linux下oracle日常维护

linux中oracle的日常维护命令2008-04-08 10:171 检查Oracle的进程
[oracle@mzl ~]$ ps -ef|grep "ora_"|grep -v grep
oracle 2791 1 0 17:22 ? 00:00:00 ora_pmon_ORCL
oracle 2793 1 0 17:22 ? 00:00:00 ora_psp0_ORCL
oracle 2795 1 0 17:22 ? 00:00:00 ora_mman_ORCL
oracle 2797 1 0 17:22 ? 00:00:00 ora_dbw0_ORCL
......
在检查Oracle的进程命令输出后，输出显示至少应包括以下一些进程：
• Oracle写数据文件的进程，输出显示为：“ora_dbw0_ORCL”
• Oracle写日志文件的进程，输出显示为：“ora_lgwr_ORCL”
• Oracle监听实例状态的进程，输出显示为：“ora_smon_ORCL”
• Oracle监听客户端连接进程状态的进程，输出显示为：“ora_pmon_ORCL”
• Oracle进行归档的进程，输出显示为：“ora_arc0_ORCL”
• Oracle进行检查点的进程，输出显示为：“ora_ckpt_ORCL”
• Oracle进行恢复的进程，输出显示为：“ora_reco_ORCL”


2
查看数据库的实例：
SQL> select instance_name,status,version,database_status from v$instance;

INSTANCE_NAME STATUS VERSION DATABASE_STATUS
---------------- ------------ ----------------- -----------------
ORCL OPEN 10.2.0.1.0 ACTIVE

其中“STATUS”表示Oracle当前的实例状态，必须为“OPEN”；“DATABASE_STATUS”表
示Oracle当前数据库的状态，必须为“ACTIVE”。

3 查看数据库的日志模式，打开模式
SQL> select name,log_mode,open_mode from v$database;

NAME LOG_MODE OPEN_MODE
--------- ------------ ----------
ORCL ARCHIVELOG READ WRITE
其中“LOG_MODE”表示Oracle当前的归档方式。“ARCHIVELOG”表示数据库运行在归档
模式下，“NOARCHIVELOG”表示数据库运行在非归档模式下。在我们的系统中数据库必须运
行在归档方式下。

4 查看数据库alert_SID.log 中的ora-的错误信息：
[oracle@mzl ~]$ cat /u01/app/oracle/admin/orcl/bdump/alert_ORCL.log|grep -i ora-
ORA-1126 signalled during: alter database archivelog ...
ORA-1126 signalled during: alter database archivelog...
ORA-01157: cannot identify/lock data file 4 - see DBWR trace file
ORA-01110: data file 4: '/u01/app/oracle/oradata/orcl/users01.dbf'

查alert_SID.log中的err的信息：
[oracle@mzl ~]$ cat $ORACLE_BASE/admin/orcl/bdump/alert_ORCL.log|grep -i err
Mem# 0 errs 0: /u01/app/oracle/oradata/orcl/redo01.log
Mem# 0 errs 0: /u01/app/oracle/oradata/orcl/redo02.log


查alert_SID.log的fail的信息：
[oracle@mzl ~]$ cat $ORACLE_BASE/admin/orcl/bdump/alert_ORCL.log|grep -i fail
PMON failed to acquire latch, see PMON dump
PMON failed to acquire latch, see PMON dump
PMON failed to acquire latch, see PMON dump


5
检查控制文件状态：
SQL> select status,name from v$controlfile;

STATUS NAME
------- --------------------------------------------------
/u01/app/oracle/oradata/orcl/control01.ctl
/u01/app/oracle/oradata/orcl/control02.ctl
/u01/app/oracle/oradata/orcl/control03.ctl
输出结果应该有3条以上（包含3条）的记录，“STATUS”应该为空。状态为空表示控制
文件状态正常。

6 查询日志状态
SQL> select group#,status,member from v$logfile;

GROUP# STATUS MEMBER
---------- ------- --------------------------------------------------
2 /u01/app/oracle/oradata/orcl/redo02.log
1 STALE /u01/app/oracle/oradata/orcl/redo01.log
3 STALE /u01/app/oracle/oradata/orcl/redo03.log
输 出结 果应该 有 3 条以上 （包 含 3 条）记 录， “ STATUS ” 应该为 非 “ INVALID” ， 非“DELETED”。

7 查询数据状态：
SQL> select file#,status,name from v$datafile;

FILE# STATUS NAME
---------- ------- --------------------------------------------------
1 SYSTEM /u01/app/oracle/oradata/orcl/system01.dbf
2 ONLINE /u01/app/oracle/oradata/orcl/undotbs01.dbf
3 ONLINE /u01/app/oracle/oradata/orcl/sysaux01.dbf
4 ONLINE /u01/app/oracle/oradata/orcl/users01.dbf
5 ONLINE /u01/app/oracle/oradata/orcl/example01.dbf
6 ONLINE /u01/app/oracle/oradata/orcl/perfstat.dbf
7 ONLINE /u01/app/oracle/oradata/orcl/risenet.dbf

7 rows selected.
“ONLINE"表示正常在线状态

8 查询表空间状态：
SQL> select tablespace_name,status from dba_tablespaces;

TABLESPACE_NAME STATUS
------------------------------ ---------
SYSTEM ONLINE
UNDOTBS1 ONLINE
SYSAUX ONLINE
TEMP ONLINE
USERS ONLINE
EXAMPLE ONLINE
PERFSTAT ONLINE
RISENET ONLINE

8 rows selected.
“ONLINE”表示正常在线状态

9
检查Oracle所有回滚段的状态

SQL> select segment_name,status from dba_rollback_segs;

SEGMENT_NAME STATUS
------------------------------ ----------------
SYSTEM ONLINE
_SYSSMU10$ ONLINE
_SYSSMU9$ ONLINE
_SYSSMU8$ ONLINE
_SYSSMU7$ ONLINE
_SYSSMU6$ ONLINE
_SYSSMU5$ ONLINE
_SYSSMU4$ ONLINE
_SYSSMU3$ ONLINE
_SYSSMU2$ ONLINE
_SYSSMU1$ ONLINE

11 rows selected.


10 检查Oracle初始化文件中相关的参数值


SQL> l
1* select resource_name,current_utilization,max_utilization,initial_allocation,limit_value from v$resource_limit
SQL> /

RESOURCE_NAME CURRENT_UTILIZATION MAX_UTILIZATION INITIAL_AL LIMIT_VALU
--------------- ------------------- --------------- ---------- ----------
processes 20 30 150 150
sessions 25 33 170 170
enqueue_locks 14 22 2300 2300
enqueue_resourc 14 37 968 UNLIMITED
es

ges_procs 0 0 0 0
ges_ress 0 0 0 UNLIMITED
ges_locks 0 0 0 UNLIMITED
ges_cache_ress 0 0 0 UNLIMITED
ges_reg_msgs 0 0 0 UNLIMITED

RESOURCE_NAME CURRENT_UTILIZATION MAX_UTILIZATION INITIAL_AL LIMIT_VALU
--------------- ------------------- --------------- ---------- ----------
ges_big_msgs 0 0 0 UNLIMITED
ges_rsv_msgs 0 0 0 0
gcs_resources 0 0 0 0
gcs_shadows 0 0 0 0
dml_locks 0 44 748 UNLIMITED
temporary_table 0 0 UNLIMITED UNLIMITED
_locks

transactions 0 10 187 UNLIMITED
branches 0 0 187 UNLIMITED
cmtcallbk 0 1 187 UNLIMITED

RESOURCE_NAME CURRENT_UTILIZATION MAX_UTILIZATION INITIAL_AL LIMIT_VALU
--------------- ------------------- --------------- ---------- ----------
sort_segment_lo 0 1 UNLIMITED UNLIMITED
cks

max_rollback_se 11 11 187 65535
gments

max_shared_serv 1 1 UNLIMITED UNLIMITED
ers

parallel_max_se 0 2 40 3600
rvers

RESOURCE_NAME CURRENT_UTILIZATION MAX_UTILIZATION INITIAL_AL LIMIT_VALU
--------------- ------------------- --------------- ---------- ----------


22 rows selected.


11
检查Oracle各个表空间的增长情况

SQL> l
1 select
2 A.tablespace_name,(1-(A.total)/B.total)*100 used_percent
3 from
4 (select tablespace_name,sum(bytes) total
5 from dba_free_space group by tablespace_name) A,
6 (select tablespace_name,sum(bytes) total
7 from dba_data_files group by tablespace_name) B
8* where A.tablespace_name=B.tablespace_name
SQL> /

TABLESPACE_NAME USED_PERCENT
------------------------------ ------------
UNDOTBS1 33.25
SYSAUX 99.0364583
RISENET .0125
USERS 62.5
SYSTEM 98.6067708
EXAMPLE 68.25
PERFSTAT 28.7625

7 rows selected.

12 检查一些扩展异常的对象

SQL> select segment_name,segment_type,tablespace_name,
2 (extents/max_extents)*100 Percent
3 from sys.DBA_segments
4 where max_extents!=0 and (extents/max_extents)*100>=95
5 order by percent;

no rows selected
如果有记录返回，则这些对象的扩展已经快达到它定义时的最大扩展值。对于这些对象
要修改它的存储结构参数.

13 检查system表空间内的内容

SQL> select distinct(owner) from dba_tables
2 where tablespace_name='SYSTEM' and
3 owner!='SYS' and owner!='SYSTEM'
4 union
5 select distinct(owner) from dba_indexes
6 where tablespace_name='SYSTEM' and
7 owner!='SYS' and owner!='SYSTEM';

OWNER
------------------------------
MDSYS
OLAPSYS
OUTLN
如果有记录返回，则表明system表空间内存在一些非system和sys用户的对象。应该进一
步检查这些对象是否与我们应用相关。如果相关请把这些对象移到非System表空间，同时应
该检查这些对象属主的缺省表空间值，

14
检查对象的下一扩展与表空间的最大扩展值
SQL> select a.table_name,a.next_extent,a.tablespace_name
2 from all_tables a,
3 (select tablespace_name,max(bytes) as big_chunk
4 from dba_free_space group by tablespace_name) f
5 where f.tablespace_name=a.tablespace_name
6 and a.next_extent>f.big_chunk
7 union
8 select a.index_name,a.next_extent,a.tablespace_name
9 from all_indexes a,
10 (select tablespace_name,max(bytes) as big_chunk
11 from dba_free_space
12 group by tablespace_name) f
13 where f.tablespace_name=a.tablespace_name
14 and a.next_extent>f.big_chunk;

no rows selected

如果有记录返回，则表明这些对象的下一个扩展大于该对象所属表空间的最大扩展值，
需调整相应表空间的存储参数