
      有趣   name   align   pos   str   堆棧   技術分享   easy   pan   
從簡單的樣例開始.如果我們要交換兩個整形變量的值，在C/C++中怎麽做呢？我們來看多種方式，哪種能夠做到.void call_by_ref(int &p,int &q) { // 

  
 

    

    
    數據庫的備份與還原系列——全備份+兩增量的備份與還原
      完全備份   增量備份   數據還原   場景：周日全備份，周一，周二增量備份，周三出現問題，數分鐘後發現；
數據庫的安裝：
yum install mariadb-server -y
systemctl start mariadb
yum install percona-xtrabackup-24-2.4. 

  
 

    

    
    jvm系列之-gc日誌檢視
       
  
  
  
  在上一篇中我們介紹的有引數的設定，那我們要了解不同的引數對於程式的影響和對於調優的影響，這個時候我們就需要檢視gc日誌來看我們設定的引數是否合理，這一章節中我們就來檢視gc日誌 
  
 一、新生代和老年代的比值對效能的影響 
  Map map = new HashMap<& 

  
 

    

    
    自己動手編寫一個Linux偵錯程式系列之4 ELF檔案格式與DWARF除錯格式
      
                目錄





在上一節中，你已經聽說了DWARF除錯格式，它是程式的除錯資訊，是一種可以更好理解原始碼的方式，而不只是解析程式。今天我們將討論原始碼級除錯資訊的細節，以準備在本教程後面的部分中使用它。

系列索引

準備工作
	斷點的設定
	暫存器和記憶體
	ELF檔案格式 

  
 

    

    
    tomcat 系列之 啟動日誌分析
      
							
							
							一下是tomcat9的啟動資訊，以及每行日誌的說明


  tomcat伺服器的版本 
  26-Sep-2017 07:35:00.036 INFO [main] org.apache.catalina.startup.VersionLoggerListene 

  
 

    

    
    大資料與Hadoop系列之壓縮(一)
      
                Hadoop壓縮簡介Hadoop作為一個較通用的海量數椐處理平臺，在使用壓縮方式方面，主要考慮壓縮速度和壓縮檔案的可分割性.所有的壓縮演算法都會考慮時間和空間的權衡，更快的壓縮和解壓縮速度通常會耗費更多的交間（壓縮比較低）例如：通過gzip命令壓縮資料時，使用者可以設定不同的 

  
 

    

    
    【oracle11g,11】redo日誌檔案2 ：日誌恢復 (重點)
      

用alter database clear logfile命令嘗試重構logfile group 1,失敗；
SQL> alter database clear logfile group 1;
alter database clear logfile group 1
*
ERROR at line 

  
 

    

    
    程式設計基石與實踐系列之失落的C語言結構體封裝藝術
      
                誰該閱讀這篇文章本文是關於削減C語言程式記憶體佔用空間的一項技術——為了減小記憶體大小而手工重新封裝C結構體宣告。你需要C語言的基本知識來讀懂本文。如果你要為記憶體有限制的嵌入式系統、或者作業系統核心寫程式碼，那麼你需要懂這項技術。如果你在處理極大的應用程式資料集，以至於你的 

  
 

    

    
    KVM原理與架構系列之四 記憶體虛擬化篇
      
                
記憶體虛擬化


1. 客戶機實體地址空間

在物理機上，虛擬地址通過Guest頁表即可轉換為實體地址。但是在虛擬化環境中，由於VMM和VM都需要獨立的地址空間，則產生了衝突。

為實現記憶體虛擬化，讓客戶機使用一個隔離的、從零開始且具有連續的記憶體空間，KVM 引入一層新 

  
 

    

    
    深入理解MySQL系列之redo log、undo log和binlog
      #### 事務的實現

redo log保證事務的永續性，undo log用來幫助事務回滾及MVCC的功能。

#### InnoDB儲存引擎體系結構
![](https://img2020.cnblogs.com/blog/2211828/202012/2211828-20201222093735507-1 

  
 

    

    
    ARCHIVELOG模式下使用者管理恢復聯機重做日誌檔案—當前活動組所有成員全部損壞
       
 
 1、在關閉狀態下 當前活動組所有成員全部損壞，需要不完全恢復然後resetlogs開啟資料庫。恢復完成後會自動建立一個丟失了的online redo logfile。 
  
   
   
    [sql] 
     view plain
    copy
    print
    

  
 

    

    
    ARCHIVELOG模式下使用者管理恢復聯機重做日誌檔案—非活動組所有成員全部損壞
       
 
 聯機重做日誌檔案至少需要兩組，oracle建議每組的成員至少要兩個，也需要多路複用的。因為每組的成員的內容的都是一樣的。同一組內只要有一個成員還存在就可以保證不丟資料的。 
   
 1、在open狀態下非活動組所有成員全部損壞，可以重建一個成員。 
  
   
   
    [sql 

  
 

    

    
    mybatis系列之XML對映檔案
      
							
							
							一、XML對映檔案
1.1 Mybatis的增刪改操作
mybatis允許增刪改直接定義以下型別返回值
Integer、Long、Boolean、void


1.2 sqlsession
sqlSessionFactory.openSession(); 獲取的 

  
 

    

    
    十四.Spark SQL總結之spark日誌檔案資料形式的轉換
      
							
							
							第一步.資料來源
找到spark的日誌資料來源,在/root/spark/spark-2.0.2-bin-hadoop2.7/logs目錄下:

通過對檔案的讀取,統計其中資料的條數:
val masterLog = sc.textFile("file:///r 

  
 

    

    
    axis2系列之services.xml檔案詳解
      
                

   原文連結（http://beyond-gzz.blog.163.com/blog/static/744735620119935948592/ ）在Axis1中部署服務時，我們使用service.wsdd檔案來配置服務。在Axis2中，不再使用service.wsdd 

  
 

    

    
    mysql備份與還原操作手冊（.sql檔案操作指南）
       
  
  
 轉載自： https://yq.aliyun.com/articles/626120?utm_content=m_1000026933 
 一、備份常用操作基本命令 
 1、備份命令mysqldump格式 
 格式：mysqldump -h主機名  -P埠 -u使用者名稱 -p密碼 –dat

RMAN備份與恢復系列之redo日誌檔案恢復

實驗環境

實驗模擬

1. INACTIVE日誌組部分member成員損壞或丟失

案例模擬

案例恢復

2. INACTIVE日誌組全部的member成員都損壞或丟失

案例模擬

案例恢復

3. ACTIVE非CURRENT日誌組的部分member成員都損壞或丟失

案例模擬

案例恢復

RMAN備份與恢復系列之redo日誌檔案恢復

RMAN備份與恢復之基於時間點的不完全恢復

數據庫的備份與還原系列——單表備份和恢復詳細完整實現

catalog方式的rman備份與恢復示例

程序設計基石與實踐系列之按值傳遞還是按引用

數據庫的備份與還原系列——全備份+兩增量的備份與還原

jvm系列之-gc日誌檢視

自己動手編寫一個Linux偵錯程式系列之4 ELF檔案格式與DWARF除錯格式

tomcat 系列之啟動日誌分析

大資料與Hadoop系列之壓縮(一)

【oracle11g,11】redo日誌檔案2 ：日誌恢復 (重點)

程式設計基石與實踐系列之失落的C語言結構體封裝藝術

KVM原理與架構系列之四記憶體虛擬化篇

深入理解MySQL系列之redo log、undo log和binlog

ARCHIVELOG模式下使用者管理恢復聯機重做日誌檔案—當前活動組所有成員全部損壞

ARCHIVELOG模式下使用者管理恢復聯機重做日誌檔案—非活動組所有成員全部損壞

mybatis系列之XML對映檔案

十四.Spark SQL總結之spark日誌檔案資料形式的轉換

axis2系列之services.xml檔案詳解

mysql備份與還原操作手冊（.sql檔案操作指南）