Oracle启动过程简析__oracle_localdomain_localhost_3939_started_

当前位置: 数据库>oracle

Oracle启动过程简析

来源: 互联网发布时间：2017-06-15

本文导语: Oracle数据库启动时，经历了三个过程：(用命名如下）startup nomount;alter database mount; alter database open;当然数据库关闭时也是经历了三个相反的过程：alter database close;alter database dismount;shutdown;（shutdown 后面跟了四个参数：norma...

Oracle数据库启动时，经历了三个过程：(用命名如下）
startup nomount;
alter database mount; alter database open;
当然数据库关闭时也是经历了三个相反的过程：
alter database close;
alter database dismount;
shutdown;（shutdown 后面跟了四个参数：normal;immediate;transactional;abort)
四种方式关闭数据库的比较：

(NO YES)

关闭方式 A I T N 允许新的连接 N N N N 等待活动会话终止 N N N Y 等待活动事务终止 N N

Y 强制进行checkpoint,关闭所有文件 N Y Y Y

分析第一个过程startup nomount:
这个过程数据库首先到参数文件（pfile/spfile)中读取数据库的设置，创建实例.
数据库所在的操作系统版本：
[oracle@localhost ~]$ lsb_release -a
LSB Version: :core-3.1-ia32:core-3.1-noarch:graphics-3.1-ia32:graphics-3.1-noarch
Distributor ID: EnterpriseEnterpriseServer
Description: Enterprise Linux Enterprise Linux Server release 5.5 (Carthage)
Release: 5.5
Codename: Carthage

数据库版本：
SQL> SELECT * FROM v$version where rownum=1;

BANNER
--------------------------------------------------------------------------------
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production

因为spfile是二进制文件，不能直接读取，在linux中，可以用命令String转储出来：
[oracle@localhost dbs]$ strings spfileorcl3939.ora
orcl3939.__db_cache_size=54525952
orcl3939.__java_pool_size=4194304
orcl3939.__large_pool_size=4194304
orcl3939.__oracle_base='/u01/app/oracle'#ORACLE_BASE set from environment
orcl3939.__pga_aggregate_target=171966464
orcl3939.__sga_target=251658240
orcl3939.__shared_io_pool_size=0
orcl3939.__shared_pool_size=176160768
orcl3939.__streams_pool_size=4194304
*.audit_file_dest='/u01/app/oracle/admin/orcl3939/adump'
*.audit_trail='db'
*.compatible='11.2.0.0.0'
*.control_files='/u01/app/o
racle/oradata/orcl3939/control01.ctl','/u01/app/oracle/flash_recovery_area/orcl3939/control02.ctl','/u01/app/oracle/oradata/orcl3939/control03.ctl'
*.db_block_size=8192
*.db_domain='localdomain'
*.db_name='orcl3939'
*.db_recovery_file_dest='/u01/app/oracle/flash_recovery_area'
*.db_recovery_file_dest_size=4039114752
*.diagnostic_dest='/u01/app/oracle'
*.dispatchers='(PROTOCOL=TCP) (SERVICE=orcl3939XDB)'
*.instance_name='ORCL3939'
*.local_listener='(ADDRESS=(PROTOCOL=TCP)(HOST = local
host.localdomain)(PORT = 1521))'
*.memory_target=423624704
*.open_cursors=300
*.processes=150
*.remote_login_passwordfile='EXCLUSIVE'
*.service_names='a,b,c,d'
*.trace_enabled=TRUE
*.undo_tablespace='UNDOTBS1'

spfile文件中你可以看到数据库在nomount时做了些什么，根据参数文件的内容，创建了instance,分配了相应的内存区域，启动了相应的后台进程。
我们再看告警日志文件（alert_.log）：读取了参数文件，启动了实例

Starting up:
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options.
Using parameter settings in server-side spfile /u01/app/oracle/product/11.2.0/dbhome_1/dbs/spfileorcl3939.ora
System parameters with non-default values:
processes = 150
memory_target = 404M
control_files = "/u01/app/oracle/oradata/orcl3939/control01.ctl"
control_files = "/u01/app/oracle/flash_recovery_area/orcl3939/control02.ctl"
control_files = "/u01/app/oracle/oradata/orcl3939/control03.ctl"
db_block_size = 8192
compatible = "11.2.0.0.0"
db_recovery_file_dest = "/u01/app/oracle/flash_recovery_area"
db_recovery_file_dest_size= 3852M
undo_tablespace = "UNDOTBS1"
remote_login_passwordfile= "EXCLUSIVE"
db_domain = "localdomain"
instance_name = "ORCL3939"
service_names = "a,b,c,d"
dispatchers = "(PROTOCOL=TCP) (SERVICE=orcl3939XDB)"
local_listener = "(ADDRESS=(PROTOCOL=TCP)(HOST = localhost.localdomain)(PORT = 1521))"
audit_file_dest = "/u01/app/oracle/admin/orcl3939/adump"
audit_trail = "DB"
db_name = "orcl3939"
open_cursors = 300
diagnostic_dest = "/u01/app/oracle"
trace_enabled = TRUE
Thu Apr 02 14:59:41 2015
PMON started with pid=2, OS id=5989
Thu Apr 02 14:59:41 2015
VKTM started with pid=3, OS id=5991 at elevated priority
VKTM running at (10)millisec precision with DBRM quantum (100)ms
Thu Apr 02 14:59:41 2015
GEN0 started with pid=4, OS id=5995
Thu Apr 02 14:59:41 2015
DIAG started with pid=5, OS id=5997
Thu Apr 02 14:59:41 2015
DBRM started with pid=6, OS id=5999
Thu Apr 02 14:59:41 2015
PSP0 started with pid=7, OS id=6001
Thu Apr 02 14:59:41 2015
DIA0 started with pid=8, OS id=6003
Thu Apr 02 14:59:41 2015
MMAN started with pid=9, OS id=6005
Thu Apr 02 14:59:41 2015
DBW0 started with pid=10, OS id=6007
Thu Apr 02 14:59:41 2015
LGWR started with pid=11, OS id=6009
Thu Apr 02 14:59:41 2015
CKPT started with pid=12, OS id=6011
Thu Apr 02 14:59:41 2015
SMON started with pid=13, OS id=6013
Thu Apr 02 14:59:41 2015
RECO started with pid=14, OS id=6015
Thu Apr 02 14:59:41 2015
MMON started with pid=15, OS id=6017
Thu Apr 02 14:59:41 2015
MMNL started with pid=16, OS id=6019
starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
starting up 1 shared server(s) ...
ORACLE_BASE from environment = /u01/app/oracle

数据库根据参数创建实例之后，后台进程依次启动，注意上面输出中包含了PID信息以及OS ID两个信息，PID代表该进程在数据库内部的标识符编号，而OS ID则代表该进程在操作系统上的进程编号。
我们可以通过oracle中的动态视图v$process，可以把后台进程和操作系统的进程想关联起来：
SQL> select addr,pid,spid,username,program from v$process;
ADDR PID SPID USERNAME PROGRAM
ADDR PID SPID USERNAME PROGRAM
-------- ---------- ------------------------------------------------------------------------ --------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------------------
393B9444 1 PSEUDO
393B9F1C 2 5989 oracle oracle@localhost.localdomain (PMON)
393BA9F4 3 5991 oracle oracle@localhost.localdomain (VKTM)
393BB4CC 4 5995 oracle oracle@localhost.localdomain (GEN0)
393BBFA4 5 5997 oracle oracle@localhost.localdomain (DIAG)
393BCA7C 6 5999 oracle oracle@localhost.localdomain (DBRM)
393BD554 7 6001 oracle oracle@localhost.localdomain (PSP0)
393BE02C 8 6003 oracle oracle@localhost.localdomain (DIA0)
393BEB04 9 6005 oracle oracle@localhost.localdomain (MMAN)
393BF5DC 10 6007 oracle oracle@localhost.localdomain (DBW0)
393C00B4 11 6009 oracle oracle@localhost.localdomain (LGWR)
393C0B8C 12 6011 oracle oracle@localhost.localdomain (CKPT)
393C1664 13 6013 oracle oracle@localhost.localdomain (SMON)
393C213C 14 6015 oracle oracle@localhost.localdomain (RECO)
393C2C14 15 6017 oracle oracle@localhost.localdomain (MMON)
393C36EC 16 6019 oracle oracle@localhost.localdomain (MMNL)
393C41C4 17 6021 oracle oracle@localhost.localdomain (D000)
393C4C9C 18 6023 oracle oracle@localhost.localdomain (S000)
393C5774 19 6757 oracle oracle@localhost.localdomain (TNS V1-V3)

已选择19行。

分析第二个过程mount:

告警日志中：
alter database mount
Thu Apr 02 15:33:03 2015
Successful mount of redo thread 1, with mount id 3864558315
Database mounted in Exclusive Mode
Lost write protection disabled
Completed: alter database mount
有参数文件中，找到了control_file的位置并锁定控制文件：
SQL> show parameter control_files;

NAME TYPE VALUE
------------------------------------ --------------------------------- ------------------------------
control_files string /u01/app/oracle/oradata/orcl39
39/control01.ctl, /u01/app/ora
cle/flash_recovery_area/orcl39
39/control02.ctl, /u01/app/ora
cle/oradata/orcl3939/control03
.ctl

这三个控制文件的大小一样，3个控制文件最好放在不同的物理磁盘上，往控制文件中写信息的时候并发同时写，所以3个控制文件的内容是相同的，但是读取的时候，只读取第一个，如果3个控制文件有一个出错了，oracle就不能启动了。在实际的生产工程中，不建议放在同一磁盘上，这样不利于数据库遇到磁盘介质损坏的恢复。
控制文件中包含了联机重做日志文件和数据文件的位置。
分析第三个过程open:
由于控制文件中记录了数据文件，日志文件的位置，检查点信息等重要的信息，在open阶段时，数据库根据控制文件中记录的这些信息找到这些文件，然后进行检查点及完整性检查。如果没有问题可以启动数据库，如果存在不一致或者文件丢失则需要恢复数据库。关于数据库的一致性性检查在这里不做阐述。

在这三个过程中，每个过程可以查些什么动态性能视图（动态性能视图是在数据库启动时自动创建）:
nomount:
只是启动了实例，启动实例的信息主要来自参数文件，参数文件中记录的信息可以查询，可以查：v$parameter,v$spparameter,v$sga,v$sgastat,v$bh,v$instance,v$option,v$version,v$process,v$session
mount:
此时控制文件被读取，和控制文件相关的视图可以查询，这要有：v$thread,v$controlfile,v$database,v$datafile,v$logfile,v$datafile_header
open:
open之后，所有的动态性能视图都可以查询。

：