» Dataguard启动的各个阶段 OracleBlog.cn -天堂向左 DBA向右

Dataguard启动的各个阶段

· 作者小荷||【转载时请务必以超链接形式标明文章原始出处和作者信息】
· 永久链接: http://www.oracleblog.cn/study-note/different-stage-on-dataguard-startup/

主库启动instance和lsnr，在备库启动lsnr后，在dg启动的各个阶段，dg都做了什么呢：启动了那些进程？什么时候会传归档？什么时候开始应用归档？

1.备库关闭时。
1.1 检查备库进程：

[oracle@standby1 arch]$ sqlplus "/ as sysdba"

SQL*Plus: Release 9.2.0.4.0 - Production on Fri Jan 11 23:52:25 2008

Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.

Connected to an idle instance.

>
> !ps -ef |grep ora_
oracle 3917 3915 0 23:52 pts/1 00:00:00 /bin/bash -c ps -ef |grep ora_
oracle 3919 3917 0 23:52 pts/1 00:00:00 grep ora_

1.2 备库arch目录：

[oracle@standby1 arch]$ ll
total 328
-rw-r----- 1 oracle dba 1536 Jan 10 01:04 oradg_0001_0000000040.arc
-rw-r----- 1 oracle dba 1536 Jan 10 01:04 oradg_0001_0000000041.arc
-rw-r----- 1 oracle dba 1024 Jan 10 01:04 oradg_0001_0000000042.arc
-rw-r----- 1 oracle dba 302080 Jan 10 01:11 oradg_0001_0000000043.arc

1.3 此时在主库switch logfile，arch不会传到备库，在主库上检查v$archived_log，也发现日志不会applied：

SQL> select SEQUENCE#,DEST_ID,ARCHIVED,APPLIED,DELETED,STATUS from v$archived_log order by SEQUENCE#
2 /

SEQUENCE# DEST_ID ARCHIV APPLIE DELETE ST
---------- ---------- ------ ------ ------ --
   20 2 YES YES NO A
   21 2 YES YES NO A
   22 2 YES YES NO A
   23 2 YES YES NO A
   24 2 YES YES NO A
   25 2 YES YES NO A
   26 2 YES YES NO A
   27 2 YES YES NO A
   28 0 YES YES NO A
   29 0 YES YES NO A
   30 0 YES YES NO A
   31 0 YES YES NO A
   32 0 YES YES NO A
   33 0 YES YES NO A
   34 0 YES YES NO A
   35 0 YES YES NO A
   36 0 YES YES NO A
   37 0 YES YES NO A
   38 0 YES YES NO A
   39 0 YES YES NO A
   40 0 YES YES NO A
   41 2 YES YES NO A
   42 2 YES YES NO A
   43 2 YES YES NO A
   44 1 YES NO NO A
   45 1 YES NO NO A
   46 1 YES NO NO A

27 rows selected.

1.4 备库alertlog只有上次的信息：

……
ARC0: Archival stopped
Sat Jan 12 00:39:57 2008
ARCH: Archiving is disabled
Shutting down archive processes
Archiving is disabled
Archive process shutdown avoided: 0 active

2.备库startup nomount
2.1 检查备库进程：

> startup nomount;
ORACLE instance started.

Total System Global Area 370218244 bytes
Fixed Size 451844 bytes
Variable Size 167772160 bytes
Database Buffers 201326592 bytes
Redo Buffers 667648 bytes
> !ps -ef |grep ora_
oracle 4251 1 1 00:42 ? 00:00:00 ora_pmon_oradg
oracle 4253 1 0 00:42 ? 00:00:00 ora_dbw0_oradg
oracle 4255 1 0 00:42 ? 00:00:00 ora_lgwr_oradg
oracle 4257 1 1 00:42 ? 00:00:00 ora_ckpt_oradg
oracle 4259 1 1 00:42 ? 00:00:00 ora_smon_oradg
oracle 4261 1 1 00:42 ? 00:00:00 ora_reco_oradg
oracle 4263 1 1 00:42 ? 00:00:00 ora_cjq0_oradg
oracle 4265 1 0 00:42 ? 00:00:00 ora_qmn0_oradg
oracle 4267 1 0 00:42 ? 00:00:00 ora_s000_oradg
oracle 4269 1 1 00:42 ? 00:00:00 ora_d000_oradg
oracle 4271 1 1 00:42 ? 00:00:00 ora_arc0_oradg
oracle 4273 1 0 00:42 ? 00:00:00 ora_arc1_oradg
oracle 4275 4245 0 00:42 pts/1 00:00:00 /bin/bash -c ps -ef |grep ora_

这些进程的数量和主库一致。

2.2 备库arch目录：

发现备库的arch没有增加，也就是说，arch没有被传输到备库。

2.3 检查主库上检查v$archived_log，发现还是没有applied日志（连传都没有传过去，当然不会applied）：

SQL> /

SEQUENCE# DEST_ID ARCHIV APPLIE DELETE ST
---------- ---------- ------ ------ ------ --
   24 2 YES YES NO A
   25 2 YES YES NO A
   26 2 YES YES NO A
   27 2 YES YES NO A
   28 0 YES YES NO A
   29 0 YES YES NO A
   30 0 YES YES NO A
   31 0 YES YES NO A
   32 0 YES YES NO A
   33 0 YES YES NO A
   34 0 YES YES NO A
   35 0 YES YES NO A
   36 0 YES YES NO A
   37 0 YES YES NO A
   38 0 YES YES NO A
   39 0 YES YES NO A
   40 0 YES YES NO A
   41 2 YES YES NO A
   42 2 YES YES NO A
   43 2 YES YES NO A
   44 1 YES NO NO A
   45 1 YES NO NO A
   46 1 YES NO NO A
   47 1 YES NO NO A
   48 1 YES NO NO A
   49 1 YES NO NO A
   50 1 YES NO NO A
   51 1 YES NO NO A
   52 1 YES NO NO A

2.4 备库alertlog上却出现了报错信息：

Sat Jan 12 00:41:43 2008
Starting ORACLE instance (normal)
LICENSE_MAX_SESSION = 0
LICENSE_SESSIONS_WARNING = 0
SCN scheme 2
Using log_archive_dest parameter default value
LICENSE_MAX_USERS = 0
SYS auditing is disabled
Starting up ORACLE RDBMS Version: 9.2.0.4.0.
System parameters with non-default values:
processes = 150
timed_statistics = TRUE
shared_pool_size = 67108864
large_pool_size = 50331648
java_pool_size = 33554432
control_files = /oracle/oradata/oradg/control01.ctl, /oracle/oradata/oradg/control02.ctl, /oracle/oradata/oradg/control03.ctl
db_block_size = 8192
db_cache_size = 201326592
compatible = 9.2.0.0.0
log_archive_start = TRUE
log_archive_dest_1 = LOCATION=/oracle/arch/
log_archive_dest_2 = SERVICE=standby optional
standby_archive_dest = /oracle/arch
log_archive_format = oradg_%T_%S.arc
db_file_multiblock_read_count= 16
fast_start_mttr_target = 300
undo_management = AUTO
undo_tablespace = UNDOTBS1
undo_retention = 10800
remote_login_passwordfile= EXCLUSIVE
db_domain =
instance_name = oradg
dispatchers = (PROTOCOL=TCP) (SERVICE=oradgXDB)
job_queue_processes = 10
hash_join_enabled = TRUE
background_dump_dest = /oracle/admin/oradg/bdump
user_dump_dest = /oracle/admin/oradg/udump
core_dump_dest = /oracle/admin/oradg/cdump
sort_area_size = 524288
db_name = oradg
open_cursors = 300
star_transformation_enabled= FALSE
query_rewrite_enabled = FALSE
pga_aggregate_target = 82837504
aq_tm_processes = 1
PMON started with pid=2
DBW0 started with pid=3
LGWR started with pid=4
CKPT started with pid=5
SMON started with pid=6
CJQ0 started with pid=8
RECO started with pid=7
Sat Jan 12 00:41:44 2008
starting up 1 shared server(s) ...
QMN0 started with pid=9
Sat Jan 12 00:41:44 2008
starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
ARCH: STARTING ARCH PROCESSES
Sat Jan 12 00:42:53 2008
ARCH: STARTING ARCH PROCESSES COMPLETE
ARC1 started with pid=13
ARC1: Archival started
ARC1: Thread not mounted
ARC0 started with pid=12
ARC0: Archival started
ARC0: Thread not mounted
Sat Jan 12 00:43:41 2008
Restarting dead background process QMN0
QMN0 started with pid=9
destination database instance is 'started' not 'mounted'
Sat Jan 12 00:49:08 2008
Restarting dead background process QMN0
QMN0 started with pid=15
Sat Jan 12 00:50:47 2008
RFS: Possible network disconnect with primary database
destination database instance is 'started' not 'mounted'
Sat Jan 12 00:50:47 2008
destination database instance is 'started' not 'mounted'
RFS: Possible network disconnect with primary database
Sat Jan 12 00:51:13 2008
RFS: Possible network disconnect with primary database
destination database instance is 'started' not 'mounted'
Sat Jan 12 00:51:13 2008
destination database instance is 'started' not 'mounted'
Sat Jan 12 00:51:13 2008
RFS: Possible network disconnect with primary database
Sat Jan 12 00:51:13 2008
RFS: Possible network disconnect with primary database
Sat Jan 12 00:51:14 2008
destination database instance is 'started' not 'mounted'-------------注意这里附近
Sat Jan 12 00:51:14 2008
destination database instance is 'started' not 'mounted'
RFS: Possible network disconnect with primary database

这里的destination database是指备机本身，这个报错信息会定时出现，即每当在主库在switch logfile的时候出现。
RFS服务是指Remote File Server服务，负责传输arch日志。此时RFS服务报错，因此arch也无法传输过来。

3.备库到mount状态。
3.1 检查备库进程：

> alter database mount standby database;

Database altered.

Elapsed: 00:00:04.23
> !ps -ef |grep ora_
oracle 4251 1 0 00:47 ? 00:00:01 ora_pmon_oradg
oracle 4253 1 0 00:47 ? 00:00:00 ora_dbw0_oradg
oracle 4255 1 0 00:47 ? 00:00:00 ora_lgwr_oradg
oracle 4257 1 0 00:47 ? 00:00:00 ora_ckpt_oradg
oracle 4259 1 0 00:47 ? 00:00:00 ora_smon_oradg
oracle 4261 1 0 00:47 ? 00:00:00 ora_reco_oradg
oracle 4263 1 0 00:47 ? 00:00:00 ora_cjq0_oradg
oracle 4267 1 0 00:47 ? 00:00:00 ora_s000_oradg
oracle 4269 1 0 00:47 ? 00:00:00 ora_d000_oradg
oracle 4271 1 0 00:47 ? 00:00:00 ora_arc0_oradg
oracle 4273 1 0 00:47 ? 00:00:00 ora_arc1_oradg
oracle 4452 4245 0 01:04 pts/1 00:00:00 /bin/bash -c ps -ef |grep ora_

发现进程数还是和原来一样。

3.2 备库arch目录：

[oracle@standby1 arch]$ ll
[oracle@standby1 arch]$ ll
total 4908
-rw-r----- 1 oracle dba 1536 Jan 10 01:04 oradg_0001_0000000040.arc
-rw-r----- 1 oracle dba 1536 Jan 10 01:04 oradg_0001_0000000041.arc
-rw-r----- 1 oracle dba 1024 Jan 10 01:04 oradg_0001_0000000042.arc
-rw-r----- 1 oracle dba 302080 Jan 10 01:11 oradg_0001_0000000043.arc
-rw-r----- 1 oracle dba 889856 Jan 12 00:04 oradg_0001_0000000044.arc
-rw-r----- 1 oracle dba 35328 Jan 12 00:04 oradg_0001_0000000045.arc
-rw-r----- 1 oracle dba 1536 Jan 12 00:04 oradg_0001_0000000046.arc
……
-rw-r----- 1 oracle dba 34816 Jan 12 01:10 oradg_0001_0000000092.arc
-rw-r----- 1 oracle dba 1024 Jan 12 01:10 oradg_0001_0000000093.arc
-rw-r----- 1 oracle dba 1536 Jan 12 01:10 oradg_0001_0000000094.arc
-rw-r----- 1 oracle dba 1024 Jan 12 01:10 oradg_0001_0000000095.arc
-rw-r----- 1 oracle dba 1024 Jan 12 01:10 oradg_0001_0000000096.arc

发现备库的arch已经传输过来。

3.3 检查主库上检查v$archived_log，发现文件虽然已经传到备库，还是没有applied日志（MRP进程没起来）：

SQL> select SEQUENCE#,DEST_ID,ARCHIVED,APPLIED,DELETED,STATUS from v$archived_log order by SEQUENCE#
2 /

SEQUENCE# DEST_ID ARCHIV APPLIE DELETE ST
---------- ---------- ------ ------ ------ --
   24 2 YES YES NO A
   25 2 YES YES NO A
   26 2 YES YES NO A
   27 2 YES YES NO A
   28 0 YES YES NO A
   29 0 YES YES NO A
   30 0 YES YES NO A
   31 0 YES YES NO A
   32 0 YES YES NO A
   33 0 YES YES NO A
   34 0 YES YES NO A
   35 0 YES YES NO A
   36 0 YES YES NO A
   37 0 YES YES NO A
   38 0 YES YES NO A
   39 0 YES YES NO A
   40 0 YES YES NO A
   41 2 YES YES NO A
   42 2 YES YES NO A
   43 2 YES YES NO A
   44 1 YES NO NO A
   44 2 YES NO NO A
   45 1 YES NO NO A
   45 2 YES NO NO A
   46 1 YES NO NO A
   46 2 YES NO NO A
   47 1 YES NO NO A
……
   94 2 YES NO NO A
   95 1 YES NO NO A
   95 2 YES NO NO A
   96 1 YES NO NO A
   96 2 YES NO NO A

3.4 备库alertlog的信息：

alter database mount standby database
Sat Jan 12 01:09:29 2008
Successful mount of redo thread 1, with mount id 2634673925.
Sat Jan 12 01:09:29 2008
Standby Database mounted.
Completed: alter database mount standby database

基本就一个启动到mount的信息，没别的。

4. 备库启动到recover managed状态

> alter database recover managed standby database disconnect from session;

Database altered.

Elapsed: 00:00:06.17
> !ps -ef |grep ora_
oracle 4528 1 0 01:11 ? 00:00:00 ora_pmon_oradg
oracle 4530 1 0 01:11 ? 00:00:00 ora_dbw0_oradg
oracle 4532 1 0 01:11 ? 00:00:00 ora_lgwr_oradg
oracle 4534 1 0 01:11 ? 00:00:00 ora_ckpt_oradg
oracle 4536 1 0 01:11 ? 00:00:00 ora_smon_oradg
oracle 4538 1 0 01:11 ? 00:00:00 ora_reco_oradg
oracle 4540 1 0 01:11 ? 00:00:00 ora_cjq0_oradg
oracle 4544 1 0 01:11 ? 00:00:00 ora_s000_oradg
oracle 4546 1 0 01:11 ? 00:00:00 ora_d000_oradg
oracle 4548 1 0 01:11 ? 00:00:00 ora_arc0_oradg
oracle 4550 1 0 01:11 ? 00:00:00 ora_arc1_oradg
oracle 5064 1 5 01:18 ? 00:00:02 ora_mrp0_oradg<-------注意此处，mrp进程已经起来
oracle 5065 4245 0 01:18 pts/1 00:00:00 /bin/bash -c ps -ef |grep ora_

发现mrp进程已经起来。

4.2 备库arch目录：

还是原来的。

4.3 检查主库上检查v$archived_log，发现归档已经applied：

SQL> /

SEQUENCE# DEST_ID ARCHIV APPLIE DELETE ST
---------- ---------- ------ ------ ------ --
   24 2 YES YES NO A
   25 2 YES YES NO A
   26 2 YES YES NO A
   27 2 YES YES NO A
   28 0 YES YES NO A
   29 0 YES YES NO A
   30 0 YES YES NO A
   31 0 YES YES NO A
   32 0 YES YES NO A
   33 0 YES YES NO A
   34 0 YES YES NO A
   35 0 YES YES NO A
   36 0 YES YES NO A
   37 0 YES YES NO A
   38 0 YES YES NO A
……
   92 1 YES NO NO A
   92 2 YES NO NO A
   93 1 YES NO NO A
   93 2 YES NO NO A
   94 1 YES NO NO A
   94 2 YES NO NO A
   95 1 YES NO NO A
   95 2 YES NO NO A
   96 1 YES NO NO A
   96 2 YES YES NO A

126 rows selected.

4.4 备库alertlog的信息：

……
alter database recover managed standby database disconnect from session
Sat Jan 12 01:18:09 2008
Attempt to start background Managed Standby Recovery process
MRP0 started with pid=16
MRP0: Background Managed Standby Recovery process started
Starting datafile 1 recovery in thread 1 sequence 44
Datafile 1: '/oracle/oradata/oradg/system01.dbf'
Starting datafile 2 recovery in thread 1 sequence 44
Datafile 2: '/oracle/oradata/oradg/undotbs01.dbf'
Starting datafile 3 recovery in thread 1 sequence 44
Datafile 3: '/oracle/oradata/oradg/cwmlite01.dbf'
Starting datafile 4 recovery in thread 1 sequence 44
Datafile 4: '/oracle/oradata/oradg/drsys01.dbf'
Starting datafile 5 recovery in thread 1 sequence 44
Datafile 5: '/oracle/oradata/oradg/example01.dbf'
Starting datafile 6 recovery in thread 1 sequence 44
……
Datafile 9: '/oracle/oradata/oradg/users01.dbf'
Starting datafile 10 recovery in thread 1 sequence 44
Datafile 10: '/oracle/oradata/oradg/xdb01.dbf'
Starting datafile 11 recovery in thread 1 sequence 44
Datafile 11: '/oracle/oradata/oradg/ts_mssm_01.dbf'
Media Recovery Log /oracle/arch/oradg_0001_0000000044.arc
Sat Jan 12 01:18:15 2008
Completed: alter database recover managed standby database di
Sat Jan 12 01:18:15 2008
Media Recovery Log /oracle/arch/oradg_0001_0000000045.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000046.arc
……
Media Recovery Log /oracle/arch/oradg_0001_0000000089.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000090.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000091.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000092.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000093.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000094.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000095.arc
Media Recovery Log /oracle/arch/oradg_0001_0000000096.arc
Media Recovery Waiting for thread 1 seq# 97
Sat Jan 12 01:22:05 2008

总结：备库在起来时：

1.nomount：rfs服务没有起，因此主库文件不会传到备库；mrp进程没有起，因此也不会做recover，归档不会applied。
2.mount：rfs服务启动，主库的归档能传输到备库（可以在主库switchlogfile几次，会发现归档已经传输过去）；mrp进程没有启动，归档不applied。
3.recover managed：rfs在上一步已经启动，归档能传输到备库；且mrp进程已经起来，apply 归档日志！

· 【文章发布信息】发表于： 2008-01-12 @ 01:36:26 · ||分类： Study note

OracleBlog.cn -天堂向左 DBA向右

Dataguard启动的各个阶段

留条评论

ACOUG Member

分类

标签云

搜索

随机文章

最新回复

归档

订阅和更新

网站管理

访客位置分布

知识共享许可