Oracle主庫歸檔丟失，備庫日誌有gap，在不重建備庫的情況下，恢復備庫
本文主要描述Oracle備庫日誌與主庫日誌之間有gap，切主庫這部分gap的歸檔日誌已經刪除或丟失，如何在不重建備庫的情況下，恢復備庫。
歡迎轉載，請註明作者、出處。
作者：張正
blog：http://space.itpub.net/26355921
QQ：176036317
如有疑問，歡迎聯絡。
在dataguard環境中，由於主庫archivelog丟失，且尚未同步到standby，問如何在避免
重建standby的情況下來將standby恢復成功的。 下面是我的測試過程，供參考！
—-主庫
SQL> SELECT database_role FROM v$database;
DATABASE_ROLE
----------------
PRIMARY
SQL> archive log list;
DATABASE log mode Archive Mode
Automatic archival Enabled
Archive destination /arch
Oldest online log SEQUENCE 382
NEXT log SEQUENCE TO archive 383
CURRENT log SEQUENCE 383
SQL> SELECT COUNT(1) FROM roger.test;
COUNT(1)
----------
30
—-備庫
SQL> SELECT database_role FROM v$database;
DATABASE_ROLE
----------------
PHYSICAL STANDBY
SQL> archive log list;
DATABASE log mode Archive Mode
Automatic archival Enabled
Archive destination /arch2
Oldest online log SEQUENCE 382
NEXT log SEQUENCE TO archive 0
CURRENT log SEQUENCE 383
SQL>
SQL> ALTER DATABASE recover managed standby DATABASE cancel;
DATABASE altered.
SQL> ALTER DATABASE OPEN READ ONLY;
DATABASE altered.
SQL> SELECT COUNT(1) FROM roger.test;
COUNT(1)
----------
30
模擬主庫丟失歸檔的情況：
—主庫
SQL> SELECT * FROM v$Log;
GROUP# THREAD# SEQUENCE# BYTES MEMBERS ARC STATUS FIRST_CHANGE# FIRST_TIME
---------- ---------- ---------- ---------- ---------- --- ---------------- ------------- ------------
1 1 383 10485760 1 NO CURRENT 740638 10-OCT-12
2 1 382 20971520 1 YES INACTIVE 740633 10-OCT-12
SQL> archive log list;
DATABASE log mode Archive Mode
Automatic archival Enabled
Archive destination /arch
Oldest online log SEQUENCE 382
NEXT log SEQUENCE TO archive 383
CURRENT log SEQUENCE 383
SQL> ALTER system SET log_archive_dest_state_2 = 'defer';
System altered.
SQL> DELETE FROM roger.test WHERE rownum < 11;
10 ROWS deleted.
SQL> commit;
Commit complete.
SQL> SELECT COUNT(*) FROM roger.test;
COUNT(*)
----------
20
SQL> ALTER system switch logfile;
System altered.
SQL> /
System altered.
SQL> /
System altered.
SQL> !
[oracle@primarydb ~]$ cd /arch
[oracle@primarydb arch]$ ls -ltr|tail -10
-rw-r----- 1 oracle dba 3456000 Oct 10 15:18 1_374_726529113.dbf
-rw-r----- 1 oracle dba 593408 Oct 10 15:23 1_375_726529113.dbf
-rw-r----- 1 oracle dba 4608 Oct 10 15:23 1_376_726529113.dbf
-rw-r----- 1 oracle dba 1024 Oct 10 15:23 1_377_726529113.dbf
-rw-r----- 1 oracle dba 68096 Oct 10 15:25 1_378_726529113.dbf
-rw-r----- 1 oracle dba 1297408 Oct 10 15:34 1_379_726529113.dbf
-rw-r----- 1 oracle dba 2048 Oct 10 18:35 1_382_726529113.dbf
-rw-r----- 1 oracle dba 166400 Oct 10 18:44 1_383_726529113.dbf
-rw-r----- 1 oracle dba 2560 Oct 10 18:44 1_384_726529113.dbf
-rw-r----- 1 oracle dba 1024 Oct 10 18:44 1_385_726529113.dbf
[oracle@primarydb arch]$ rm 1_383_726529113.dbf ---刪除archivelog
[oracle@primarydb arch]$ rm 1_384_726529113.dbf
[oracle@primarydb arch]$ exit
exit
SQL> ALTER system SET log_archive_dest_state_2 = 'enable';
System altered.
SQL>
—備庫
SQL> ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION;
DATABASE altered.
SQL>
此時alert log資訊如下：
Tue Nov 13 14:45:26 2012
MRP0: Background Managed Standby Recovery process started (test)
Managed Standby Recovery NOT USING REAL TIME Apply
parallel recovery started WITH 2 processes
Tue Nov 13 14:45:31 2012
Waiting FOR ALL non-CURRENT ORLs TO be archived...
Media Recovery Waiting FOR thread 1 SEQUENCE 383
Tue Nov 13 14:45:32 2012
Completed: ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION
Tue Nov 13 14:45:59 2012
Redo Shipping Client Connected AS PUBLIC
-- Connected User is Valid
RFS[2]: Assigned TO RFS process 9542
RFS[2]: IDENTIFIED DATABASE TYPE AS 'physical standby'
RFS[2]: Archived Log: '/arch2/1_385_726529113.dbf'
Tue Nov 13 14:46:02 2012
Fetching gap SEQUENCE IN thread 1, gap SEQUENCE 383-384
Tue Nov 13 14:47:02 2012
FAL[client]: Failed TO request gap SEQUENCE
GAP - thread 1 SEQUENCE 383-384
DBID 2024668720 branch 726529113
FAL[client]: ALL defined FAL servers have been attempted.
-------------------------------------------------------------
CHECK that the CONTROL_FILE_RECORD_KEEP_TIME initialization
parameter IS defined TO a VALUE that IS sufficiently LARGE
enough TO maintain adequate log switch information TO resolve
archivelog gaps.
如何在重建standby的情況下搞好備庫呢？mos上,其實也有文章進行描述的,就是利用rman進行增量scn的恢復,下面我來進行展示：
1)首先定位到scn
SQL> SELECT SEQUENCE#,FIRST_CHANGE#,NEXT_CHANGE# FROM v$archived_log WHERE SEQUENCE# > 382 ORDER BY 1;
SEQUENCE# FIRST_CHANGE# NEXT_CHANGE#
---------- ------------- ------------
383 740638 740911
384 740911 740915
385 740915 740917
385 740915 740917
2)根據scn，進行rman增量備份
[oracle@primarydb ~]$ cd $ORACLE_HOME/bin
[oracle@primarydb bin]$ ./rman target /
Recovery Manager: Release 10.2.0.4.0 - Production on Wed Oct 10 18:51:57 2012
Copyright (c) 1982, 2007, Oracle. All rights reserved.
connected to target database: TEST (DBID=2024668720)
RMAN> backup device type disk incremental from scn 740638 database format '/tmp/test_db_incre.bbk';
Starting backup at 10-OCT-12
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=147 devtype=DISK
channel ORA_DISK_1: starting full datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
input datafile fno=00004 name=/oracle/product/oradata/test/test01.dbf
input datafile fno=00007 name=/oracle/product/oradata/test/sysaux01.dbf
input datafile fno=00009 name=/oracle/product/oradata/test/test03.dbf
input datafile fno=00006 name=/oracle/product/oradata/test/undo02.dbf
input datafile fno=00001 name=/oracle/product/oradata/test/system01.dbf
input datafile fno=00005 name=/oracle/product/oradata/test/perfstat.dbf
input datafile fno=00003 name=/oracle/product/oradata/test/rman.dbf
input datafile fno=00002 name=/oracle/product/oradata/test/undo01.dbf
input datafile fno=00008 name=/oracle/product/oradata/test/test02.dbf
input datafile fno=00010 name=/oracle/product/oradata/test/test04.dbf
input datafile fno=00011 name=/oracle/product/oradata/test/test05.dbf
channel ORA_DISK_1: starting piece 1 at 10-OCT-12
channel ORA_DISK_1: finished piece 1 at 10-OCT-12
piece handle=/tmp/test_db_incre.bbk tag=TAG20121010T185204 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:03:26
Finished backup at 10-OCT-12
RMAN>
3) 複製增量備份到standby
[oracle@primarydb bin]$ cd /tmp
[oracle@primarydb tmp]$ scp test_db_incre.bbk 192.168.3.176:/tmp/backup
The authenticity of host '192.168.3.176 (192.168.3.176)' can't be established.
RSA key fingerprint is a4:54:6b:bf:12:34:42:73:f5:ba:5f:38:c7:28:9c:b5.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added '192.168.3.176' (RSA) to the list of known hosts.
oracle@192.168.3.176's password:
test_db_incre.bbk 100% 736KB 736.0KB/s 00:00
[oracle@primarydb tmp]$
4) standby進行recover
[oracle@standbydb bin]$ sqlplus "/as sysdba"
SQL*Plus: Release 10.2.0.4.0 - Production ON Tue Nov 13 15:03:46 2012
Copyright (c) 1982, 2007, Oracle. ALL Rights Reserved.
Connected TO:
Oracle DATABASE 10g Enterprise Edition Release 10.2.0.4.0 - Production
WITH the Partitioning, OLAP, DATA Mining AND REAL Application Testing options
SQL> ALTER DATABASE recover managed standby DATABASE cancel;
DATABASE altered.
SQL> exit
Disconnected FROM Oracle DATABASE 10g Enterprise Edition Release 10.2.0.4.0 - Production
WITH the Partitioning, OLAP, DATA Mining AND REAL Application Testing options
[oracle@standbydb ~]$ cd $ORACLE_HOME/bin
[oracle@standbydb bin]$ ./rman target /
Recovery Manager: Release 10.2.0.4.0 - Production ON Tue Nov 13 14:58:25 2012
Copyright (c) 1982, 2007, Oracle. ALL rights reserved.
connected TO target DATABASE: TEST (DBID=2024668720, NOT OPEN)
RMAN> catalog backuppiece '/tmp/backup/test_db_incre.bbk';
USING target DATABASE control file instead OF recovery catalog
cataloged backuppiece
backup piece handle=/tmp/backup/test_db_incre.bbk recid=49 stamp=799253957
RMAN> recover DATABASE noredo;
Starting recover at 13-NOV-12
USING target DATABASE control file instead OF recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=97 devtype=DISK
channel ORA_DISK_1: starting incremental datafile backupset restore
channel ORA_DISK_1: specifying datafile(s) TO restore FROM backup SET
destination FOR restore OF datafile 00001: /oracle/product/oradata/test/system01.dbf
destination FOR restore OF datafile 00002: /oracle/product/oradata/test/undo01.dbf
destination FOR restore OF datafile 00003: /oracle/product/oradata/test/rman.dbf
destination FOR restore OF datafile 00004: /oracle/product/oradata/test/test01.dbf
destination FOR restore OF datafile 00005: /oracle/product/oradata/test/perfstat.dbf
destination FOR restore OF datafile 00006: /oracle/product/oradata/test/undo02.dbf
destination FOR restore OF datafile 00007: /oracle/product/oradata/test/sysaux01.dbf
destination FOR restore OF datafile 00008: /oracle/product/oradata/test/test02.dbf
destination FOR restore OF datafile 00009: /oracle/product/oradata/test/test03.dbf
destination FOR restore OF datafile 00010: /oracle/product/oradata/test/test04.dbf
destination FOR restore OF datafile 00011: /oracle/product/oradata/test/test05.dbf
channel ORA_DISK_1: reading FROM backup piece /tmp/backup/test_db_incre.bbk
channel ORA_DISK_1: restored backup piece 1
piece handle=/tmp/backup/test_db_incre.bbk tag=TAG20121010T185204
channel ORA_DISK_1: restore complete, elapsed TIME: 00:00:01
Finished recover at 13-NOV-12
RMAN>
如下資訊是recover時，alert log的記錄資訊：
Tue Nov 13 15:04:05 2012
Completed: ALTER DATABASE recover managed standby DATABASE cancel
Tue Nov 13 15:04:55 2012
Incremental restore complete OF datafile 2 /oracle/product/oradata/test/undo01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 8 /oracle/product/oradata/test/test02.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 10 /oracle/product/oradata/test/test04.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 11 /oracle/product/oradata/test/test05.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 3 /oracle/product/oradata/test/rman.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 5 /oracle/product/oradata/test/perfstat.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 1 /oracle/product/oradata/test/system01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 6 /oracle/product/oradata/test/undo02.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 4 /oracle/product/oradata/test/test01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 7 /oracle/product/oradata/test/sysaux01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 9 /oracle/product/oradata/test/test03.dbf
checkpoint IS 741134
5) 開啟standby同步，檢查是否ok。
SQL> ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION;
DATABASE altered.
SQL> SELECT database_role,open_mode FROM v$database;
DATABASE_ROLE OPEN_MODE
---------------- ----------
PHYSICAL STANDBY MOUNTED
--主庫切換日誌
SQL> SELECT database_role,open_mode FROM v$database;
DATABASE_ROLE OPEN_MODE
---------------- ----------
PRIMARY READ WRITE
SQL>
SQL> ALTER system switch logfile;
System altered.
SQL> SELECT MAX(al.SEQUENCE#) "Last Seq Recieved",
2 MAX(lh.SEQUENCE#) "Last Seq Applied"
3 FROM v$archived_log al, v$log_history lh;
LAST Seq Recieved LAST Seq Applied
----------------- ----------------
386 386
SQL>
--備庫
SQL> SELECT MAX(al.SEQUENCE#) "Last Seq Recieved",
2 MAX(lh.SEQUENCE#) "Last Seq Applied"
3 FROM v$archived_log al, v$log_history lh;
LAST Seq Recieved LAST Seq Applied
----------------- ----------------
386 382
SQL>
SQL> SELECT * FROM v$archive_gap;
THREAD# LOW_SEQUENCE# HIGH_SEQUENCE#
---------- ------------- --------------
1 383 384
備庫archivelog：
[oracle@standbydb arch2]$ ls -ltr |tail -5
-rw-r----- 1 oracle dba 2048 Nov 13 14:33 1_381_726529113.dbf
-rw-r----- 1 oracle dba 2048 Nov 13 14:34 1_382_726529113.dbf
-rw-r----- 1 oracle dba 1024 Nov 13 14:45 1_385_726529113.dbf
-rw-r----- 1 oracle dba 1262080 Nov 13 15:06 1_386_726529113.dbf
備庫alert log此時的資訊：
Tue Nov 13 15:06:17 2012
ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION
Tue Nov 13 15:06:17 2012
Attempt TO START background Managed Standby Recovery process (test)
MRP0 started WITH pid=19, OS id=13497
Tue Nov 13 15:06:17 2012
MRP0: Background Managed Standby Recovery process started (test)
Managed Standby Recovery NOT USING REAL TIME Apply
parallel recovery started WITH 2 processes
Tue Nov 13 15:06:22 2012
Waiting FOR ALL non-CURRENT ORLs TO be archived...
Media Recovery Waiting FOR thread 1 SEQUENCE 383
Fetching gap SEQUENCE IN thread 1, gap SEQUENCE 383-384
Tue Nov 13 15:06:23 2012
Completed: ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION
Tue Nov 13 15:06:54 2012
Redo Shipping Client Connected AS PUBLIC
-- Connected User is Valid
RFS[3]: Assigned TO RFS process 13596
RFS[3]: IDENTIFIED DATABASE TYPE AS 'physical standby'
PRIMARY DATABASE IS IN MAXIMUM PERFORMANCE mode
PRIMARY DATABASE IS IN MAXIMUM PERFORMANCE mode
RFS[3]: Successfully opened standby log 4: '/oracle/product/oradata/test/redo04.log'
Tue Nov 13 15:06:54 2012
RFS[2]: Successfully opened standby log 3: '/oracle/product/oradata/test/redo03.log'
Tue Nov 13 15:07:23 2012
FAL[client]: Failed TO request gap SEQUENCE
GAP - thread 1 SEQUENCE 383-384
DBID 2024668720 branch 726529113
FAL[client]: ALL defined FAL servers have been attempted.
-------------------------------------------------------------
CHECK that the CONTROL_FILE_RECORD_KEEP_TIME initialization
parameter IS defined TO a VALUE that IS sufficiently LARGE
enough TO maintain adequate log switch information TO resolve
archivelog gaps.
我們可以看到，雖然備庫，仍然在提示383,384是gap ，但是實際上已經是同步的了。
6)最後來驗證下資料
SQL> ALTER DATABASE recover managed standby DATABASE cancel;
DATABASE altered.
SQL> SELECT file#,checkpoint_change# FROM v$datafile ORDER BY 1;
FILE# CHECKPOINT_CHANGE#
---------- ------------------
1 740638
2 740638
3 740638
4 740638
5 740638
6 740638
7 740638
8 740638
9 740638
10 740638
11 740638
11 ROWS selected.
SQL> SELECT file#,checkpoint_change# FROM v$datafile_header ORDER BY 1;
FILE# CHECKPOINT_CHANGE#
---------- ------------------
1 741134
2 741134
3 741134
4 741134
5 741134
6 741134
7 741134
8 741134
9 741134
10 741134
11 741134
11 ROWS selected.
SQL> ALTER DATABASE OPEN READ ONLY;
ALTER DATABASE OPEN READ ONLY
*
ERROR at line 1:
ORA-16004: backup DATABASE requires recovery
ORA-01152: file 1 was NOT restored FROM a sufficiently OLD backup
ORA-01110: DATA file 1: '/oracle/product/oradata/test/system01.dbf'
重建備庫的standby controlfile：
SQL> ALTER DATABASE CREATE standby controlfile AS '/tmp/standby.ctl';
DATABASE altered.
SQL> exit
Disconnected FROM Oracle DATABASE 10g Enterprise Edition Release 10.2.0.4.0 - Production
WITH the Partitioning, OLAP, DATA Mining AND REAL Application Testing options
[oracle@primarydb tmp]$ scp standby.ctl 192.168.3.176:/tmp/backup
oracle@192.168.3.176's password:
standby.ctl 100% 3928KB 3.8MB/s 00:00
[oracle@primarydb tmp]$
[oracle@standbydb bin]$ ./rman target /
Recovery Manager: Release 10.2.0.4.0 - Production on Tue Nov 13 15:19:09 2012
Copyright (c) 1982, 2007, Oracle. All rights reserved.
connected to target database (not started)
RMAN> startup nomount
Oracle instance started
Total System Global Area 264241152 bytes
Fixed Size 1266944 bytes
Variable Size 209718016 bytes
Database Buffers 50331648 bytes
Redo Buffers 2924544 bytes
RMAN> restore controlfile from '/tmp/backup/standby.ctl';
Starting restore at 13-NOV-12
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=101 devtype=DISK
channel ORA_DISK_1: copied control file copy
output filename=/oracle/product/oradata/test/control01.ctl
Finished restore at 13-NOV-12
RMAN> startup mount
database is already started
database mounted
released channel: ORA_DISK_1
RMAN> exit
Recovery Manager complete.
[oracle@standbydb bin]$ sqlplus "/as sysdba"
SQL*Plus: Release 10.2.0.4.0 - Production on Tue Nov 13 15:20:02 2012
Copyright (c) 1982, 2007, Oracle. All Rights Reserved.
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
SQL> alter database recover managed standby database disconnect from session;
Database altered.
SQL> alter database recover managed standby database cancel;
Database altered.
SQL>
SQL>
SQL> alter database open read only;
Database altered.
SQL> select count(1) from roger.test;
COUNT(1)
----------
20
SQL>
最後，我們可以看到，在主庫archivelog丟失無法同步到備庫時，可以利用增量scn的方式，來避免重建standby。
本文是參考Roger的部落格：
重建standby的情況下來將standby恢復成功的。 下面是我的測試過程，供參考！
—-主庫
SQL> SELECT database_role FROM v$database;
DATABASE_ROLE
----------------
PRIMARY
SQL> archive log list;
DATABASE log mode Archive Mode
Automatic archival Enabled
Archive destination /arch
Oldest online log SEQUENCE 382
NEXT log SEQUENCE TO archive 383
CURRENT log SEQUENCE 383
SQL> SELECT COUNT(1) FROM roger.test;
COUNT(1)
----------
30
—-備庫
SQL> SELECT database_role FROM v$database;
DATABASE_ROLE
----------------
PHYSICAL STANDBY
SQL> archive log list;
DATABASE log mode Archive Mode
Automatic archival Enabled
Archive destination /arch2
Oldest online log SEQUENCE 382
NEXT log SEQUENCE TO archive 0
CURRENT log SEQUENCE 383
SQL>
SQL> ALTER DATABASE recover managed standby DATABASE cancel;
DATABASE altered.
SQL> ALTER DATABASE OPEN READ ONLY;
DATABASE altered.
SQL> SELECT COUNT(1) FROM roger.test;
COUNT(1)
----------
30
模擬主庫丟失歸檔的情況：
—主庫
SQL> SELECT * FROM v$Log;
GROUP# THREAD# SEQUENCE# BYTES MEMBERS ARC STATUS FIRST_CHANGE# FIRST_TIME
---------- ---------- ---------- ---------- ---------- --- ---------------- ------------- ------------
1 1 383 10485760 1 NO CURRENT 740638 10-OCT-12
2 1 382 20971520 1 YES INACTIVE 740633 10-OCT-12
SQL> archive log list;
DATABASE log mode Archive Mode
Automatic archival Enabled
Archive destination /arch
Oldest online log SEQUENCE 382
NEXT log SEQUENCE TO archive 383
CURRENT log SEQUENCE 383
SQL> ALTER system SET log_archive_dest_state_2 = 'defer';
System altered.
SQL> DELETE FROM roger.test WHERE rownum < 11;
10 ROWS deleted.
SQL> commit;
Commit complete.
SQL> SELECT COUNT(*) FROM roger.test;
COUNT(*)
----------
20
SQL> ALTER system switch logfile;
System altered.
SQL> /
System altered.
SQL> /
System altered.
SQL> !
[oracle@primarydb ~]$ cd /arch
[oracle@primarydb arch]$ ls -ltr|tail -10
-rw-r----- 1 oracle dba 3456000 Oct 10 15:18 1_374_726529113.dbf
-rw-r----- 1 oracle dba 593408 Oct 10 15:23 1_375_726529113.dbf
-rw-r----- 1 oracle dba 4608 Oct 10 15:23 1_376_726529113.dbf
-rw-r----- 1 oracle dba 1024 Oct 10 15:23 1_377_726529113.dbf
-rw-r----- 1 oracle dba 68096 Oct 10 15:25 1_378_726529113.dbf
-rw-r----- 1 oracle dba 1297408 Oct 10 15:34 1_379_726529113.dbf
-rw-r----- 1 oracle dba 2048 Oct 10 18:35 1_382_726529113.dbf
-rw-r----- 1 oracle dba 166400 Oct 10 18:44 1_383_726529113.dbf
-rw-r----- 1 oracle dba 2560 Oct 10 18:44 1_384_726529113.dbf
-rw-r----- 1 oracle dba 1024 Oct 10 18:44 1_385_726529113.dbf
[oracle@primarydb arch]$ rm 1_383_726529113.dbf ---刪除archivelog
[oracle@primarydb arch]$ rm 1_384_726529113.dbf
[oracle@primarydb arch]$ exit
exit
SQL> ALTER system SET log_archive_dest_state_2 = 'enable';
System altered.
SQL>
—備庫
SQL> ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION;
DATABASE altered.
SQL>
此時alert log資訊如下：
Tue Nov 13 14:45:26 2012
MRP0: Background Managed Standby Recovery process started (test)
Managed Standby Recovery NOT USING REAL TIME Apply
parallel recovery started WITH 2 processes
Tue Nov 13 14:45:31 2012
Waiting FOR ALL non-CURRENT ORLs TO be archived...
Media Recovery Waiting FOR thread 1 SEQUENCE 383
Tue Nov 13 14:45:32 2012
Completed: ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION
Tue Nov 13 14:45:59 2012
Redo Shipping Client Connected AS PUBLIC
-- Connected User is Valid
RFS[2]: Assigned TO RFS process 9542
RFS[2]: IDENTIFIED DATABASE TYPE AS 'physical standby'
RFS[2]: Archived Log: '/arch2/1_385_726529113.dbf'
Tue Nov 13 14:46:02 2012
Fetching gap SEQUENCE IN thread 1, gap SEQUENCE 383-384
Tue Nov 13 14:47:02 2012
FAL[client]: Failed TO request gap SEQUENCE
GAP - thread 1 SEQUENCE 383-384
DBID 2024668720 branch 726529113
FAL[client]: ALL defined FAL servers have been attempted.
-------------------------------------------------------------
CHECK that the CONTROL_FILE_RECORD_KEEP_TIME initialization
parameter IS defined TO a VALUE that IS sufficiently LARGE
enough TO maintain adequate log switch information TO resolve
archivelog gaps.
如何在重建standby的情況下搞好備庫呢？mos上,其實也有文章進行描述的,就是利用rman進行增量scn的恢復,下面我來進行展示：
1)首先定位到scn
SQL> SELECT SEQUENCE#,FIRST_CHANGE#,NEXT_CHANGE# FROM v$archived_log WHERE SEQUENCE# > 382 ORDER BY 1;
SEQUENCE# FIRST_CHANGE# NEXT_CHANGE#
---------- ------------- ------------
383 740638 740911
384 740911 740915
385 740915 740917
385 740915 740917
2)根據scn，進行rman增量備份
[oracle@primarydb ~]$ cd $ORACLE_HOME/bin
[oracle@primarydb bin]$ ./rman target /
Recovery Manager: Release 10.2.0.4.0 - Production on Wed Oct 10 18:51:57 2012
Copyright (c) 1982, 2007, Oracle. All rights reserved.
connected to target database: TEST (DBID=2024668720)
RMAN> backup device type disk incremental from scn 740638 database format '/tmp/test_db_incre.bbk';
Starting backup at 10-OCT-12
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=147 devtype=DISK
channel ORA_DISK_1: starting full datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
input datafile fno=00004 name=/oracle/product/oradata/test/test01.dbf
input datafile fno=00007 name=/oracle/product/oradata/test/sysaux01.dbf
input datafile fno=00009 name=/oracle/product/oradata/test/test03.dbf
input datafile fno=00006 name=/oracle/product/oradata/test/undo02.dbf
input datafile fno=00001 name=/oracle/product/oradata/test/system01.dbf
input datafile fno=00005 name=/oracle/product/oradata/test/perfstat.dbf
input datafile fno=00003 name=/oracle/product/oradata/test/rman.dbf
input datafile fno=00002 name=/oracle/product/oradata/test/undo01.dbf
input datafile fno=00008 name=/oracle/product/oradata/test/test02.dbf
input datafile fno=00010 name=/oracle/product/oradata/test/test04.dbf
input datafile fno=00011 name=/oracle/product/oradata/test/test05.dbf
channel ORA_DISK_1: starting piece 1 at 10-OCT-12
channel ORA_DISK_1: finished piece 1 at 10-OCT-12
piece handle=/tmp/test_db_incre.bbk tag=TAG20121010T185204 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:03:26
Finished backup at 10-OCT-12
RMAN>
3) 複製增量備份到standby
[oracle@primarydb bin]$ cd /tmp
[oracle@primarydb tmp]$ scp test_db_incre.bbk 192.168.3.176:/tmp/backup
The authenticity of host '192.168.3.176 (192.168.3.176)' can't be established.
RSA key fingerprint is a4:54:6b:bf:12:34:42:73:f5:ba:5f:38:c7:28:9c:b5.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added '192.168.3.176' (RSA) to the list of known hosts.
oracle@192.168.3.176's password:
test_db_incre.bbk 100% 736KB 736.0KB/s 00:00
[oracle@primarydb tmp]$
4) standby進行recover
[oracle@standbydb bin]$ sqlplus "/as sysdba"
SQL*Plus: Release 10.2.0.4.0 - Production ON Tue Nov 13 15:03:46 2012
Copyright (c) 1982, 2007, Oracle. ALL Rights Reserved.
Connected TO:
Oracle DATABASE 10g Enterprise Edition Release 10.2.0.4.0 - Production
WITH the Partitioning, OLAP, DATA Mining AND REAL Application Testing options
SQL> ALTER DATABASE recover managed standby DATABASE cancel;
DATABASE altered.
SQL> exit
Disconnected FROM Oracle DATABASE 10g Enterprise Edition Release 10.2.0.4.0 - Production
WITH the Partitioning, OLAP, DATA Mining AND REAL Application Testing options
[oracle@standbydb ~]$ cd $ORACLE_HOME/bin
[oracle@standbydb bin]$ ./rman target /
Recovery Manager: Release 10.2.0.4.0 - Production ON Tue Nov 13 14:58:25 2012
Copyright (c) 1982, 2007, Oracle. ALL rights reserved.
connected TO target DATABASE: TEST (DBID=2024668720, NOT OPEN)
RMAN> catalog backuppiece '/tmp/backup/test_db_incre.bbk';
USING target DATABASE control file instead OF recovery catalog
cataloged backuppiece
backup piece handle=/tmp/backup/test_db_incre.bbk recid=49 stamp=799253957
RMAN> recover DATABASE noredo;
Starting recover at 13-NOV-12
USING target DATABASE control file instead OF recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=97 devtype=DISK
channel ORA_DISK_1: starting incremental datafile backupset restore
channel ORA_DISK_1: specifying datafile(s) TO restore FROM backup SET
destination FOR restore OF datafile 00001: /oracle/product/oradata/test/system01.dbf
destination FOR restore OF datafile 00002: /oracle/product/oradata/test/undo01.dbf
destination FOR restore OF datafile 00003: /oracle/product/oradata/test/rman.dbf
destination FOR restore OF datafile 00004: /oracle/product/oradata/test/test01.dbf
destination FOR restore OF datafile 00005: /oracle/product/oradata/test/perfstat.dbf
destination FOR restore OF datafile 00006: /oracle/product/oradata/test/undo02.dbf
destination FOR restore OF datafile 00007: /oracle/product/oradata/test/sysaux01.dbf
destination FOR restore OF datafile 00008: /oracle/product/oradata/test/test02.dbf
destination FOR restore OF datafile 00009: /oracle/product/oradata/test/test03.dbf
destination FOR restore OF datafile 00010: /oracle/product/oradata/test/test04.dbf
destination FOR restore OF datafile 00011: /oracle/product/oradata/test/test05.dbf
channel ORA_DISK_1: reading FROM backup piece /tmp/backup/test_db_incre.bbk
channel ORA_DISK_1: restored backup piece 1
piece handle=/tmp/backup/test_db_incre.bbk tag=TAG20121010T185204
channel ORA_DISK_1: restore complete, elapsed TIME: 00:00:01
Finished recover at 13-NOV-12
RMAN>
如下資訊是recover時，alert log的記錄資訊：
Tue Nov 13 15:04:05 2012
Completed: ALTER DATABASE recover managed standby DATABASE cancel
Tue Nov 13 15:04:55 2012
Incremental restore complete OF datafile 2 /oracle/product/oradata/test/undo01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 8 /oracle/product/oradata/test/test02.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 10 /oracle/product/oradata/test/test04.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 11 /oracle/product/oradata/test/test05.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 3 /oracle/product/oradata/test/rman.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 5 /oracle/product/oradata/test/perfstat.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 1 /oracle/product/oradata/test/system01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 6 /oracle/product/oradata/test/undo02.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 4 /oracle/product/oradata/test/test01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 7 /oracle/product/oradata/test/sysaux01.dbf
checkpoint IS 741134
Incremental restore complete OF datafile 9 /oracle/product/oradata/test/test03.dbf
checkpoint IS 741134
5) 開啟standby同步，檢查是否ok。
SQL> ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION;
DATABASE altered.
SQL> SELECT database_role,open_mode FROM v$database;
DATABASE_ROLE OPEN_MODE
---------------- ----------
PHYSICAL STANDBY MOUNTED
--主庫切換日誌
SQL> SELECT database_role,open_mode FROM v$database;
DATABASE_ROLE OPEN_MODE
---------------- ----------
PRIMARY READ WRITE
SQL>
SQL> ALTER system switch logfile;
System altered.
SQL> SELECT MAX(al.SEQUENCE#) "Last Seq Recieved",
2 MAX(lh.SEQUENCE#) "Last Seq Applied"
3 FROM v$archived_log al, v$log_history lh;
LAST Seq Recieved LAST Seq Applied
----------------- ----------------
386 386
SQL>
--備庫
SQL> SELECT MAX(al.SEQUENCE#) "Last Seq Recieved",
2 MAX(lh.SEQUENCE#) "Last Seq Applied"
3 FROM v$archived_log al, v$log_history lh;
LAST Seq Recieved LAST Seq Applied
----------------- ----------------
386 382
SQL>
SQL> SELECT * FROM v$archive_gap;
THREAD# LOW_SEQUENCE# HIGH_SEQUENCE#
---------- ------------- --------------
1 383 384
備庫archivelog：
[oracle@standbydb arch2]$ ls -ltr |tail -5
-rw-r----- 1 oracle dba 2048 Nov 13 14:33 1_381_726529113.dbf
-rw-r----- 1 oracle dba 2048 Nov 13 14:34 1_382_726529113.dbf
-rw-r----- 1 oracle dba 1024 Nov 13 14:45 1_385_726529113.dbf
-rw-r----- 1 oracle dba 1262080 Nov 13 15:06 1_386_726529113.dbf
備庫alert log此時的資訊：
Tue Nov 13 15:06:17 2012
ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION
Tue Nov 13 15:06:17 2012
Attempt TO START background Managed Standby Recovery process (test)
MRP0 started WITH pid=19, OS id=13497
Tue Nov 13 15:06:17 2012
MRP0: Background Managed Standby Recovery process started (test)
Managed Standby Recovery NOT USING REAL TIME Apply
parallel recovery started WITH 2 processes
Tue Nov 13 15:06:22 2012
Waiting FOR ALL non-CURRENT ORLs TO be archived...
Media Recovery Waiting FOR thread 1 SEQUENCE 383
Fetching gap SEQUENCE IN thread 1, gap SEQUENCE 383-384
Tue Nov 13 15:06:23 2012
Completed: ALTER DATABASE recover managed standby DATABASE disconnect FROM SESSION
Tue Nov 13 15:06:54 2012
Redo Shipping Client Connected AS PUBLIC
-- Connected User is Valid
RFS[3]: Assigned TO RFS process 13596
RFS[3]: IDENTIFIED DATABASE TYPE AS 'physical standby'
PRIMARY DATABASE IS IN MAXIMUM PERFORMANCE mode
PRIMARY DATABASE IS IN MAXIMUM PERFORMANCE mode
RFS[3]: Successfully opened standby log 4: '/oracle/product/oradata/test/redo04.log'
Tue Nov 13 15:06:54 2012
RFS[2]: Successfully opened standby log 3: '/oracle/product/oradata/test/redo03.log'
Tue Nov 13 15:07:23 2012
FAL[client]: Failed TO request gap SEQUENCE
GAP - thread 1 SEQUENCE 383-384
DBID 2024668720 branch 726529113
FAL[client]: ALL defined FAL servers have been attempted.
-------------------------------------------------------------
CHECK that the CONTROL_FILE_RECORD_KEEP_TIME initialization
parameter IS defined TO a VALUE that IS sufficiently LARGE
enough TO maintain adequate log switch information TO resolve
archivelog gaps.
我們可以看到，雖然備庫，仍然在提示383,384是gap ，但是實際上已經是同步的了。
6)最後來驗證下資料
SQL> ALTER DATABASE recover managed standby DATABASE cancel;
DATABASE altered.
SQL> SELECT file#,checkpoint_change# FROM v$datafile ORDER BY 1;
FILE# CHECKPOINT_CHANGE#
---------- ------------------
1 740638
2 740638
3 740638
4 740638
5 740638
6 740638
7 740638
8 740638
9 740638
10 740638
11 740638
11 ROWS selected.
SQL> SELECT file#,checkpoint_change# FROM v$datafile_header ORDER BY 1;
FILE# CHECKPOINT_CHANGE#
---------- ------------------
1 741134
2 741134
3 741134
4 741134
5 741134
6 741134
7 741134
8 741134
9 741134
10 741134
11 741134
11 ROWS selected.
SQL> ALTER DATABASE OPEN READ ONLY;
ALTER DATABASE OPEN READ ONLY
*
ERROR at line 1:
ORA-16004: backup DATABASE requires recovery
ORA-01152: file 1 was NOT restored FROM a sufficiently OLD backup
ORA-01110: DATA file 1: '/oracle/product/oradata/test/system01.dbf'
重建備庫的standby controlfile：
SQL> ALTER DATABASE CREATE standby controlfile AS '/tmp/standby.ctl';
DATABASE altered.
SQL> exit
Disconnected FROM Oracle DATABASE 10g Enterprise Edition Release 10.2.0.4.0 - Production
WITH the Partitioning, OLAP, DATA Mining AND REAL Application Testing options
[oracle@primarydb tmp]$ scp standby.ctl 192.168.3.176:/tmp/backup
oracle@192.168.3.176's password:
standby.ctl 100% 3928KB 3.8MB/s 00:00
[oracle@primarydb tmp]$
[oracle@standbydb bin]$ ./rman target /
Recovery Manager: Release 10.2.0.4.0 - Production on Tue Nov 13 15:19:09 2012
Copyright (c) 1982, 2007, Oracle. All rights reserved.
connected to target database (not started)
RMAN> startup nomount
Oracle instance started
Total System Global Area 264241152 bytes
Fixed Size 1266944 bytes
Variable Size 209718016 bytes
Database Buffers 50331648 bytes
Redo Buffers 2924544 bytes
RMAN> restore controlfile from '/tmp/backup/standby.ctl';
Starting restore at 13-NOV-12
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=101 devtype=DISK
channel ORA_DISK_1: copied control file copy
output filename=/oracle/product/oradata/test/control01.ctl
Finished restore at 13-NOV-12
RMAN> startup mount
database is already started
database mounted
released channel: ORA_DISK_1
RMAN> exit
Recovery Manager complete.
[oracle@standbydb bin]$ sqlplus "/as sysdba"
SQL*Plus: Release 10.2.0.4.0 - Production on Tue Nov 13 15:20:02 2012
Copyright (c) 1982, 2007, Oracle. All Rights Reserved.
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
SQL> alter database recover managed standby database disconnect from session;
Database altered.
SQL> alter database recover managed standby database cancel;
Database altered.
SQL>
SQL>
SQL> alter database open read only;
Database altered.
SQL> select count(1) from roger.test;
COUNT(1)
----------
20
SQL>
最後，我們可以看到，在主庫archivelog丟失無法同步到備庫時，可以利用增量scn的方式，來避免重建standby。
本文是參考Roger的部落格：
來自 “ ITPUB部落格 ” ，連結：http://blog.itpub.net/90618/viewspace-1978849/，如需轉載，請註明出處，否則將追究法律責任。
相關文章
- [Dataguard]主庫歸檔丟失，備庫不需重建實驗
- dataguard 歸檔丟失,備庫基於SCN恢復
- 【DATAGUARD】物理dg在主庫丟失歸檔檔案的情況下的恢復(七)
- 有整庫備份的情況rman恢復全庫
- OracleDG備庫恢復–gapOracle
- oracle歸檔日誌丟失後的資料庫恢復Oracle資料庫
- 丟失已歸檔日誌檔案下恢復資料庫資料庫
- 利用增量備份恢復gap歸檔丟失DG
- 【備份恢復】恢復 丟失已歸檔重做日誌檔案
- 用物理備庫的檔案來恢復主庫
- RMAN資料庫恢復 之歸檔模式有(無)備份-丟失資料檔案的恢復資料庫模式
- 【故障處理】DG環境主庫丟失歸檔情況下資料檔案的恢復
- rman全庫備份備份歸檔日誌檔案
- 備庫跳歸檔恢復的有趣案例
- 【DATAGUARD】手工恢復備庫日誌中斷
- Oracle RMAN 不完全恢復（只有資料檔案備份，丟失歸檔日誌備份）Oracle
- REDO檔案丟失的恢復__沒有任何備份的情況
- 利用增量備份恢復因歸檔丟失造成的DG gap
- 資料庫在沒有備份的情況下的資料檔案損壞的恢復資料庫
- 沒備份，歸檔日誌存在，丟失資料檔案的恢復
- dataguard 主備庫出現gap
- 沒有自動備份的情況下控制檔案全部丟失的恢復
- rman 非歸檔模式下停庫備份與恢復模式
- 測試在丟失歸檔日誌的情況下，跳過部分歸檔日誌進行資料恢復資料恢復
- undo表空間檔案丟失恢復（2）--無備份有redo的情況下恢復
- 使用冷備份與冷備份後的資料庫歸檔日誌檔案進行資料庫不完整恢復資料庫
- DG 主庫丟失歸檔
- RMAN備份整庫和歸檔日誌的方法
- 第5章：從開啟的資料庫備份與恢復之備份歸檔日誌檔案資料庫
- 恢復之利用備份在所有控制檔案丟失情況下恢復（四）
- 恢復之利用備份在所有控制檔案丟失情況下恢復（三）
- 恢復之利用備份在所有控制檔案丟失情況下恢復（二）
- 恢復之利用備份在所有控制檔案丟失情況下恢復（一）
- 在歸檔模式下丟失日誌檔案的恢復模式
- RAC環境備份歸檔日誌和RMAN恢復啟動資料庫資料庫
- 【備份恢復】在 ARCHIVELOG 模式下執行資料庫還原和恢復操作（源庫備份源庫恢復）Hive模式資料庫
- rman 非歸檔模式下open庫備份與mount恢復模式
- 備份＆恢復之八：RMAN備份歸檔模式下損壞（丟失）多個資料檔案，進行整個資料庫的恢復模式資料庫