Oracle 数据库启动 startup和startup force的区别

来源：互联网发布：蒙古帝国和元朝知乎编辑：程序博客网时间：2024/06/05 11:59

今天在的我的群里有人对Oracle 数据库启动 startup和startup force进行了讨论
其实这种命令类型的问题也没什么可讨论的，只要自己执行下startup force的同时看下alert日志，就可以总结出来区别，而且很有说服力。
1. startup
就是正常启动数据库，没什么好说的。

2. startup force
是shutdown abort + startup的组合，即强制关闭数据库+ 正常启动数据库，想快速重启数据库时胆子大的人用的。
startup force测试
在一个窗口执行startup force 命令

点击(此处)折叠或打开

[oracle@vm012odb018 ~]$ sqlplus "/as sysdba"

SQL*Plus: Release 11.2.0.4.0 Production on Fri Dec 20 10:16:29 2013

Copyright (c) 1982, 2013, Oracle. All rights reserved.


Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Oracle Label Security, OLAP, Data Mining
and Real Application Testing options

SQL> startup force
ORACLE instance started.

Total System Global Area 3340451840 bytes
Fixed Size         2257840 bytes
Variable Size         1929382992 bytes
Database Buffers     1392508928 bytes
Redo Buffers         16302080 bytes
Database mounted.
Database opened.
SQL> 

同时监控后台数据库的alert日志

点击(此处)折叠或打开

[root@vm012odb018 ~]# su - oracle
[oracle@vm012odb018 ~]$ cd app/diag/rdbms/ledb01/LEDB01/trace/
[oracle@vm012odb018 trace]$ tailf alert_LEDB01.log 
Archived Log entry 344 added for thread 1 sequence 344 ID 0xc87c495a dest 1:
Fri Dec 20 02:00:00 2013
Closing scheduler window
Closing Resource Manager plan via scheduler window
Clearing Resource Manager plan via parameter
Fri Dec 20 02:00:08 2013
Thread 1 advanced to log sequence 346 (LGWR switch)
  Current log# 1 seq# 346 mem# 0: /home/oracle/app/oradata/LEDB01/redo01.log
Fri Dec 20 02:00:09 2013
Archived Log entry 345 added for thread 1 sequence 345 ID 0xc87c495a dest 1:

             以下为startup force的日志



Fri Dec 20 10:17:17 2013                       --数据库被shutdown abort 

Shutting down instance (abort)                              
License high water mark = 9
USER (ospid: 23328): terminating the instance
Instance terminated by USER, pid = 23328
Fri Dec 20 10:17:18 2013
Instance shutdown complete
Fri Dec 20 10:17:19 2013                                    --启动数据库startup      
Starting ORACLE instance (normal)
LICENSE_MAX_SESSION = 0
LICENSE_SESSIONS_WARNING = 0
Initial number of CPU is 1
CELL communication is configured to use 0 interface(s):
CELL IP affinity details:
    NUMA status: non-NUMA system
    cellaffinity.ora status: N/A
CELL communication will use 1 IP group(s):
    Grp 0: 
Picked latch-free SCN scheme 3
Autotune of undo retention is turned on. 
IMODE=BR
ILAT =51
LICENSE_MAX_USERS = 0
SYS auditing is disabled
Starting up:
Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Oracle Label Security, OLAP, Data Mining
and Real Application Testing options.
ORACLE_HOME = /home/oracle/app/product/11.2.0.4/db_1
System name:    Linux
Node name:    vm012odb018
Release:    2.6.18-238.el5
Version:    #1 SMP Thu Jan 13 15:51:15 EST 2011
Machine:    x86_64
Using parameter settings in server-side spfile /home/oracle/app/product/11.2.0.4/db_1/dbs/spfileLEDB01.ora
System parameters with non-default values:
  processes = 300
  sessions = 472
  event = ""
  nls_language = "AMERICAN"
  nls_territory = "AMERICA"
  memory_target = 3200M
  control_files = "/home/oracle/app/oradata/LEDB01/control01.ctl"
  control_files = "/home/oracle/app/flash_recovery_area/LEDB01/control02.ctl"
  db_block_size = 8192
  compatible = "11.2.0.0.0"
  log_archive_dest_1 = "location=/home/oracle/archivelog"
  db_recovery_file_dest = "/home/oracle/app/flash_recovery_area"
  db_recovery_file_dest_size= 3882M
  undo_tablespace = "UNDOTBS1"
  remote_login_passwordfile= "EXCLUSIVE"
  db_domain = ""
  dispatchers = "(PROTOCOL=TCP) (SERVICE=LEDB01XDB)"
  audit_file_dest = "/home/oracle/app/admin/LEDB01/adump"
  audit_trail = "DB"
  db_name = "LEDB01"
  open_cursors = 300
  deferred_segment_creation= FALSE
  diagnostic_dest = "/home/oracle/app"
Fri Dec 20 10:17:21 2013
PMON started with pid=2, OS id=23423 
Fri Dec 20 10:17:21 2013
PSP0 started with pid=3, OS id=23427 
Fri Dec 20 10:17:22 2013
VKTM started with pid=4, OS id=23431 
VKTM running at (100ms) precision 
Fri Dec 20 10:17:22 2013
GEN0 started with pid=5, OS id=23437 
Fri Dec 20 10:17:22 2013
DIAG started with pid=6, OS id=23441 
Fri Dec 20 10:17:22 2013
DBRM started with pid=7, OS id=23445 
Fri Dec 20 10:17:22 2013
DIA0 started with pid=8, OS id=23449 
Fri Dec 20 10:17:22 2013
MMAN started with pid=9, OS id=23453 
Fri Dec 20 10:17:22 2013
DBW0 started with pid=10, OS id=23457 
Fri Dec 20 10:17:23 2013
LGWR started with pid=11, OS id=23461 
Fri Dec 20 10:17:23 2013
CKPT started with pid=12, OS id=23465 
Fri Dec 20 10:17:23 2013
SMON started with pid=13, OS id=23469 
Fri Dec 20 10:17:23 2013
RECO started with pid=14, OS id=23473 
Fri Dec 20 10:17:23 2013
MMON started with pid=15, OS id=23477 
starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
Fri Dec 20 10:17:23 2013
MMNL started with pid=16, OS id=23481 
starting up 1 shared server(s) ...
ORACLE_BASE from environment = /home/oracle/app
Fri Dec 20 10:17:23 2013
ALTER DATABASE MOUNT
Successful mount of redo thread 1, with mount id 3372810867
Database mounted in Exclusive Mode
Lost write protection disabled
Completed: ALTER DATABASE MOUNT
Fri Dec 20 10:17:27 2013
ALTER DATABASE OPEN
Beginning crash recovery of 1 threads
Started redo scan
Completed redo scan
 read 129 KB redo, 50 data blocks need recovery
Started redo application at
 Thread 1: logseq 346, block 45315
Recovery of Online Redo Log: Thread 1 Group 1 Seq 346 Reading mem 0
  Mem# 0: /home/oracle/app/oradata/LEDB01/redo01.log
Completed redo application of 0.03MB
Completed crash recovery at
 Thread 1: logseq 346, block 45573, scn 10747900
 50 data blocks read, 50 data blocks written, 129 redo k-bytes read
LGWR: STARTING ARCH PROCESSES
Fri Dec 20 10:17:28 2013
ARC0 started with pid=20, OS id=23510 
ARC0: Archival started
LGWR: STARTING ARCH PROCESSES COMPLETE
ARC0: STARTING ARCH PROCESSES
Fri Dec 20 10:17:29 2013
ARC1 started with pid=21, OS id=23514 
Fri Dec 20 10:17:29 2013
ARC2 started with pid=22, OS id=23518 
Thread 1 advanced to log sequence 347 (thread open)
Thread 1 opened at log sequence 347
  Current log# 2 seq# 347 mem# 0: /home/oracle/app/oradata/LEDB01/redo02.log
Successful open of redo thread 1
MTTR advisory is disabled because FAST_START_MTTR_TARGET is not set
SMON: enabling cache recovery
ARC1: Archival started
ARC2: Archival started
ARC1: Becoming the 'no FAL' ARCH
ARC1: Becoming the 'no SRL' ARCH
ARC2: Becoming the heartbeat ARCH
Fri Dec 20 10:17:29 2013
ARC3 started with pid=23, OS id=23522 
Archived Log entry 346 added for thread 1 sequence 346 ID 0xc87c495a dest 1:
[23506] Successfully onlined Undo Tablespace 2.
Undo initialization finished serial:0 start:261251664 end:261251884 diff:220 (2 seconds)
Verifying file header compatibility for 11g tablespace encryption..
Verifying 11g file header compatibility for tablespace encryption completed
SMON: enabling tx recovery
ARC3: Archival started
ARC0: STARTING ARCH PROCESSES COMPLETE
Database Characterset is AL32UTF8
No Resource Manager plan active
replication_dependency_tracking turned off (no async multimaster replication found)
Starting background process QMNC
Fri Dec 20 10:17:32 2013
QMNC started with pid=24, OS id=23530 
Completed: ALTER DATABASE OPEN
Fri Dec 20 10:17:36 2013
db_recovery_file_dest_size of 3882 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.
Fri Dec 20 10:17:36 2013
Starting background process CJQ0
Fri Dec 20 10:17:36 2013
CJQ0 started with pid=25, OS id=23560 
Fri Dec 20 10:22:32 2013
Starting background process SMCO
Fri Dec 20 10:22:32 2013
SMCO started with pid=28, OS id=23620 

oracle正常启动，startup force的全过程都在日志中体现。

0 0