Oracle 12C Active Data Guard Far Sync 配置

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：Oracle 12C Active Data Guard Far Sync 配置

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

Active Data Guard Far Sync是Oracle 12c的新功能（也称为Far Sync Standby），Far Sync功能的实现是通过在距离主库(Primary Database)相对较近的地点配置Far Sync实例，主库(Primary Database) 同步(synchronous)传输redo到Far Sync实例，然后Far Sync实例再将redo异步(asynchronous)传输到终端备库(Standby Database)。这样既可以保证零数据丢失又可以降低主库压力。Far Sync实例只有密码文件，init参数文件和控制文件，而没有数据文件。 如果redo 传输采用Maximum Availability模式，我们可以在距离生产中心(Primary Database)相对较近的地点配置Far Sync实例，主库(Primary Database)同步(synchronous)传输redo到Far Sync实例，保证零数据丢失（zero data loss），同时主库和Far Sync距离较近，网络延时很小，因此对主库性能影响很小。然后Far Sync实例再将redo异步(asynchronous)发送到终端备库(Standby Database)。 如果redo 传输采用Maximum Performance模式，我们可以在距离生产中心(Primary Database)相对较近的地点配置Far Sync实例，主库(Primary Database) 异步传输redo到Far Sync实例，然后Far Sync实例再负责传输redo到其他多个终端备库(Standby Database)。这样可以减少主库向多个终端备库(Standby Database)传输redo的压力（offload）。 Far Sync配置对于Data Guard 角色转换(role transitions)是透明的，即switchover/failover命令方式与12c之前相同。
1.主库配置fra

SQL> alter system set DB_RECOVERY_FILE_DEST_SIZE=200G;
System altered.
SQL> alter system set db_recovery_file_dest='/u01/app/oracle/fast_recovery_area';
System altered.

2.启用归档模式和强制日志

SQL> shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup mount
ORACLE instance started.
Total System Global Area 2516582400 bytes
Fixed Size                  4585912 bytes
Variable Size             671090248 bytes
Database Buffers         1828716544 bytes
Redo Buffers               12189696 bytes
Database mounted.
SQL> alter database archivelog;
Database altered.
SQL> alter database force logging;
Database altered.
SQL> alter database open;
Database altered.

3.主机规划

192.168.137.121  ora1221   --->主库(ORCL12C)
192.168.137.122  ora1222   --->Far Sync实例(ORCL12CFS)
192.168.137.123  ora1223   --->备库(ORCL12CDG)

4.tns配置

ORCL12C =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = ora1221)(PORT = 1521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = orcl12c)
    )
  )
ORCL12CDG =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = ora1223)(PORT = 1521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = orcl12c)
    )
  )
ORCL12CFS =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = ora1222)(PORT = 1521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = orcl12c)
    )
  )

5.参数文件配置

--主库
db_unique_name='orcl12c'
service_names='orcl12c'
log_archive_dest_1='LOCATION=USE_DB_RECOVERY_FILE_DEST valid_for=(all_logfiles,all_roles) db_unique_name=orcl12c'
log_archive_dest_2='service=orcl12cfs lgwr  sync  AFFIRM MAX_FAILURE=1 ALTERNATE=LOG_ARCHIVE_DEST_3 valid_for=(online_logfiles,primary_role) db_unique_name=orcl12cfs'
LOG_ARCHIVE_DEST_3='SERVICE=orcl12cdg  lgwr ASYNC ALTERNATE=LOG_ARCHIVE_DEST_2 VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=orcl12cdg'
LOG_ARCHIVE_DEST_STATE_3=ALTERNATE
log_archive_config='dg_config=(orcl12c,orcl12cdg,orcl12cfs)'
standby_file_management=auto
db_file_name_convert='/u01/app/oracle/oradata/orcl12c/','/u01/app/oracle/oradata/orcl12c/'
log_file_name_convert='/u01/app/oracle/oradata/orcl12c/','/u01/app/oracle/oradata/orcl12c/'
fal_server='orcl12cdg','orcl12cfs'
--Far Sync实例
db_unique_name='orcl12cfs'
service_names='orcl12c'
log_archive_dest_1='LOCATION=USE_DB_RECOVERY_FILE_DEST valid_for=(all_logfiles,all_roles) db_unique_name=orcl12cfs'
log_archive_dest_2='service=orcl12cdg lgwr  async valid_for=(STANDBY_LOGFILES,STANDBY_ROLE) db_unique_name=orcl12cdg'
log_archive_config='dg_config=(orcl12c,orcl12cdg,orcl12cfs)'
standby_file_management=manual
fal_server='orcl12c'
--备库
db_unique_name='orcl12cdg'
service_names='orcl12c'
log_archive_dest_1='LOCATION=USE_DB_RECOVERY_FILE_DEST valid_for=(all_logfiles,all_roles) db_unique_name=orcl12cdg'
log_archive_dest_2='service=orcl12c lgwr  async   valid_for=(online_logfiles,primary_role) db_unique_name=orcl12c'
log_archive_config='dg_config=(orcl12c,orcl12cdg,orcl12cfs)'
standby_file_management=auto
db_file_name_convert='/u01/app/oracle/oradata/orcl12c/','/u01/app/oracle/oradata/orcl12c/'
log_file_name_convert='/u01/app/oracle/oradata/orcl12c/','/u01/app/oracle/oradata/orcl12c/'
fal_server='orcl12c','orcl12cfs'

6.密码文件
拷贝主库的密码文件到Far Sync实例和备库$ORACLE_HOME/dbs下(如果是win在%ORACLE_HOME%/database中)

7.创建Far Sync实例
创建和参数文件匹配的fra,adump目录

--主库
ALTER DATABASE CREATE FAR SYNC INSTANCE CONTROLFILE AS '/tmp/controlfs01.ctl';
拷贝到Far Sync实例对应的控制文件位置
--Far Sync实例
startup pfile='/tmp/pfile' mount
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 10 '/u01/app/oracle/oradata/orcl12c/std_redo10.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 11 '/u01/app/oracle/oradata/orcl12c/std_redo11.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 12 '/u01/app/oracle/oradata/orcl12c/std_redo12.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 13 '/u01/app/oracle/oradata/orcl12c/std_redo13.log' size 50m reuse;
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT  LOGFILE DISCONNECT FROM SESSION;

8.创建备库
创建和参数文件匹配的fra,adump目录

--主库备份（传输至备库）
backup database format '/tmp/ora12c_%U';
--主库创建standby ctl（传输至备库和备库参数文件路径一致）
alter database create standby controlfile as '/tmp/controlst.ctl';
--启动备库至mount
startup pfile='/tmp/pfile' mount;
--备库注册备份
catalog start with '/tmp/xifenfei/';
--备库还原数据文件并恢复
restore database;
recover database;
--增加standby redo
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 10 '/u01/app/oracle/oradata/orcl12c/std_redo10.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 11 '/u01/app/oracle/oradata/orcl12c/std_redo11.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 12 '/u01/app/oracle/oradata/orcl12c/std_redo12.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 13 '/u01/app/oracle/oradata/orcl12c/std_redo13.log' size 50m reuse;
--备库开启mrp进程
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT  LOGFILE DISCONNECT FROM SESSION;
--主库增加standby redo
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 10 '/u01/app/oracle/oradata/orcl12c/std_redo10.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 11 '/u01/app/oracle/oradata/orcl12c/std_redo11.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 12 '/u01/app/oracle/oradata/orcl12c/std_redo12.log' size 50m reuse;
ALTER DATABASE ADD STANDBY LOGFILE thread 1  GROUP 13 '/u01/app/oracle/oradata/orcl12c/std_redo13.log' size 50m reuse;

9.配置结果

SQL> select * from V$DATAGUARD_CONFIG;
DB_UNIQUE_NAME                 PARENT_DBUN                    DEST_ROLE         CURRENT_SCN     CON_ID
------------------------------ ------------------------------ ----------------- ----------- ----------
orcl12c                        NONE                           PRIMARY DATABASE      1950551          0
orcl12cfs                      orcl12c                        FAR SYNC INSTANCE     1950390          0
orcl12cdg                      orcl12cfs                      PHYSICAL STANDBY      1950390          0
SQL> select group#, status, thread#, sequence#, first_change#, next_change# from v$standby_log;
    GROUP# STATUS        THREAD#  SEQUENCE# FIRST_CHANGE# NEXT_CHANGE#
---------- ---------- ---------- ---------- ------------- ------------
        10 ACTIVE              1         27       1863140
        11 UNASSIGNED          1          0
        12 UNASSIGNED          1          0
        13 UNASSIGNED          1          0
SQL> select PROTECTION_MODE,PROTECTION_LEVEL from v$database;
PROTECTION_MODE      PROTECTION_LEVEL
-------------------- --------------------
MAXIMUM PERFORMANCE  MAXIMUM PERFORMANCE

10.调整保护级别
从MAXIMUM PERFORMANCE调整为MAXIMUM AVAILABILITY

--主库
SQL> startup mount;
ORACLE instance started.
Total System Global Area 2516582400 bytes
Fixed Size                  4585912 bytes
Variable Size             671090248 bytes
Database Buffers         1828716544 bytes
Redo Buffers               12189696 bytes
Database mounted.
SQL> alter database set standby database to maximize availability;
Database altered.
SQL>  select PROTECTION_MODE,PROTECTION_LEVEL from v$database;
PROTECTION_MODE      PROTECTION_LEVEL
-------------------- --------------------
MAXIMUM AVAILABILITY UNPROTECTED
SQL> alter database open;
Database altered.
SQL>  select PROTECTION_MODE,PROTECTION_LEVEL from v$database;
PROTECTION_MODE      PROTECTION_LEVEL
-------------------- --------------------
MAXIMUM AVAILABILITY MAXIMUM AVAILABILITY
--Far Sync实例
SQL>  select PROTECTION_MODE,PROTECTION_LEVEL from v$database;
PROTECTION_MODE      PROTECTION_LEVEL
-------------------- --------------------
MAXIMUM AVAILABILITY MAXIMUM AVAILABILITY
--备库
SQL> select PROTECTION_MODE,PROTECTION_LEVEL from v$database;
PROTECTION_MODE      PROTECTION_LEVEL
-------------------- --------------------
MAXIMUM AVAILABILITY RESYNCHRONIZATION

分享一例由于主库逻辑坏块导致dataguard容灾失效

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：分享一例由于主库逻辑坏块导致dataguard容灾失效

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

有朋友数据库配置了物理dataguard，但是由于触发了某种Oracle bug/或者其他原因导致block出现大量逻辑坏块，结果是主备库(主库启动后就crash,备库failover后也启动后就crash)都无法正常工作,请求给予技术支持。
数据库配置了物理dataguard，但是主库在进行了某些操作之后，主库直接crash.重启主库发现，数据库启动之后，稍后数据库继续CRASH

Wed Jul 08 16:32:22 2015
Thread 1 advanced to log sequence 401531 (LGWR switch)
  Current log# 6 seq# 401531 mem# 0: /opt/oracle/database/fast_recovery_area/xifenfei/onlinelog/o1_mf_6_b9p7qhrm_.log
Archived Log entry 605194 added for thread 1 sequence 401530 ID 0xfbe74bc9 dest 1:
ARC3: Standby redo logfile selected for thread 1 sequence 401530 for destination LOG_ARCHIVE_DEST_2
******************************************************************
LGWR: Setting 'active' archival for destination LOG_ARCHIVE_DEST_2
******************************************************************
Wed Jul 08 16:33:02 2015
Errors in file /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/trace/xifenfei_p032_413611.trc  (incident=201910):
ORA-00600: internal error code, arguments: [17114], [0x7FB03BB6DBD0], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7FB03BB6DBF8], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/incident/incdir_201910/xifenfei_p032_413611_i201910.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Errors in file /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/trace/xifenfei_p032_413611.trc  (incident=201911):
ORA-00600: internal error code, arguments: [kghfrempty:ds], [0x7FB03BB6DBE8], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17114], [0x7FB03BB6DBD0], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7FB03BB6DBF8], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/incident/incdir_201911/xifenfei_p032_413611_i201911.trc
Wed Jul 08 16:34:23 2015
SMON: slave died unexpectedly, downgrading to serial recovery
Errors in file /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/trace/xifenfei_smon_413207.trc  (incident=201547):
ORA-00600: internal error code, arguments: [17182], [0x7F24D680D7A0], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/incident/incdir_201547/xifenfei_smon_413207_i201547.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Block recovery from logseq 401530, block 119 to scn 73226510040
Recovery of Online Redo Log: Thread 1 Group 5 Seq 401530 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/xifenfei/onlinelog/o1_mf_5_b9p7qhl0_.log
Recovery of Online Redo Log: Thread 1 Group 6 Seq 401531 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/xifenfei/onlinelog/o1_mf_6_b9p7qhrm_.log
Block recovery completed at rba 401531.1882.16, scn 17.212066009
ORACLE Instance xifenfei (pid = 16) - Error 600 encountered while recovering transaction (13, 14) on object 135520.
Errors in file /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/trace/xifenfei_smon_413207.trc:
ORA-00600: internal error code, arguments: [17182], [0x7F24D680D7A0], [], [], [], [], [], [], [], [], [], []
Wed Jul 08 16:34:24 2015
Dumping diagnostic data in directory=[cdmp_20150708163424], requested by (instance=1, osid=413207 (SMON)), summary=[incident=201547].
Errors in file /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/trace/xifenfei_smon_413207.trc  (incident=201548):
ORA-00600: internal error code, arguments: [KSMFPG2], [0x7F24D680D000], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7F24D680D7A0], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/incident/incdir_201548/xifenfei_smon_413207_i201548.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Fatal internal error happened while SMON was doing active transaction recovery.
Errors in file /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/trace/xifenfei_smon_413207.trc:
ORA-00600: internal error code, arguments: [KSMFPG2], [0x7F24D680D000], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7F24D680D7A0], [], [], [], [], [], [], [], [], [], []
SMON (ospid: 413207): terminating the instance due to error 474
System state dump requested by (instance=1, osid=413207 (SMON)), summary=[abnormal instance termination].
System State dumped to trace file /opt/oracle/database/diag/rdbms/xifenfei/xifenfei/trace/xifenfei_diag_413167_20150708163426.trc
Dumping diagnostic data in directory=[cdmp_20150708163426], requested by (instance=1, osid=413207 (SMON)), summary=[abnormal instance termination].
Instance terminated by SMON, pid = 413207

由于主库不能正常open，备库直接failover方式激活

Wed Jul 08 17:56:41 2015
alter database recover managed standby database finish
Terminal Recovery: request posted (xffdb)
Wed Jul 08 17:56:45 2015
Begin: Standby Redo Logfile archival
End: Standby Redo Logfile archival
Terminal Recovery timestamp is '07/08/2015 17:56:45'
Terminal Recovery: applying standby redo logs.
Terminal Recovery: thread 1 seq# 401533 redo required
Terminal Recovery:
Recovery of Online Redo Log: Thread 1 Group 7 Seq 401533 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_7_bb7x31lk_.log
Identified End-Of-Redo (failover) for thread 1 sequence 401533 at SCN 0xffff.ffffffff
Incomplete Recovery applied until change 73226530800 time 07/08/2015 16:56:40
Terminal Recovery: successful completion
Wed Jul 08 17:56:45 2015
ARCH: Archival stopped, error occurred. Will continue retrying
Forcing ARSCN to IRSCN for TR 17:212086768
ORACLE Instance xffdb - Archival Error
Attempt to set limbo arscn 17:212086768 irscn 17:212086768 ORA-16014: log 7 sequence# 401533 not archived, no available destinations
ORA-00312: online log 7 thread 1: '/opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_7_bb7x31lk_.log'
Resetting standby activation ID 4226239433 (0xfbe74bc9)
Wed Jul 08 17:56:45 2015
MRP0: Media Recovery Complete (xffdb)
MRP0: Background Media Recovery process shutdown (xffdb)
Terminal Recovery: completion detected (xffdb)
Completed: alter database recover managed standby database finish
Wed Jul 08 17:56:58 2015
alter database commit to switchover to primary
ALTER DATABASE SWITCHOVER TO PRIMARY (xffdb)
Maximum wait for role transition is 15 minutes.
Backup controlfile written to trace file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_ora_485094.trc
Standby terminal recovery start SCN: 73226530482
RESETLOGS after incomplete recovery UNTIL CHANGE 73226530800
Online logfile pre-clearing operation disabled by switchover
Online log /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_1_bb7x30gw_.log: Thread 1 Group 1 was previously cleared
Online log /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_2_bb7x30js_.log: Thread 1 Group 2 was previously cleared
Online log /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_3_bb7x310q_.log: Thread 1 Group 3 was previously cleared
Online log /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_4_bb7x312r_.log: Thread 1 Group 4 was previously cleared
Online log /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_5_bb7x317f_.log: Thread 1 Group 5 was previously cleared
Online log /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_6_bb7x31cz_.log: Thread 1 Group 6 was previously cleared
Standby became primary SCN: 73226530481
Wed Jul 08 17:56:58 2015
Setting recovery target incarnation to 3
Switchover: Complete - Database mounted as primary
Completed: alter database commit to switchover to primary

数据库激活成功后，重启激活之后数据库发现和主库出现类似情况

Wed Jul 08 17:57:25 2015
ALTER DATABASE   MOUNT
Successful mount of redo thread 1, with mount id 4243462021
Database mounted in Exclusive Mode
Lost write protection disabled
Completed: ALTER DATABASE   MOUNT
Wed Jul 08 17:57:30 2015
ALTER DATABASE OPEN
Assigning activation ID 4243462021 (0xfcee1785)
LGWR: STARTING ARCH PROCESSES
Wed Jul 08 17:57:30 2015
ARC0 started with pid=23, OS id=485230
ARC0: Archival started
LGWR: STARTING ARCH PROCESSES COMPLETE
ARC0: STARTING ARCH PROCESSES
Thread 1 advanced to log sequence 2 (thread open)
Wed Jul 08 17:57:31 2015
ARC1 started with pid=24, OS id=485236
Wed Jul 08 17:57:31 2015
ARC2 started with pid=25, OS id=485240
Wed Jul 08 17:57:31 2015
ARC3 started with pid=26, OS id=485244
ARC1: Archival started
ARC2: Archival started
ARC1: Becoming the 'no FAL' ARCH
ARC1: Becoming the 'no SRL' ARCH
ARC2: Becoming the heartbeat ARCH
Thread 1 opened at log sequence 2
  Current log# 2 seq# 2 mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_2_bb7x30js_.log
Successful open of redo thread 1
MTTR advisory is disabled because FAST_START_MTTR_TARGET is not set
SMON: enabling cache recovery
Wed Jul 08 17:57:31 2015
NSA2 started with pid=27, OS id=485248
[485226] Successfully onlined Undo Tablespace 2.
Undo initialization finished serial:0 start:3453451748 end:3453452018 diff:270 (2 seconds)
Dictionary check beginning
Dictionary check complete
Verifying file header compatibility for 11g tablespace encryption..
Verifying 11g file header compatibility for tablespace encryption completed
SMON: enabling tx recovery
Database Characterset is AL32UTF8
No Resource Manager plan active
replication_dependency_tracking turned off (no async multimaster replication found)
Wed Jul 08 17:57:32 2015
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_p032_485380.trc  (incident=64481):
ORA-00600: internal error code, arguments: [17182], [0x7FE96B50DBF8], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xffdb/xffdb/incident/incdir_64481/xffdb_p032_485380_i64481.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Starting background process QMNC
Wed Jul 08 17:57:32 2015
QMNC started with pid=92, OS id=485512
LOGSTDBY: Validating controlfile with logical metadata
LOGSTDBY: Validation complete
Dumping diagnostic data in directory=[cdmp_20150708175733], requested by (instance=1, osid=485380 (P032)), summary=[incident=64481].
Thread 1 advanced to log sequence 3 (LGWR switch)
  Current log# 3 seq# 3 mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_3_bb7x310q_.log
ARC3: STARTING ARCH PROCESSES
Wed Jul 08 17:57:34 2015
ARC4 started with pid=93, OS id=485516
Wed Jul 08 17:57:35 2015
db_recovery_file_dest_size of 204800 MB is 0.41% used. This is a
user-specified limit on the amount of space that will be used by this
database for recovery-related files, and does not reflect the amount of
space available in the underlying filesystem or ASM diskgroup.
ARC4: Archival started
ARC3: STARTING ARCH PROCESSES COMPLETE
krsk_srl_archive_int: Enabling archival of deferred physical standby SRLs
Archived Log entry 273963 added for thread 1 sequence 2 ID 0xfcee1785 dest 1:
Archived Log entry 273964 added for thread 1 sequence 401533 ID 0xfbe74bc9 dest 1:
Completed: ALTER DATABASE OPEN
Wed Jul 08 17:57:36 2015
Starting background process CJQ0
Wed Jul 08 17:57:36 2015
CJQ0 started with pid=95, OS id=485554
Shutting down archive processes
ARCH shutting down
ARC4: Archival stopped
Wed Jul 08 17:57:41 2015
Thread 1 advanced to log sequence 4 (LGWR switch)
  Current log# 4 seq# 4 mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_4_bb7x312r_.log
Wed Jul 08 17:57:41 2015
Archived Log entry 273965 added for thread 1 sequence 3 ID 0xfcee1785 dest 1:
Wed Jul 08 17:58:30 2015
Sweep [inc][64481]: completed
Sweep [inc2][64481]: completed
Wed Jul 08 17:58:31 2015
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_p032_485380.trc  (incident=64482):
ORA-00600: internal error code, arguments: [17114], [0x7FE96B50DBD0], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7FE96B50DBF8], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xffdb/xffdb/incident/incdir_64482/xffdb_p032_485380_i64482.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Wed Jul 08 17:58:32 2015
Dumping diagnostic data in directory=[cdmp_20150708175832], requested by (instance=1, osid=485380 (P032)), summary=[incident=64482].
Wed Jul 08 17:58:36 2015
Thread 1 advanced to log sequence 5 (LGWR switch)
  Current log# 5 seq# 5 mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_5_bb7x317f_.log
Wed Jul 08 17:59:02 2015
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_p032_485380.trc  (incident=64483):
ORA-00600: internal error code, arguments: [17114], [0x7FE96B50DBD0], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17114], [0x7FE96B50DBD0], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7FE96B50DBF8], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xffdb/xffdb/incident/incdir_64483/xffdb_p032_485380_i64483.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Wed Jul 08 17:59:04 2015
Dumping diagnostic data in directory=[cdmp_20150708175904], requested by (instance=1, osid=485380 (P032)), summary=[incident=64483].
Wed Jul 08 17:59:29 2015
Sweep [inc][64483]: completed
Sweep [inc][64482]: completed
Sweep [inc2][64483]: completed
Sweep [inc2][64482]: completed
Wed Jul 08 17:59:30 2015
Block recovery from logseq 2, block 104 to scn 73226531646
Recovery of Online Redo Log: Thread 1 Group 2 Seq 2 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_2_bb7x30js_.log
Recovery of Online Redo Log: Thread 1 Group 3 Seq 3 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_3_bb7x310q_.log
Recovery of Online Redo Log: Thread 1 Group 4 Seq 4 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_4_bb7x312r_.log
Recovery of Online Redo Log: Thread 1 Group 5 Seq 5 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_5_bb7x317f_.log
Block recovery stopped at EOT rba 5.765.16
Block recovery completed at rba 5.765.16, scn 17.212087614
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_p032_485380.trc  (incident=64484):
ORA-00600: internal error code, arguments: [KSMFPG2], [0x7FE96B50D000], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17114], [0x7FE96B50DBD0], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17114], [0x7FE96B50DBD0], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7FE96B50DBF8], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xffdb/xffdb/incident/incdir_64484/xffdb_p032_485380_i64484.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Dumping diagnostic data in directory=[cdmp_20150708175934], requested by (instance=1, osid=485380 (P032)), summary=[incident=64487].
Wed Jul 08 17:59:36 2015
SMON: slave died unexpectedly, downgrading to serial recovery
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_smon_485190.trc  (incident=64129):
ORA-00600: internal error code, arguments: [17182], [0x7F5AED10D7A0], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xffdb/xffdb/incident/incdir_64129/xffdb_smon_485190_i64129.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Block recovery from logseq 2, block 104 to scn 73226531646
Recovery of Online Redo Log: Thread 1 Group 2 Seq 2 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_2_bb7x30js_.log
Recovery of Online Redo Log: Thread 1 Group 3 Seq 3 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_3_bb7x310q_.log
Recovery of Online Redo Log: Thread 1 Group 4 Seq 4 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_4_bb7x312r_.log
Recovery of Online Redo Log: Thread 1 Group 5 Seq 5 Reading mem 0
  Mem# 0: /opt/oracle/database/fast_recovery_area/DBHLDB/onlinelog/o1_mf_5_bb7x317f_.log
Block recovery completed at rba 5.765.16, scn 17.212087615
ORACLE Instance xffdb (pid = 16) - Error 600 encountered while recovering transaction (13, 14) on object 135520.
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_smon_485190.trc:
ORA-00600: internal error code, arguments: [17182], [0x7F5AED10D7A0], [], [], [], [], [], [], [], [], [], []
Dumping diagnostic data in directory=[cdmp_20150708175937], requested by (instance=1, osid=485190 (SMON)), summary=[incident=64129].
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_smon_485190.trc  (incident=64130):
ORA-00600: internal error code, arguments: [KSMFPG2], [0x7F5AED10D000], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7F5AED10D7A0], [], [], [], [], [], [], [], [], [], []
Incident details in: /opt/oracle/database/diag/rdbms/xffdb/xffdb/incident/incdir_64130/xffdb_smon_485190_i64130.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Fatal internal error happened while SMON was doing active transaction recovery.
Errors in file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_smon_485190.trc:
ORA-00600: internal error code, arguments: [KSMFPG2], [0x7F5AED10D000], [], [], [], [], [], [], [], [], [], []
ORA-00600: internal error code, arguments: [17182], [0x7F5AED10D7A0], [], [], [], [], [], [], [], [], [], []
SMON (ospid: 485190): terminating the instance due to error 474
System state dump requested by (instance=1, osid=485190 (SMON)), summary=[abnormal instance termination].
System State dumped to trace file /opt/oracle/database/diag/rdbms/xffdb/xffdb/trace/xffdb_diag_485150_20150708175939.trc
Dumping diagnostic data in directory=[cdmp_20150708175939], requested by (instance=1, osid=485190 (SMON)), summary=[abnormal instance termination].
Instance terminated by SMON, pid = 485190

通过进一步分析确定是67号文件有异常,使用dbv检查该文件发现

DBVERIFY: Release 11.2.0.4.0 - Production on Thu Jul 9 11:08:27 2015
Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.
DBVERIFY - Verification starting : FILE = /opt/oracle/database/oradata/xffdb/dk_adv_ts_main_25.dbf
Block Checking: DBA = 283035270, Block Type = KTB-managed data block
data header at 0x7f87813ea064
kdbchk: the amount of space used is not equal to block size
        used=7383 fsc=0 avsp=873 dtl=8088
Page 2016902 failed with check code 6110
Block Checking: DBA = 283035271, Block Type = KTB-managed data block
data header at 0x7f87813ec064
kdbchk: the amount of space used is not equal to block size
        used=8676 fsc=0 avsp=832 dtl=8088
Page 2016903 failed with check code 6110
Block Checking: DBA = 283035272, Block Type = KTB-managed data block
data header at 0x7f87813ee064
kdbchk: avsp(1306) > tosp(894)
Page 2016904 failed with check code 6128
Block Checking: DBA = 283035273, Block Type = KTB-managed data block
data header at 0x7f87813f0064
kdbchk: the amount of space used is not equal to block size
        used=7506 fsc=0 avsp=815 dtl=8088
Page 2016905 failed with check code 6110
Block Checking: DBA = 283035274, Block Type = KTB-managed data block
data header at 0x7f87813f2064
kdbchk: the amount of space used is not equal to block size
        used=7892 fsc=0 avsp=884 dtl=8088
Page 2016906 failed with check code 6110
…………
Block Checking: DBA = 283035368, Block Type = KTB-managed data block
data header at 0x7f87814ae064
kdbchk: the amount of space used is not equal to block size
        used=7934 fsc=0 avsp=837 dtl=8088
Page 2017000 failed with check code 6110
Block Checking: DBA = 283035369, Block Type = KTB-managed data block
data header at 0x7f87814b0064
kdbchk: the amount of space used is not equal to block size
        used=7683 fsc=0 avsp=883 dtl=8088
Page 2017001 failed with check code 6110
Block Checking: DBA = 283035370, Block Type = KTB-managed data block
data header at 0x7f87814b2064
kdbchk: the amount of space used is not equal to block size
        used=8556 fsc=0 avsp=841 dtl=8088
Page 2017002 failed with check code 6110
Block Checking: DBA = 283035371, Block Type = KTB-managed data block
data header at 0x7f87814b4064
kdbchk: the amount of space used is not equal to block size
        used=7460 fsc=0 avsp=822 dtl=8088
Page 2017003 failed with check code 6110
DBVERIFY - Verification complete
Total Pages Examined         : 4063232
Total Pages Processed (Data) : 2724435
Total Pages Failing   (Data) : 102
Total Pages Processed (Index): 1064839
Total Pages Failing   (Index): 0
Total Pages Processed (Other): 273957
Total Pages Processed (Seg)  : 0
Total Pages Failing   (Seg)  : 0
Total Pages Empty            : 1
Total Pages Marked Corrupt   : 0
Total Pages Influx           : 0
Total Pages Encrypted        : 0
Highest block SCN            : 212065913 (17.212065913)

把datafile 67给offline之后，数据库open正常，也不再crash。这里比较明显，导致主库和被动都出现异常,主库直接crash，然后重启后，一会儿就crash；备库激活后，重启后一会儿也就crash；两个库现象相同。最后通过dbv定位到是由于某个文件出现大量逻辑坏块，导致数据库open之后，进行回滚之时crash.对于此类故障，可以通过屏蔽事务回滚，并且通过alert日志和trace文件定位到异常对象，可以删除异常对象可以暂时解决该问题。而导致数据库出现类似问题。
通过这个案例，可以的出来，由于oracle某种bug或者其他原因，导致block 逻辑损坏，而且这个损坏会传输到备库，导致备库也异常，oracle的备份，容灾不能全部依赖物理dataguard容灾。因此在条件允许情况下，建议增加物理备份和逻辑容灾（类似OGG)

10g dataguard 只读和应用归档互换脚本

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：10g dataguard 只读和应用归档互换脚本

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

在Oracle 11g中能够通过active dataguard特性实现备库只读而且同步应用日志,10g及其以前版本,在应用日志的时候库是不能把打开到read only模式,对于一些实时性要求不是特别高的应用,比如他们可以实现在白天只读,查询业务,晚上应用日志.如果人工每天做类似处理工作量太大,这里因为有朋友需要,基于win平台写了类似功能脚本,通过计划任务调用来实现白天只读,晚上应用日志的功能
数据库从应用归档模式转换为只读模式

--change_open.bat
sqlplus / as sysdba @C:\oracle\product\script\change_open.sql
--change_open.sql
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE cancel;
alter database open read only;
exit;

数据库从只读模式转换到应用归档模式

--change_mount.bat
sqlplus / as sysdba @C:\oracle\product\script\change_mount.sql
--change_mount.sql
shutdown immediate;
startup mount ;
alter system register;
alter system register;
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT
LOGFILE DISCONNECT FROM SESSION;
conn sys/oracle@ntsy as sysdba
alter system set log_archive_dest_state=defer;
alter system set log_archive_dest_state=enable;
alter system switch logfile;
exit;

物理备库在read only时报ORA-01552错误处理

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：物理备库在read only时报ORA-01552错误处理

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

物理备库在read only时报ORA-01552错误

Tue Jan 06 11:53:38 中国标准时间 2015
alter database open read only
Tue Jan 06 11:53:38 中国标准时间 2015
SMON: enabling cache recovery
Tue Jan 06 11:53:39 中国标准时间 2015
Database Characterset is ZHS16GBK
Opening with internal Resource Manager plan
replication_dependency_tracking turned off (no async multimaster replication found)
Physical standby database opened for read only access.
Completed: alter database open read only
Tue Jan 06 11:54:04 中国标准时间 2015
Errors in file c:\oracle\product\10.2.0\admin\ntsy\udump\ntsy_ora_9080.trc:
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01552: 非系统表空间 'MY_SPACE' 不能使用系统回退段
ORA-06512: 在 line 2

分析trace文件

*** ACTION NAME:() 2015-01-06 11:54:04.828
*** MODULE NAME:(sqlplus.exe) 2015-01-06 11:54:04.828
*** SERVICE NAME:(SYS$USERS) 2015-01-06 11:54:04.828
*** SESSION ID:(1284.9) 2015-01-06 11:54:04.828
Error in executing triggers on connect internal
*** 2015-01-06 11:54:04.828
ksedmp: internal or fatal error
ORA-00604: 递归 SQL 级别 1 出现错误
ORA-01552: 非系统表空间 'MY_SPACE' 不能使用系统回退段
ORA-06512: 在 line 2
*** 2015-01-06 11:54:05.843
Process diagnostic dump for ORACLE.EXE (MMNL), OS id=10492,
pid: 13, proc_ser: 1, sid: <no session>

这里可以看出来，是由于执行触发器导致该问题,根据经验第一感觉很可能是logon之类的触发器导致。

查询触发器

SQL> select trigger_name,trigger_type,OWNER from dba_triggers where owner='OP';
TRIGGER_NAME                   TRIGGER_TYPE     OWNER
------------------------------ ---------------- ------------------------------
LOGAD                          AFTER EVENT      OP
TR_TRACE_DDL                   AFTER EVENT      OP

只有这两个触发器是基于事件的,另外从名字和dba_source中确定

SQL> select text from dba_source where name='LOGAD';
TEXT
--------------------------------------------------------------------------------
TRIGGER "OP".logad after logon on database
begin
insert into logad values (SYS_CONTEXT('USERENV', 'SESSION_USER'), SYSDATE,SYS_CO
NTEXT('USERENV','IP_ADDRESS')) ;
end;
已选择6行。
SQL> select text from dba_source where name='TR_TRACE_DDL';
TEXT
--------------------------------------------------------------------------------
TRIGGER "OP".tr_trace_ddl
AFTER ddl
ON database
DECLARE
sql_text ora_name_list_t;
state_sql ddl$trace.ddl_sql%TYPE;
BEGIN
FOR i IN 1..ora_sql_txt(sql_text) LOOP
state_sql := state_sql||sql_text(i);
END LOOP;
TEXT
--------------------------------------------------------------------------------
INSERT INTO ddl$trace(login_user,audsid,machine,ipaddress,
schema_user,schema_object,ddl_time,ddl_sql)
VALUES(ora_login_user,userenv('SESSIONID'),sys_context('userenv','host'),
sys_context('userenv','ip_address'),ora_dict_obj_owner,ora_dict_obj_name,SYSDATE
,state_sql);
EXCEPTION
WHEN OTHERS THEN
-- sp_write_log('捕获DDL语句异常错误：'||SQLERRM);
null;
END tr_trace_ddl;

基本上确定LOGAD是登录触发器,tr_trace_ddl是记录ddl触发器,那现在问题应该出在LOGAD的触发器上.因为该触发器在备库上当有用户登录之时,他也会工作插入记录到logad表中,由于数据库是只读,因此就出现了类似ORA-01552错误

解决方法
在触发器中加判断数据库角色条件,当数据库为物理备库之时才执行dml操作

SQL> CREATE OR REPLACE TRIGGER "OP".logad
  2  AFTER LOGON on database
  3  declare
  4  db_role varchar2(30);
  5  begin
  6  select database_role into db_role from v$database;
  7  If db_role <> 'PHYSICAL STANDBY' then
  8  insert into op.logad values (SYS_CONTEXT('USERENV', 'SESSION_USER'),
  9  SYSDATE,SYS_CONTEXT('USERENV','IP_ADDRESS')) ;
 10  end if;
 11  end;
 12  /
Warning: Trigger created with compilation errors.
SQL> show error;
Errors for TRIGGER "OP".logad:
LINE/COL ERROR
-------- -----------------------------------------------------------------
4/1      PL/SQL: SQL Statement ignored
4/40     PL/SQL: ORA-00942: table or view does not exist
SQL> conn / as sysdba
Connected.
SQL> grant select on v_$database to op;
Grant succeeded.
SQL> CREATE OR REPLACE TRIGGER "OP".logad
  2  AFTER LOGON on database
  3  declare
  4  db_role varchar2(30);
  5  begin
  6  select database_role into db_role from v$database;
  7  If db_role <> 'PHYSICAL STANDBY' then
  8  insert into op.logad values (SYS_CONTEXT('USERENV', 'SESSION_USER'),
  9 SYSDATE,SYS_CONTEXT('USERENV','IP_ADDRESS')) ;
 10  end if;
 12  end;
 12  /
Trigger created.

数据库open正常

Completed: ALTER DATABASE RECOVER MANAGED STANDBY DATABASE cancel
Tue Jan 06 13:51:20 中国标准时间 2015
alter database open read only
Tue Jan 06 13:51:21 中国标准时间 2015
SMON: enabling cache recovery
Tue Jan 06 13:51:21 中国标准时间 2015
Database Characterset is ZHS16GBK
Opening with internal Resource Manager plan
replication_dependency_tracking turned off (no async multimaster replication found)
Physical standby database opened for read only access.
Tue Jan 06 13:51:23 中国标准时间 2015
db_recovery_file_dest_size of 102400 MB is 0.00% used. This is a
user-specified limit on the amount of space that will be used by this
database for recovery-related files, and does not reflect the amount of
space available in the underlying filesystem or ASM diskgroup.
Tue Jan 06 13:51:23 中国标准时间 2015
Completed: alter database open read only

win平台rman备份和删除dg备库归档日志脚本

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：win平台rman备份和删除dg备库归档日志脚本

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

总觉得使用windows跑oracle是不靠谱的事情,可以这个世界上总有很多人喜欢做类似这样的事情,对于数据库比较常见的两件事情:rman和删除dg备库归档日志,在linux/unix平台上使用shell实现很简单,可是跑到win里面,就变的烦了,不是因为其麻烦,而是因为用的人少,不知道怎么下手处理该事情,我编写了简单的实现初级功能的win下面rman备份和删除备库归档日志脚本,供大家参考,也更加欢迎朋友提出来更加好的处理方法(win是真心的不懂)
rman备份脚本

--backup_oracle.bat文件
rman target / cmdfile=D:\backup\rman\backup_db.rman
log=d:/backup/rman/logfile/rmanlog%date:~0,4%%date:~5,2%%date:~8,2%.log
--backup_db.rman文件
CONFIGURE RETENTION POLICY TO REDUNDANCY = 2;
CONFIGURE DEVICE TYPE DISK PARALLELISM 2;
CONFIGURE DEFAULT DEVICE TYPE TO DISK;
backup filesperset = 5 as compressed backupset database format 'd:/backup/rman/full_%U.rman';
sql 'alter system archive log current';
backup  filesperset = 50 as compressed backupset archivelog  all format 'd:/backup/rman/arch_%U.rman'  delete input;
DELETE noprompt OBSOLETE;
crosscheck backup;
delete noprompt expired backup;
backup  format 'd:/backup/rman/ctl_%U.rman' current controlfile;
backup spfile format 'd:/backup/rman/spfile_%U.rman' ;
exit;

backup_oracle.bat文件加入到计划任务即可

删除dg备库归档日志(已经应用)

--delete_dg_archivelog.bat
rem 注意修改 部署目录
cd D:\win_xifenfei
d:
rem 注意delete_archive.sql 查询是否有记录
echo delete archivelog staring > delete_archivelog.bak
sqlplus / as sysdba @delete_archive.sql
echo rman target / cmdfile=rman_checkcross.rman>>delete_archivelog.bat
delete_archivelog.bat >>delete_dg_archivelog_%DATE:~0,4%%DATE:~5,2%%DATE:~8,2%".log
exit
--delete_archive.sql
set lines 150
col name for a150
set pagesize 0 feedback off verify off heading off echo off
spool delete_archivelog.bat
select 'del '||name from v$archived_log where APPLIED='YES' AND NAME IS NOT NULL and DEST_ID=1;
spool off
exit;
--rman_checkcross.rman
crosscheck archivelog all;
delete noprompt expired archivelog all;
exit

delete_dg_archivelog.bat加入到计划任务即可

ORACLE 12C Windows-Linux 部署DATAGURAD

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：ORACLE 12C Windows-Linux 部署DATAGURAD

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

环境描述
win 64中的ORACLE 12C(primary)与Linux 64中的ORACLE 12C(standby)搭建datagurad

primary force logging

C:\Users\XIFENFEI>sqlplus / as sysdba
SQL*Plus: Release 12.1.0.1.0 Production on 星期六 8月 24 16:59:53 2013
Copyright (c) 1982, 2013, Oracle.  All rights reserved.
连接到:
Oracle Database 12c Enterprise Edition Release 12.1.0.1.0 - 64bit Production
With the Partitioning, OLAP, Advanced Analytics and Real Application Testing options
CDB_CDB$ROOT@SYS> select banner from v$version;
BANNER
--------------------------------------------------------------------------------
Oracle Database 12c Enterprise Edition Release 12.1.0.1.0 - 64bit Production
PL/SQL Release 12.1.0.1.0 - Production
CORE    12.1.0.1.0      Production
TNS for 64-bit Windows: Version 12.1.0.1.0 - Production
NLSRTL Version 12.1.0.1.0 - Production
CDB_CDB$ROOT@SYS> alter database force logging ;
数据库已更改。

primary rman backup

backup filesperset = 5 as compressed backupset database format 'd:/backup/rman/full_%U.rman';

primary 生成standby controlfile

CDB_CDB$ROOT@SYS> ALTER DATABASE CREATE STANDBY CONTROLFILE AS 'c:/control01.ctl';
数据库已更改。
CDB_CDB$ROOT@SYS> create pfile='e:/pfile.txt' from spfile;
文件已创建。

standby 参数文件

DB_CREATE_FILE_DEST='+DATA'
db_create_online_log_dest_1='+DATA'
db_unique_name='cdb_dg'
service_names='cdb'
log_archive_dest_1='LOCATION=/u02/app/oracle/archivelog/ valid_for=(all_logfiles,all_roles)'
log_archive_dest_2='service=primary lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=cdb'
log_archive_config='dg_config=(cdb,cdb_dg)'
standby_file_management=auto
db_file_name_convert='E:\APP\XIFENFEI\ORADATA\','+DATA\CDB_DG\DATAFILE\'
log_file_name_convert='E:\APP\XIFENFEI\ORADATA\','+DATA\CDB_DG\LOGFILE\'
fal_server=primary

primary 修改参数

CDB_CDB$ROOT@SYS> alter system set log_archive_config='dg_config=(cdb,cdb_dg)';
系统已更改。
CDB_CDB$ROOT@SYS> alter system set log_archive_dest_2='
 2 service=standby lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=cdb_dg';
系统已更改。
CDB_CDB$ROOT@SYS> alter system set log_archive_dest_1=
  2 'LOCATION=USE_DB_RECOVERY_FILE_DEST valid_for=(all_logfiles,all_roles)';
系统已更改。
CDB_CDB$ROOT@SYS> alter system set fal_server=standby;
系统已更改。
CDB_CDB$ROOT@SYS> alter system set standby_file_management=auto;
系统已更改。
CDB_CDB$ROOT@SYS> alter system set db_file_name_convert='+DATA\CDB_DG\DATAFILE\','E:\APP\XIFENFEI\ORADATA\' scope=spfile;
系统已更改。
CDB_CDB$ROOT@SYS> alter system set log_file_name_convert='+DATA\CDB_DG\LOGFILE\','E:\APP\XIFENFEI\ORADATA\' scope=spfile;
系统已更改。

primary and standby tns

STANDBY =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.30.32 )(PORT = 1521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = cdb)
    )
  )
PRIMARY =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.30.1 )(PORT = 1521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = cdb)
    )
  )

standby restore controlfile

ASMCMD> cp /tmp/CONTROL01.CTL control01.ctl
copying /tmp/CONTROL01.CTL -> +data/cdb/control01.ctl

standby password file

[oracle@xifenfei dbs]$ pwd
/u01/app/oracle/product/12.1/db_1/dbs
[oracle@xifenfei dbs]$ cp /tmp/rman/PWDcdb.ora  orapwcdb

standby mount

SYS% cdb> create spfile from pfile='initcdb.ora';
File created.
SYS% cdb> startup mount;
ORACLE instance started.
Total System Global Area  521936896 bytes
Fixed Size                  2290264 bytes
Variable Size             314576296 bytes
Database Buffers          197132288 bytes
Redo Buffers                7938048 bytes
Database mounted.
SYS% cdb> select banner from v$version;
BANNER
--------------------------------------------------------------------------------
Oracle Database 12c Enterprise Edition Release 12.1.0.1.0 - 64bit Production
PL/SQL Release 12.1.0.1.0 - Production
CORE    12.1.0.1.0      Production
TNS for Linux: Version 12.1.0.1.0 - Production
NLSRTL Version 12.1.0.1.0 - Production
SYS% cdb> select name from v$datafile;
NAME
--------------------------------------------------------------------------------
E:\APP\XIFENFEI\ORADATA\CDB\SYSTEM01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\PDBSEED\SYSTEM01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\SYSAUX01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\PDBSEED\SYSAUX01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\UNDOTBS01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\USERS01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\PDB\SYSTEM01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\PDB\SYSAUX01.DBF
E:\APP\XIFENFEI\ORADATA\CDB\PDB\PDB_USERS01.DBF
9 rows selected.

standby rman restore

--清理控制文件中的备份集垃圾
DELETE noprompt OBSOLETE;
crosscheck backup;
delete noprompt expired backup;
--注册新备份集
catalog start with '/tmp/rman/';
--还原数据文件
run
{
set newname for database to '+data';
restore database;
switch datafile all;
}

standby clear redo

SYS% cdb> select group# from v$log;
    GROUP#
----------
         4
         6
         5
SYS% cdb> alter database clear logfile group 4;
Database altered.
SYS% cdb> alter database clear logfile group 5;
Database altered.
SYS% cdb> alter database clear logfile group 6;
Database altered.

standby add standby redolog

SYS% cdb> ALTER DATABASE ADD STANDBY LOGFILE GROUP 10  size 50M;
Database altered.
SYS% cdb> ALTER DATABASE ADD STANDBY LOGFILE GROUP 11  size 50M;
Database altered.
SYS% cdb> ALTER DATABASE ADD STANDBY LOGFILE GROUP 12  size 50M;
Database altered.
SYS% cdb> ALTER DATABASE ADD STANDBY LOGFILE GROUP 13  size 50M;
Database altered.

primary add standby redolog

CDB_CDB$ROOT@SYS> ALTER DATABASE ADD STANDBY LOGFILE GROUP 10  'E:\APP\XIFENFEI\ORADATA\CDB\std_redo10.log' size 50M;
数据库已更改。
CDB_CDB$ROOT@SYS> ALTER DATABASE ADD STANDBY LOGFILE GROUP 11  'E:\APP\XIFENFEI\ORADATA\CDB\std_redo11.log' size 50M;
数据库已更改。
CDB_CDB$ROOT@SYS> ALTER DATABASE ADD STANDBY LOGFILE GROUP 12  'E:\APP\XIFENFEI\ORADATA\CDB\std_redo12.log' size 50M;
数据库已更改。
CDB_CDB$ROOT@SYS> ALTER DATABASE ADD STANDBY LOGFILE GROUP 13  'E:\APP\XIFENFEI\ORADATA\CDB\std_redo13.log' size 50M;
数据库已更改。

standby readonly

SYS% cdb> ALTER DATABASE OPEN READ ONLY;
Database altered.

standby start mrp

SYS% cdb> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT FROM SESSION;
Database altered.

DATAGURAD 是否正常检查

--primary
CDB_CDB$ROOT@SYS> archive log list;
数据库日志模式            存档模式
自动存档             启用
存档终点            USE_DB_RECOVERY_FILE_DEST
最早的联机日志序列     374
下一个存档日志序列   376
当前日志序列           376
--standby
Mon Aug 12 13:56:51 2013
All non-current ORLs have been archived.
Mon Aug 12 13:56:53 2013
Media Recovery Log /u02/app/oracle/archivelog/1_370_820595806.dbf
Mon Aug 12 13:56:57 2013
Media Recovery Log /u02/app/oracle/archivelog/1_371_820595806.dbf
Mon Aug 12 13:57:02 2013
Media Recovery Log /u02/app/oracle/archivelog/1_372_820595806.dbf
Mon Aug 12 13:57:04 2013
Media Recovery Log /u02/app/oracle/archivelog/1_373_820595806.dbf
Mon Aug 12 13:57:05 2013
Media Recovery Log /u02/app/oracle/archivelog/1_374_820595806.dbf
Media Recovery Waiting for thread 1 sequence 375
Mon Aug 12 13:57:19 2013
Primary database is in MAXIMUM PERFORMANCE mode
RFS[2]: Assigned to RFS process (PID:26114)
RFS[2]: No standby redo logfiles created for thread 1
RFS[2]: Opened log for thread 1 sequence 376 dbid 1937199326 branch 820595806
Mon Aug 12 13:57:19 2013
RFS[3]: Assigned to RFS process (PID:26118)
RFS[3]: Opened log for thread 1 sequence 375 dbid 1937199326 branch 820595806
Mon Aug 12 13:57:19 2013
Archived Log entry 16 added for thread 1 sequence 375 rlc 820595806 ID 0x7377d8de dest 2:
Mon Aug 12 13:57:22 2013
Media Recovery Log /u02/app/oracle/archivelog/1_375_820595806.dbf
Media Recovery Waiting for thread 1 sequence 376 (in transit)

11G RAC TO 11G RAC ADG配置

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：11G RAC TO 11G RAC ADG配置

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

RDBMS tnsnames.ora配置
在主备库的数据库的tns中均增加如下配置

q9dbdg =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = q9db01-dg-vip)(PORT = 1522))
    (ADDRESS = (PROTOCOL = TCP)(HOST = q9db02-dg-vip)(PORT = 1522))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = q9db)
    )
)
q9adgdg =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = q9adg01-dg-vip)(PORT = 1522))
    (ADDRESS = (PROTOCOL = TCP)(HOST = q9adg02-dg-vip)(PORT = 1522))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = q9db)
    )
)

主库修改参数配置

alter system set DB_UNIQUE_NAME=q9db scope=spfile;
alter system set LOG_ARCHIVE_CONFIG='DG_CONFIG=(q9db,q9db_adg)' scope=both;
alter system set LOG_ARCHIVE_DEST_1='LOCATION=+arch/q9db/ARCHIVELOG VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=q9db' scope=both;
alter system set LOG_ARCHIVE_DEST_2='SERVICE=q9adg1dg LGWR SYNC AFFIRM VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=q9db_adg';
alter system set standby_file_management='AUTO' sid='*' scope=both;
alter system set fal_client='q9dbdg';
alter system set fal_server='q9adgdg';
alter system set db_file_name_convert='+DATA','+DATA' SCOPE=SPFILE;
ALTER SYSTEM SET log_file_name_convert='+DATA','+DATA' SCOPE=SPFILE;

备库修改参数配置

alter system set DB_UNIQUE_NAME=q9db_adg scope=spfile;
alter system set LOG_ARCHIVE_CONFIG='DG_CONFIG=(q9db,q9db_adg)' scope=both;
alter system set LOG_ARCHIVE_DEST_1='LOCATION=+arch/q9db/ARCHIVELOG VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=q9db_adg';
alter system set LOG_ARCHIVE_DEST_2='SERVICE=q9db1dg LGWR SYNC AFFIRM VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=q9db';
alter system set standby_file_management='AUTO' sid='*' scope=both;
alter system set fal_client='q9adgdg';
alter system set fal_server='q9dbdg';
alter system set service_names='q9db';
alter system set db_file_name_convert='+DATA','+DATA' SCOPE=SPFILE;
ALTER SYSTEM SET log_file_name_convert='+DATA','+DATA' SCOPE=SPFILE;

主库增加standby redo log

alter database add standby logfile THREAD 1 group 51 '+DATA' size 512m;
…………
alter database add standby logfile THREAD 1 group 61 '+DATA' size 512m;
alter database add standby logfile THREAD 2 group 71 '+DATA' size 512m;
…………
alter database add standby logfile THREAD 2 group 81 '+DATA' size 512m;

主库修改强制日志

SQL> ALTER DATABASE FROCE LOGGING;

RMAN备份主库

rman target /
run{
configure device type disk parallelism 16;
backup database filesperset = 10  format '/q9bak/df_%U';
backup current controlfile format '/q9bak/ctl_%U';
}

创建standby controlfile

ALTER DATABASE CREATE STANDBY CONTROLFILE AS '/q9bak/stb_ctl';

还原standby controlfile

ASMCMD> cp /q9bak/stb_ctl control01.ctl
copying /q9bak/stb_ctl -> +data/Q9DB/control01.ctl
ASMCMD> cp /q9bak/stb_ctl control02.ctl
copying /q9bak/stb_ctl -> +arch/q9db/control02.ctl

还原备库

SQL>startup mount;
rman target / nocatalog <<EOF
run{
configure default device type to disk;
configure device type disk parallelism 16;
restore database;
}
<<EOF

修改备库文件名

RMAN>catalog start with '+DATA/Q9DB/DATAFILE/';
RMAN> switch database to copy;

清理redo

--清理redo(所有的group#)
SQL> select group# from v$log;
SQL> alter database clear logfile group 1;
……
--清理standby redo(所有的group#)
SQL> select group# from v$standby_log;
SQL> alter database clear logfile group 51;
……

开启备库实时应用模式

SQL> alter database recover managed standby database using current logfile disconnect from session;

备库注册DB服务(oracle用户)

srvctl add database -d q9db_adg -o $ORACLE_HOME
srvctl add instance -d q9db_adg -i q9db1 -n q9adg01
srvctl add instance -d q9db_adg -i q9db2 -n q9adg02
srvctl modify database -d q9db_adg -a DATA,ARCH

11G RAC TO 11G RAC ADG SWITCHOVER

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：11G RAC TO 11G RAC ADG SWITCHOVER

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

11G RAC TO 11G RAC ADG切换过程
主库准备工作

SQL> select inst_id,database_role,OPEN_MODE from  gv$database;
   INST_ID DATABASE_ROLE    OPEN_MODE
---------- ---------------- --------------------
         2 PRIMARY          READ WRITE
         1 PRIMARY          READ WRITE
[oracle@q9db02 ~]$ srvctl stop instance -d q9db -i q9db2
SQL> select inst_id,database_role,OPEN_MODE from  gv$database;
   INST_ID DATABASE_ROLE    OPEN_MODE
---------- ---------------- --------------------
         1 PRIMARY          READ WRITE
SQL> CREATE RESTORE POINT SWITCHOVER_START_GRP GUARANTEE FLASHBACK DATABASE;
Restore point created.

备库准备工作

SQL> select inst_id,database_role,OPEN_MODE from  gv$database;
   INST_ID DATABASE_ROLE    OPEN_MODE
---------- ---------------- --------------------
         2 PHYSICAL STANDBY READ ONLY WITH APPLY
         1 PHYSICAL STANDBY READ ONLY WITH APPL
[oracle@q9adg02 ~]$ srvctl stop instance -d q9db_adg -i q9db2
SQL> select inst_id,database_role,OPEN_MODE from  gv$database;
   INST_ID DATABASE_ROLE    OPEN_MODE
---------- ---------------- --------------------
         1 PHYSICAL STANDBY READ ONLY WITH APPL
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
Database altered.
SQL> CREATE RESTORE POINT SWITCHOVER_START_GRP GUARANTEE FLASHBACK DATABASE;
Restore point created.
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT;
Database altered.

主库切换日志,观察备库

--主库
SQL> alter system switch logfile;
System altered.
SQL> /
System altered.
--备库
[oracle@q9adg01 trace]$ tail -f alert_q9db1.log
Tue Jun 25 15:35:27 2013
RFS[10]: Selected log 52 for thread 1 sequence 4777 dbid 844605368 branch 817913807
Tue Jun 25 15:35:28 2013
Archived Log entry 4889 added for thread 1 sequence 4776 ID 0x3545ffea dest 1:
Tue Jun 25 15:35:28 2013
Media Recovery Waiting for thread 1 sequence 4777 (in transit)
Tue Jun 25 15:35:28 2013
RFS[11]: Selected log 72 for thread 2 sequence 1630 dbid 844605368 branch 817913807
Recovery of Online Redo Log: Thread 1 Group 52 Seq 4777 Reading mem 0
  Mem# 0: +DATA/q9db_adg/onlinelog/group_52.1564.818724635
Media Recovery Waiting for thread 2 sequence 1630 (in transit)
Recovery of Online Redo Log: Thread 2 Group 72 Seq 1630 Reading mem 0
  Mem# 0: +DATA/q9db_adg/onlinelog/group_72.1575.818724653
Tue Jun 25 15:35:30 2013
Archived Log entry 4890 added for thread 2 sequence 1629 ID 0x3545ffea dest 1:

几乎同步进行表示主备日志传输应用正常

主库切换

SQL> SELECT SWITCHOVER_STATUS FROM V$DATABASE;
SWITCHOVER_STATUS
--------------------
TO STANDBY
SQL> ALTER DATABASE COMMIT TO SWITCHOVER TO STANDBY WITH SESSION SHUTDOWN;
Database altered.

备库切换

SQL> SELECT SWITCHOVER_STATUS FROM V$DATABASE;
SWITCHOVER_STATUS
--------------------
TO PRIMARY
SQL> ALTER DATABASE COMMIT TO SWITCHOVER TO PRIMARY WITH SESSION SHUTDOWN;
Database altered.
SQL> ALTER DATABASE OPEN;
Database altered.

继续处理主库

SQL> shutdown immediate
ORA-01092: ORACLE instance terminated. Disconnection forced
SQL> exit
Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production
With the Partitioning, Real Application Clusters, Automatic Storage Management, Data Mining
and Real Application Testing options
[oracle@q9db01 ogg]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.3.0 Production on Tue Jun 25 14:13:58 2013
Copyright (c) 1982, 2011, Oracle.  All rights reserved.
Connected to an idle instance.
SQL> startup mount
ORACLE instance started.
Total System Global Area 1.6034E+11 bytes
Fixed Size                  2236968 bytes
Variable Size            2.5770E+10 bytes
Database Buffers         1.3422E+11 bytes
Redo Buffers              352468992 bytes
Database mounted.
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT;
Database altered.
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE cancel;
Database altered.
SQL> alter database open;
Database altered.
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT;
Database altered.

清理快照

--主库
SQL> DROP RESTORE POINT SWITCHOVER_START_GRP;
Restore point dropped.
--备库
SQL> startup mount
ORACLE instance started.
Total System Global Area 1.6034E+11 bytes
Fixed Size                  2236968 bytes
Variable Size            2.7380E+10 bytes
Database Buffers         1.3261E+11 bytes
Redo Buffers              352468992 bytes
Database mounted.
SQL> DROP RESTORE POINT SWITCHOVER_START_GRP;
Restore point dropped.
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT;
Database altered.
SQL>  ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
Database altered.
SQL> alter database open;
Database altered.
SQL>  ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT;
Database altered.

启动主备另外节点

--主库
[oracle@q9db01 ~]$ srvctl start instance -d q9db -i q9db2
--备库
[oracle@q9adg02 ~]$ srvctl start instance -d q9db_adg -i q9db2

补充说明:如果出现日志切换暂时不能传输

备库执行(因为重启动态监听没有马上别识别)
alter system register;
主库执行
alter system set log_archive_dest_state_2=enable;

修改dataguard主库redo组数和大小

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：修改dataguard主库redo组数和大小

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

在一个dg环境中，配置的是实时同步,需要增加主库的redo大小和组数,本来是一个很简单的问题,解决思路是：先备库增加standby redo删除老standby redo,然后主库增加redo删除老redo,备库增加新redo删除老redo,最后主库增加standby redo。但是在实施过程中，遇到了一些细节性的问题，主要是学习到了log_file_name_convert如果不配置,将导致备库redo 文件不能被删除

standby redo log管理
增加standby redo log

SQL> alter database add standby logfile group 8 ('/data/oradata/wasudb/st_redo08.log') size 200M;
alter database add standby logfile group 8 ('/data/oradata/wasudb/st_redo08.log') size 200M
*
ERROR at line 1:
ORA-01156: recovery in progress may need access to files
SQL> alter database recover managed standby database cancel;
Database altered.
SQL> alter database add standby logfile group 9 ('/data/oradata/wasudb/st_redo09.log') size 200M;
Database altered.
SQL> alter database add standby logfile group 10 ('/data/oradata/wasudb/st_redo10.log') size 200M;
Database altered.
SQL> alter database add standby logfile group 11 ('/data/oradata/wasudb/st_redo11.log') size 200M;
Database altered.
SQL> alter database add standby logfile group 12 ('/data/oradata/wasudb/st_redo12.log') size 200M;
Database altered.

删除standby redo log

SQL> alter database drop logfile group 4;
alter database drop logfile group 4
*
ERROR at line 1:
ORA-00261: log 4 of thread 1 is being archived or modified
ORA-00312: online log 4 thread 1: '/data/oradata/wasudb/st_redo04.log'
SQL> alter database drop logfile group 5;;
Database altered.
SQL> alter database drop logfile group 6;
Database altered.
SQL> alter database drop logfile group 7;
Database altered.
--在主库多次执行switch logfile
SQL>  alter database drop logfile group 4;
Database altered.

主库redo log 管理
增加redo log

SQL> alter database add logfile group 21 ('/data/oradata/wasudb/redo21.log') size 200m;
Database altered.
SQL> alter database add logfile group 22 ('/data/oradata/wasudb/redo22.log') size 200m;
Database altered.
SQL> alter database add logfile group 23 ('/data/oradata/wasudb/redo23.log') size 200m;
Database altered.
SQL> alter database add logfile group 24 ('/data/oradata/wasudb/redo24.log') size 200m;
Database altered.

删除redo log

SQL> alter database drop logfile group 1;
Database altered.
SQL> alter database drop logfile group 2;
Database altered.
SQL> alter database drop logfile group 3;
Database altered.
--这里涉及到多次switch logfile,需要确定redo是inactive才能够删除

备库redo log管理
增加redo log

SQL> alter database add logfile group 21 ('/data/oradata/wasudb/redo21.log') size 200m;
alter database add logfile group 21 ('/data/oradata/wasudb/redo21.log') size 200m
*
ERROR at line 1:
ORA-01275: Operation ADD LOGFILE is not allowed if standby file management is automatic.
SQL>  alter system set standby_file_management=manual;
System altered.
SQL> alter database add logfile group 21 ('/data/oradata/wasudb/redo21.log') size 200m;
Database altered.
SQL> alter database add logfile group 22 ('/data/oradata/wasudb/redo22.log') size 200m;
Database altered.
SQL> alter database add logfile group 23 ('/data/oradata/wasudb/redo23.log') size 200m;
Database altered.
SQL> alter database add logfile group 24 ('/data/oradata/wasudb/redo24.log') size 200m;
Database altered.

删除redo log

SQL> alter database drop logfile group 1;
alter database drop logfile group 1
*
ERROR at line 1:
ORA-01624: log 1 needed for crash recovery of instance wasudb (thread 1)
ORA-00312: online log 1 thread 1: '/data/oradata/wasudb/redo01.log'
SQL>  ALTER DATABASE CLEAR LOGFILE GROUP 1;
 ALTER DATABASE CLEAR LOGFILE GROUP 1
*
ERROR at line 1:
ORA-19527: physical standby redo log must be renamed
ORA-00312: online log 1 thread 1: '/data/oradata/wasudb/redo01.log'
SQL> show parameter NAME_CONVERT
NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
db_file_name_convert                 string
log_file_name_convert                string
SQL> alter system set log_file_name_convert='/data/oradata/wasudb','/data/oradata/wasudb' scope=spfile;
System altered.
SQL>  alter system set db_file_name_convert='/data/oradata/wasudb','/data/oradata/wasudb' scope=spfile;
System altered.
--重启数据库
SQL> show parameter file_name_convert;
NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
db_file_name_convert                 string      /data/oradata/wasudb, /data/or
                                                 adata/wasudb
log_file_name_convert                string      /data/oradata/wasudb, /data/or
                                                 adata/wasudb
SQL> alter database drop logfile group 1;
alter database drop logfile group 1
*
ERROR at line 1:
ORA-01624: log 1 needed for crash recovery of instance wasudb (thread 1)
ORA-00312: online log 1 thread 1: '/data/oradata/wasudb/redo01.log'
SQL> ALTER DATABASE CLEAR LOGFILE GROUP 1;
Database altered.
SQL> alter database drop logfile group 1;
Database altered.
SQL> ALTER DATABASE CLEAR LOGFILE GROUP 2;
Database altered.
SQL> alter database drop logfile group 2;
Database altered.
SQL> ALTER DATABASE CLEAR LOGFILE GROUP 3;
Database altered.
SQL> alter database drop logfile group 3;
Database altered.

主库standby redo log管理
增加standby redo

SQL> alter database add logfile group 21 ('/data/oradata/wasudb/redo21.log') size 200m;
Database altered.
SQL> alter database add logfile group 22 ('/data/oradata/wasudb/redo22.log') size 200m;
Database altered.
SQL> alter database add logfile group 23 ('/data/oradata/wasudb/redo23.log') size 200m;
Database altered.
SQL> alter database add logfile group 24 ('/data/oradata/wasudb/redo24.log') size 200m;
Database altered.

删除standby redo

SQL> alter database drop logfile group 4;
Database altered.
SQL> alter database drop logfile group 5;
Database altered.
SQL> alter database drop logfile group 6;
Database altered.
SQL> alter database drop logfile group 7;
Database altered.

后续工作

SQL> alter system set standby_file_management=auto;
System altered.
SQL> alter database recover managed standby database using current logfile disconnect from session;
Database altered.

至此修改dataguard环境中的redo大小和增加redo组数的操作基本完成.在这里主要修正了自己以前对log_file_name_convert的认识，我以前以为如果我的主备库redo对应的目录一致不用配置该参数,今天通过查询MOS发现从10.2开始数据库为了能实现dg的快速切换在mrp启动的时候会去尝试清理备库redo,如果备库没有redo,或者log_file_name_convert配置不正确导致不能正常执行这个清理工作,数据库就会报ORA-19527,特别是在mrp进程启动之时.对于本次出现执行CLEAR LOGFILE命令也出现该问题,确实有点过犹不及了.这里也就是提醒我们:就算redo file,datafile主备位置相同,也建议配置log_file_name_convert和db_file_name_convert参数,提高dg健壮性.

EXADATA与非EXADATA搭建DATAGURAD关于EHCC特性测试

联系：手机/微信(+86 17813235971) QQ(107644445)

标题：EXADATA与非EXADATA搭建DATAGURAD关于EHCC特性测试

作者：惜分飞©版权所有[未经本人同意,不得以任何形式转载,否则有进一步追究法律责任的权利.]

随着xd的越来越普及，不少的企业使用了xd，但是不少企业因为资金有限，只有一台xd，但是为了实现数据的容灾，可能会使用一台非xd的机器来通过dataguard来实现容灾,但是因为xd的ehcc新特性,官方宣传是只在xd中支持,如果dg的备库不是xd。那么会怎么样,这里通过测试得出如下一些结论：xd与非xd可以构造dg,ehcc功能在xd上无法高效使用。对于这样的环境条件下,使用ORACLE自带压缩效率更高.针对ehcc压缩效率很低,个人猜测,是因为xd检查到备库是非xd环境,直接对ehcc进行了降级压缩处理,从而出现了ehcc的压缩效率比oltp还低(牺牲了xd的性能,确保了数据的安全,看来xd的设计还是考虑的比较全面)
xd基本信息

SQL> select * from v$version;
BANNER
--------------------------------------------------------------------------------
Oracle Database 11g Enterprise Edition Release 11.2.0.2.0 - 64bit Production
PL/SQL Release 11.2.0.2.0 - Production
CORE    11.2.0.2.0      Production
TNS for Linux: Version 11.2.0.2.0 - Production
NLSRTL Version 11.2.0.2.0 - Production
SQL> show parameter clu;
NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
cluster_database                     boolean     TRUE
cluster_database_instances           integer     2
cluster_interconnects                string
SQL> SELECT OPEN_MODE, DATABASE_ROLE ,NAME FROM V$DATABASE;
OPEN_MODE            DATABASE_ROLE    NAME
-------------------- ---------------- ---------
READ WRITE           PRIMARY          xxxxxx
SQL> !uname -a
Linux dm01db02 2.6.18-194.3.1.0.4.el5 #1 SMP Sat Feb 19 03:38:37 EST 2011 x86_64 x86_64 x86_64 GNU/Linux

xd创建模拟表

SQL> create table t_FF_c compress as select  * from dba_objects;
Table created.
SQL> create table t_FF_c_o compress for oltp  as select  * from dba_objects;
Table created.
SQL> create table t_FF_q_l compress for query low as select  * from dba_objects;
Table created.
SQL> create table t_FF_q_h compress for query high  as select  * from dba_objects;
Table created.
SQL> create table t_FF_a_l compress for archive low  as select  * from dba_objects;
Table created.
SQL> create table t_FF_a_h compress for archive high as select * from dba_objects;
Table created.
SQL> create table t_ff as select * from dba_objects;
Table created.

xd查询模拟表

SQL> select s.owner,segment_name,s.bytes/1024/1024 t_size,compress_for
from dba_segments s,dba_tables t
where s.owner=t.owner and t.table_name=s.segment_name  and t.table_name like 'T_FF%';  2    3
OWNER                          SEGMENT_NAME                       T_SIZE COMPRESS_FOR
------------------------------ ------------------------------ ---------- ------------
SYS                            T_FF                                   11
SYS                            T_FF_A_H                               10 ARCHIVE HIGH
SYS                            T_FF_A_L                               10 ARCHIVE LOW
SYS                            T_FF_Q_H                               10 QUERY HIGH
SYS                            T_FF_Q_L                               10 QUERY LOW
SYS                            T_FF_C_O                                4 OLTP
SYS                            T_FF_C                                  4 BASIC

通过这里发现,带有非dg的xd使用ehcc压缩效率都低了很多

非xd备库基本信息

SQL> SELECT OPEN_MODE, DATABASE_ROLE ,NAME FROM V$DATABASE;
OPEN_MODE            DATABASE_ROLE    NAME
-------------------- ---------------- ---------
READ ONLY WITH APPLY PHYSICAL STANDBY xxxxxx
SQL> show parameter clu;
NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
cluster_database                     boolean     FALSE
cluster_database_instances           integer     1
cluster_interconnects                string
SQL> !uname -a
Linux oradg 2.6.18-238.el5xen #1 SMP Sun Dec 19 14:42:02 EST 2010 x86_64 x86_64 x86_64 GNU/Linux

查询非xd dg备库

SQL> select s.owner,segment_name,s.bytes/1024/1024 t_size,compress_for
from dba_segments s,dba_tables t
where s.owner=t.owner and t.table_name=s.segment_name  and t.table_name like 'T_FF%';
OWNER                          SEGMENT_NAME                       T_SIZE COMPRESS_FOR
------------------------------ ------------------------------ ---------- ------------
SYS                            T_FF                                   11
SYS                            T_FF_A_H                               10 ARCHIVE HIGH
SYS                            T_FF_A_L                               10 ARCHIVE LOW
SYS                            T_FF_Q_H                               10 QUERY HIGH
SYS                            T_FF_Q_L                               10 QUERY LOW
SYS                            T_FF_C_O                                4 OLTP
SYS                            T_FF_C                                  4 BASIC
SQL> SELECT COUNT(*) FROM T_FF_Q_L;
  COUNT(*)
----------
     94709
SQL> SELECT COUNT(*) FROM T_FF_Q_H;
  COUNT(*)
----------
     94710
SQL> SELECT COUNT(*) FROM T_FF_C_O;
  COUNT(*)
----------
     94708
SQL> SELECT COUNT(*) FROM T_FF_C;
  COUNT(*)
----------
     94707
SQL> SELECT COUNT(*) FROM T_FF_A_L;
  COUNT(*)
----------
     94711
SQL> SELECT COUNT(*) FROM T_FF_A_H;
  COUNT(*)
----------
     94712
SQL> select count(*) from t_FF;
  COUNT(*)
----------
    94713

通过这里测试证明,对于非xd dg库,可以正常的查询xd上的ehcc相关表,而且相关大小也相同(物理dg当然相同了)

测试xd与非xd dg测试ehcc的dml操作

--xd 主库
SQL> update t_ff_a_h set owner='www.xifenfei.com';
94712 rows updated.
SQL> commit;
Commit complete.
SQL>  select s.owner,segment_name,s.bytes/1024/1024 t_size,compress_for
from dba_segments s,dba_tables t
where s.owner=t.owner and t.table_name=s.segment_name  and t.table_name like 'T_FF%';  2    3
OWNER                          SEGMENT_NAME                       T_SIZE COMPRESS_FOR
------------------------------ ------------------------------ ---------- ------------
SYS                            T_FF                                   11
SYS                            T_FF_A_H                               11 ARCHIVE HIGH
SYS                            T_FF_A_L                               10 ARCHIVE LOW
SYS                            T_FF_Q_H                               10 QUERY HIGH
SYS                            T_FF_Q_L                               10 QUERY LOW
SYS                            T_FF_C_O                                4 OLTP
SYS                            T_FF_C                                  4 BASIC
7 rows selected.
SQL> alter system switch logfile;
System altered.
--非xd 备库
SQL> select s.owner,segment_name,s.bytes/1024/1024 t_size,compress_for
from dba_segments s,dba_tables t
where s.owner=t.owner and t.table_name=s.segment_name  and t.table_name like 'T_FF%';
OWNER                          SEGMENT_NAME                       T_SIZE COMPRESS_FOR
------------------------------ ------------------------------ ---------- ------------
SYS                            T_FF                                   11
SYS                            T_FF_A_H                               11 ARCHIVE HIGH
SYS                            T_FF_A_L                               10 ARCHIVE LOW
SYS                            T_FF_Q_H                               10 QUERY HIGH
SYS                            T_FF_Q_L                               10 QUERY LOW
SYS                            T_FF_C_O                                4 OLTP
SYS                            T_FF_C                                  4 BASIC

证明对于xd与非xd构成的dg环境,可以执行dml操作.

测试xd与非xd dg的ehcc的append操作

--xd准备三张测试空表
SQL> create table t_FF_a_l_1 compress for archive low as select * from t_FF  where 1=0;
Table created.
SQL> create table t_FF_a_h_1 compress for archive high  as select * from t_FF  where 1=0;
Table created.
SQL>truncate table t_FF;
Table truncated.
--插入数据(每个表执行5次)
SQL> insert /*+ APPEND */ into t_FF_a_l_1 select * from dba_objects;
94714 rows created.
SQL> commit;
Commit complete.
SQL> insert /*+ APPEND */ into t_FF_h_l_1 select * from dba_objects;
94714 rows created.
SQL> commit;
Commit complete.
SQL> insert /*+ APPEND */ into t_FF select * from dba_objects;
94714 rows created.
SQL> commit;
Commit complete.
--查看相关表数据量
SQL> select count(*) from t_FF_a_l_1;
  COUNT(*)
----------
    473570
SQL> select count(*) from t_FF_a_h_1;
  COUNT(*)
----------
    473570
SQL> select count(*) from t_FF;
  COUNT(*)
----------
    473570
--查看xd主库
SQL> select s.owner,segment_name,s.bytes/1024/1024 t_size,compress_for
from dba_segments s,dba_tables t
where s.owner=t.owner and t.table_name=s.segment_name  and t.table_name like 'T_FF%';
OWNER                          SEGMENT_NAME                       T_SIZE COMPRESS_FOR
------------------------------ ------------------------------ ---------- ------------
SYS                            T_FF                                   52
SYS                            T_FF_A_H                               11 ARCHIVE HIGH
SYS                            T_FF_A_L                               10 ARCHIVE LOW
SYS                            T_FF_Q_H                               10 QUERY HIGH
SYS                            T_FF_Q_L                               10 QUERY LOW
SYS                            T_FF_C_O                                4 OLTP
SYS                            T_FF_C                                  4 BASIC
SYS                            T_FF_A_H_1                             47 ARCHIVE HIGH
SYS                            T_FF_A_L_1                             47 ARCHIVE LOW
--查看非xd备库
SQL> select s.owner,segment_name,s.bytes/1024/1024 t_size,compress_for
from dba_segments s,dba_tables t
where s.owner=t.owner and t.table_name=s.segment_name  and t.table_name like 'T_FF%';
OWNER                          SEGMENT_NAME                       T_SIZE COMPRESS_FOR
------------------------------ ------------------------------ ---------- ------------
SYS                            T_FF                                   52
SYS                            T_FF_A_H                               11 ARCHIVE HIGH
SYS                            T_FF_A_L                               10 ARCHIVE LOW
SYS                            T_FF_Q_H                               10 QUERY HIGH
SYS                            T_FF_Q_L                               10 QUERY LOW
SYS                            T_FF_C_O                                4 OLTP
SYS                            T_FF_C                                  4 BASIC
SYS                            T_FF_A_H_1                             47 ARCHIVE HIGH
SYS                            T_FF_A_L_1                             47 ARCHIVE LOW

试验整体结论
1.xd可以与非xd机器构建dg容灾环境(不会因为非dg不支持ehcc而导致无法搭建他们之间的dg)
2.xd与非xd的dg,ehcc功能大大缩水,基本上和非压缩状态差不多,比OLTP低很多
3.xd与非xd的dg在备库中支持select,dml,hint append等操作,这些操作是因为ehcc表在xd端就进行了ehcc降级导致