扫二维码与项目经理沟通
我们在微信上24小时期待你的声音
解答本文疑问/技术咨询/运营咨询/技术建议/互联网交流
站在用户的角度思考问题，与客户深入沟通，找到文圣网站设计与文圣网站推广的解决方案，凭借多年的经验，让设计与互联网技术结合，创造个性化、用户体验好的作品，建站类型包括：网站设计、成都网站制作、企业官网、英文网站、手机端网站、网站推广、域名注册、网页空间、企业邮箱。业务覆盖文圣地区。
本文文档结构：
这里配置的过程中需要注意的一项是多看看rac1和rac2以及dg的告警日志会对配置过程有更深刻的理解。。。配置oracle rac的dg环境说白了首先将主库的数据恢复到从库上，然后在主从库上配置一些DG需要用到的参数
注意：
该步骤主要是① 将rac主库修改为归档+forcelogging模式 ② 利用rman备份db、archivelog、control file 、pfile 和口令文件
和下一步一起做
alter database force logging;
alter database archivelog; ---归档模式下修改
SQL> alter database force logging;
Database altered.
SQL> select name , open_mode, log_mode,force_logging from gv$database;
NAME OPEN_MODE LOG_MODE FOR
--------- -------------------- ------------ ---
RACDB READ WRITE NOARCHIVELOG YES
RACDB READ WRITE NOARCHIVELOG YES
SQL> exit
[oracle@rac1 ~]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.3.0 Production on Fri Oct 3 15:30:17 2014
Copyright (c) 1982, 2011, Oracle. All rights reserved.
Connected to an idle instance.
SQL> startup mount
ORACLE instance started.
Total System Global Area 784998400 bytes
Fixed Size 2232472 bytes
Variable Size 532680552 bytes
Database Buffers 247463936 bytes
Redo Buffers 2621440 bytes
Database mounted.
SQL> select name , open_mode, log_mode,force_logging from gv$database;
NAME OPEN_MODE LOG_MODE FOR
--------- -------------------- ------------ ---
RACDB MOUNTED NOARCHIVELOG YES
SQL> alter database archivelog;
Database altered.
SQL> shutdown immediate;
ORA-01109: database not open
Database dismounted.
ORACLE instance shut down.
SQL>
SQL> select name , open_mode, log_mode,force_logging from gv$database;
NAME OPEN_MODE LOG_MODE FOR
--------- -------------------- ------------ ---
RACDB READ WRITE ARCHIVELOG YES
RACDB READ WRITE ARCHIVELOG YES
SQL>
run {
allocate channel c1 type disk;
allocate channel c2 type disk;
allocate channel c3 type disk;
allocate channel c4 type disk;
backup database format '/rman_backup/FULL_%U.bak';
backup archivelog all format '/rman_backup/ARC_%U.bak';
release channel c1;
release channel c2;
release channel c3;
release channel c4;
}
Rman下：backup device type disk format '/rman_backup/standby_%U.ctl' current controlfile for standby;
或：SQL>alter database create standby controlfile as '/dats/backup/standby.ctl'
create pfile ='/rman_backup/initphydb.ora' from spfile;
SQL> show parameter spfile
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
spfile string +DATA/racdb/spfileracdb.ora
SQL>
SQL> show parameter cluster_database
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
cluster_database boolean TRUE
cluster_database_instances integer 2
SQL>
最好是将其中一个口令文件复制到另外一个节点上，最后复制到dg下，确保口令文件是一样的
将备份文件全部传递到dg库：
----rac1
scp /rman_backup/*.bak oracle@192.168.59.140:/rman_backup/
scp /u01/app/oracle/product/11.2.0/dbhome_1/dbs/orapwracdb1 oracle@192.168.59.140:/u01/app/oracle/product/11.2.0/dbhome_1/dbs/orapwphydb
上一步已做，不需要创建了
----------dg
[oracle@dg rman_backup]$ cp initphydb.ora $ORACLE_HOME/dbs/
[oracle@dg rman_backup]$ cd $ORACLE_HOME/dbs
这个是比较重要的步骤，我把行号弄出来了，当然写的时候是没有前边的行号的
物理备库初始化参数文件修改：
[oracle@dg dbs]$ mkdir -p /u01/app/oracle/admin/phydb/adump
[oracle@dg dbs]$ mkdir -p /u01/app/oracle/diag/rdbms/phydb/cdump
[oracle@dg dbs]$
[grid@dg ~]$ asmcmd
ASMCMD> ls
DATA/
FRA/
GRIDDG/
ASMCMD> cd DATA
ASMCMD> ls
ASMCMD> mkdir phydb
ASMCMD> ls
phydb/
ASMCMD>
将以下内容复制到rac1，rac2和dg库：
racdb =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.59.139)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = racdb.lhr.com)
)
)
racdb1 =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.59.135)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = racdb.lhr.com)
(INSTANCE_NAME = racdb1)
)
)
racdb2 =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.59.136)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = racdb.lhr.com)
(INSTANCE_NAME = racdb2)
)
)
phydb =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.59.140)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = phydb)
)
)
验证一下：
分别在3个库上执行：
sqlplus sys/lhr@racdb as sysdba
sqlplus sys/lhr@racdb1 as sysdba
sqlplus sys/lhr@racdb2 as sysdba
确保3个库可以连接
startup nomount
[oracle@dg dbs]$ rman target /
Recovery Manager: Release 11.2.0.3.0 - Production on Fri Oct 3 19:06:52 2014
Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.
connected to target database: RACDB (not mounted)
RMAN> restore standby controlfile from '/rman_backup/standby_0tpk53jq_1_1.ctl';
Starting restore at 2014-10-03 19:06:55
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=16 device type=DISK
channel ORA_DISK_1: restoring control file
channel ORA_DISK_1: restore complete, elapsed time: 00:00:25
output file name=+DATA/phydb/controlfile/cont.ctl
Finished restore at 2014-10-03 19:07:22
RMAN>
SQL> alter database mount;
run {
allocate channel c1 type disk;
allocate channel c2 type disk;
allocate channel c3 type disk;
allocate channel c4 type disk;
restore database ;
release channel c1;
release channel c2;
release channel c3;
release channel c4;
}
RMAN> restore database ;
Starting restore at 2014-10-03 19:13:48
Starting implicit crosscheck backup at 2014-10-03 19:13:48
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=32 device type=DISK
Crosschecked 6 objects
Finished implicit crosscheck backup at 2014-10-03 19:13:49
Starting implicit crosscheck copy at 2014-10-03 19:13:49
using channel ORA_DISK_1
Finished implicit crosscheck copy at 2014-10-03 19:13:49
searching for all files in the recovery area
cataloging files...
no files cataloged
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00001 to +DATA/phydb/datafile/system.256.859849357
channel ORA_DISK_1: restoring datafile 00004 to +DATA/phydb/datafile/users.259.859849357
channel ORA_DISK_1: restoring datafile 00005 to +DATA/phydb/datafile/example.264.859849549
channel ORA_DISK_1: reading from backup piece /rman_backup/FULL_0npk5395_1_1.bak
channel ORA_DISK_1: piece handle=/rman_backup/FULL_0npk5395_1_1.bak tag=TAG20141003T164524
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:04:23
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00002 to +DATA/phydb/datafile/sysaux.257.859849357
channel ORA_DISK_1: restoring datafile 00003 to +DATA/phydb/datafile/undotbs1.258.859849357
channel ORA_DISK_1: restoring datafile 00006 to +DATA/phydb/datafile/undotbs2.265.859849971
channel ORA_DISK_1: reading from backup piece /rman_backup/FULL_0opk53c4_1_1.bak
channel ORA_DISK_1: piece handle=/rman_backup/FULL_0opk53c4_1_1.bak tag=TAG20141003T164524
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:01:45
Finished restore at 2014-10-03 19:20:01
RMAN>
SQL> select name from v$datafile
2 union
3 select name from v$controlfile
4 union
5 select member from v$logfile;
NAME
--------------------------------------------------------------------------------
+DATA/phydb/controlfile/cont.ctl
+DATA/phydb/datafile/example.258.860008433
+DATA/phydb/datafile/sysaux.260.860008695
+DATA/phydb/datafile/system.257.860008433
+DATA/phydb/datafile/undotbs1.261.860008697
+DATA/phydb/datafile/undotbs2.262.860008697
+DATA/phydb/datafile/users.259.860008435
+DATA/phydb/onlinelog/group_1.261.859849493
+DATA/phydb/onlinelog/group_2.262.859849495
+DATA/phydb/onlinelog/group_3.266.859850179
+DATA/phydb/onlinelog/group_4.267.859850183
+FRA/racdb/onlinelog/group_1.257.859849493
+FRA/racdb/onlinelog/group_2.258.859849495
+FRA/racdb/onlinelog/group_3.259.859850181
+FRA/racdb/onlinelog/group_4.260.859850187
15 rows selected.
SQL>
创建standbylogfile，这里的大小应根据主库的v$log中的bytes列来计算：
alter database add standby logfile thread 1 group 5 size 50M ,group 6 size 50M ,group 7 size 50M ;
alter database add standby logfile thread 2 group 8 size 50M ,group 9 size 50M ,group 10 size 50M ;
创建完成后查询：
select * from v$standby_log;
SQL> set linesize 1200
SQL> select * from v$standby_log;
GROUP# DBID THREAD# SEQUENCE# BYTES BLOCKSIZE USED ARC STATUS FIRST_CHANGE# FIRST_TIM NEXT_CHANGE# NEXT_TIME LAST_CHANGE# LAST_TIME
---------- ---------------------------------------- ---------- ---------- ---------- ---------- ---------- --- ---------- ------------- --------- ------------ --------- ------------ ---------
5 UNASSIGNED 1 0 52428800 512 0 NO UNASSIGNED
6 UNASSIGNED 1 0 52428800 512 0 NO UNASSIGNED
7 UNASSIGNED 1 0 52428800 512 0 YES UNASSIGNED
8 UNASSIGNED 2 0 52428800 512 0 NO UNASSIGNED
9 UNASSIGNED 2 0 52428800 512 0 NO UNASSIGNED
10 UNASSIGNED 2 0 52428800 512 0 YES UNASSIGNED
6 rows selected.
SQL>
SQL> col member for a50
SQL> select * from v$logfile;
GROUP# STATUS TYPE MEMBER IS_
---------- ------- ------- -------------------------------------------------- ---
2 ONLINE +DATA/phydb/onlinelog/group_2.270.860171617 NO
2 ONLINE +FRA/phydb/onlinelog/group_2.294.860171767 YES
1 ONLINE +DATA/phydb/onlinelog/group_1.269.860171391 NO
1 ONLINE +FRA/phydb/onlinelog/group_1.293.860171601 YES
3 ONLINE +DATA/phydb/onlinelog/group_3.271.860171811 NO
3 ONLINE +FRA/phydb/onlinelog/group_3.295.860171893 YES
4 ONLINE +DATA/phydb/onlinelog/group_4.272.860171905 NO
4 ONLINE +FRA/phydb/onlinelog/group_4.296.860172007 YES
5 STANDBY +DATA/phydb/onlinelog/group_5.263.860014755 NO
5 STANDBY +FRA/phydb/onlinelog/group_5.256.860014769 YES
6 STANDBY +DATA/phydb/onlinelog/group_6.264.860014775 NO
6 STANDBY +FRA/phydb/onlinelog/group_6.257.860014789 YES
7 STANDBY +DATA/phydb/onlinelog/group_7.265.860014795 NO
7 STANDBY +FRA/phydb/onlinelog/group_7.258.860014811 YES
8 STANDBY +DATA/phydb/onlinelog/group_8.266.860090543 NO
8 STANDBY +FRA/phydb/onlinelog/group_8.259.860090581 YES
9 STANDBY +DATA/phydb/onlinelog/group_9.267.860090607 NO
9 STANDBY +FRA/phydb/onlinelog/group_9.260.860090659 YES
10 STANDBY +DATA/phydb/onlinelog/group_10.268.860090669 NO
10 STANDBY +FRA/phydb/onlinelog/group_10.261.860090715 YES
20 rows selected.
SQL>
这里可能会报一个错误：
ORA-15041: diskgroup "DATA" space exhausted
就是磁盘不足，具体大家上网搜吧，解决办法就是添加磁盘，不详解了：
alter diskgroup DATA add disk '/dev/raw/raw5';
[oracle@rac2 ~]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.3.0 Production on Fri Oct 10 15:59:46 2014
Copyright (c) 1982, 2011, Oracle. All rights reserved.
Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production
With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP,
Data Mining and Real Application Testing options
SQL> show parameter spfile
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
spfile string +DATA/racdb/spfileracdb.ora
SQL>
--rac1
alter system set log_archive_dest_2='service=phydb lgwr sync valid_for=(online_logfiles,primary_role) db_unique_name=phydb' sid='*';
alter system set log_archive_config='dg_config=(racdb,phydb)' sid='*';
配置后的结果：
SQL> show parameter log_archive_dest_2
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
log_archive_dest_2 string service=phydb lgwr sync valid_
for=(online_logfiles,primary_r
ole) db_unique_name=phydb
SQL> show parameter log_archive_config
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
log_archive_config string dg_config=(racdb,phydb)
SQL>
SQL> select sequence#,name,applied from v$archived_log;
SEQUENCE# NAME APPLIED
---------- ---------------------------------------------------------------------------------------------------- ---------
39 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_39.327.860599799 NO
40 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_40.328.860599799 NO
41 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_41.329.860599953 NO
49 +FRA/phydb/archivelog/2014_10_10/thread_1_seq_49.330.860600751 NO
50 +FRA/phydb/archivelog/2014_10_10/thread_1_seq_50.331.860600755 NO
51 +FRA/phydb/archivelog/2014_10_10/thread_1_seq_51.332.860600869 NO
42 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_42.333.860601127 NO
43 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_43.334.860601131 NO
44 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_44.335.860601415 NO
9 rows selected.
SQL>
这里如果备库无法顺利接收主库日志，则主库的alert日志中有如下提示：
------------------------------------------------------------
Check that the primary and standby are using a password file
and remote_login_passwordfile is set to SHARED or EXCLUSIVE,
and that the SYS password is same in the password files.
returning error ORA-16191
------------------------------------------------------------
PING[ARC2]: Heartbeat failed to connect to standby 'phydb'. Error is 16191.
Process J000 died, see its trace file
kkjcre1p: unable to spawn jobq slave process
Errors in file /u01/app/oracle/diag/rdbms/racdb/racdb2/trace/racdb2_cjq0_14236.trc:
Process NSS2 died, see its trace file
Process J000 died, see its trace file
kkjcre1p: unable to spawn jobq slave process
Errors in file /u01/app/oracle/diag/rdbms/racdb/racdb2/trace/racdb2_cjq0_14236.trc:
Error 443 for archive log file 3 to 'phydb'
LGWR: Failed to archive log 3 thread 2 sequence 25 (443)
Process O000 died, see its trace file
该节点的用户口令和备库的口令不一致，最好的办法是生成一份口令文件之后，然后将该口令文件拷贝到其它节点，重新生成口令有的时候也报错的
---- 3 个节点
orapwd file=/u01/app/oracle/product/11.2.0/dbhome_1/dbs/orapwracdb2 password=lhr force=y
SQL> alter database recover managed standby database using current logfile disconnect from session;
Database altered.
过几分钟后：
SQL> select sequence#,name,applied from v$archived_log;
SEQUENCE# NAME APPLIED
---------- ---------------------------------------------------- ---------
39 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_39.327.860599799 YES
40 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_40.328.860599799 YES
41 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_41.329.860599953 YES
49 +FRA/phydb/archivelog/2014_10_10/thread_1_seq_49.330.860600751 YES
50 +FRA/phydb/archivelog/2014_10_10/thread_1_seq_50.331.860600755 YES
51 +FRA/phydb/archivelog/2014_10_10/thread_1_seq_51.332.860600869 YES
42 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_42.333.860601127 NO
43 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_43.334.860601131 NO
44 +FRA/phydb/archivelog/2014_10_10/thread_2_seq_44.335.860601415 NO
alter database recover managed standby database cancel;
alter database open;
alter database recover managed standby database using current logfile disconnect from session;
SQL> col name for a30
SQL> select dbid,name,current_scn,protection_mode,database_role,force_logging,open_mode,switchover_status from v$database;
DBID NAME CURRENT_SCN PROTECTION_MODE DATABASE_ROLE FOR OPEN_MODE SWITCHOVER_STATUS
---------- ------------------------------ ----------- -------------------- ---------------- --- -------------------- --------------------
857466254 RACDB 1871595 MAXIMUM PERFORMANCE PHYSICAL STANDBY YES READ ONLY WITH APPLY NOT ALLOWED
SQL>
至此，我们已经顺利的给RAC主库配置ActivePhysicalDataguard成功！
注意：11g以前的版本在执行复制命令dupilicate时，必须在备库上存在有备份文件，即在主库上备份完成后将文件拷到备库相同目录下
但是在11g一切都变的简单，只需要在复制命令后加上FROM ACTIVE DATABASE参数即可。
例如为实现数据库的文件回复，可以执行命令：
$ rman target / auxiliary sys/herenit@racheren_standby
RMAN>duplicate target database for standby from active database;
如果原目录下有对应文件已经删除，但是数据字典中存在，可以使用nofilenamecheck参数取消对文件名的检查
RMAN>duplicate target database for standby from active database nofilenamecheck;
恢复完成以后，检查备库状态
SQL> select status from v$instance;
STATUS$dEQ5b/C#x22494861------------ITPUB个人空间 T c:v_y!RMOUNTED
SQL> select open_mode from v$database;
OPEN_MODEITPUB个人空间FU?tN2FE5v:_wU}----------ITPUB个人空间1R_G@5k/kMOUNTED
SQL> select member from v$logfile;
SQL> select name from v$datafile;
SQL> select name from v$tempfile;
DATA Guard在最大保护和最高可用性模式下，Standby数据库必须配置 Standby Redo Log
备库创建并配置 Standby Redo Log 是为了能够接收到主库传过来的日志，用作恢复.
主库创建并配置 Standby Redo Log 是为了在主备切换之后，能够接收到原备库（切换之后的主库）传过来的日志
注意：为了主备切换后能正常，主备库都必须创建standby redo log
创建原则：建议Standby Redologs日志组数量基于Primary数据库的线程数来确定（这里的线程数可以理解为RAC环境中的节点数）。
有一个推荐的公式可供参考：（每线程的日志组数+1）×最大线程数。
使用这个公式可以降低Primary数据库实例LGWR进程锁住的可能性。
创建原则和单实例一样，大小相等，但是日志组数量要比primary数据库多一组，比如在我的环境中，有2个节点，每个节点有4组redo，那么现在要创建的standby redo log组数为：（4+1）*2=10个
Standby Redologs的操作方式与Online Redologs几乎一模一样，只不过在创建或删除时需要多指定一个Standby关键字
SQL>alter database add standby logfile thread 1 group 5 '+ORAFLASH/racheren/onlinelog/group_5.log' size 50M;
SQL>alter database add standby logfile thread 2 group 6 '+ORAFLASH/racheren/onlinelog/group_6.log' size 50M;
单实例上不需要thread标识
SQL>alter database add standby logfile group 6 '/oradata/racheren/onlinelog/group_6.log' size 50M;
删除方式：
SQL>alter database drop standby logfile grop 6
查询方式：
SQL>select group#,thread#,sequence#,archived,status from v$standby_log;
ERROR at line 1:
ORA-01156: recovery or flashback in progress may need access to files
--在备库添加standby redo log需要先停MRP
SQL> alter database recover managed standby database cancel;
select tablespace_name, file_name from dba_data_files;
create tablespace dg_tbs datafile size 5m;
drop tablespace dg_tbs including contents and datafiles;
①主库创建测试表：
② 备库查看测试表同步：
Create table tmp_test(id number);
Insert into tmp_test values(1);
Commit;
Select * from tmp_test;
即rac主库切换为物理备库，物理备库切换为rac主库。
首先需要确保主库采用spfile启动，然后
Rac1上：
alter system set fal_client='rac1' sid='rac1';
Rac2上：
alter system set fal_client='rac2' sid='rac2';
alter system set fal_server='phydb' sid='*';
我们在微信上24小时期待你的声音
解答本文疑问/技术咨询/运营咨询/技术建议/互联网交流
微信二维码
移动版官网