Oracle监控的关键指标有哪些？

1、监控事例的等待

select event,sum(decode(wait_Time,0,0,1)) "Prev",      
sum(decode(wait_Time,0,1,0)) "Curr",count(*) "Tot"      
from v$session_Wait      
group by event order by 4;     

2、回滚段的争用情况

select name, waits, gets, waits/gets "Ratio"      
from v$rollstat a, v$rollname b      
where a.usn = b.usn;     

3、监控表空间的 I/O 比例

select df.tablespace_name name,df.file_name "file",f.phyrds pyr,     
f.phyblkrd pbr,f.phywrts pyw, f.phyblkwrt pbw     
from v$filestat f, dba_data_files df     
where f.file# = df.file_id     
order by df.tablespace_name;     

4、监控文件系统的 I/O 比例

select substr(a.file#,1,2) "#", substr(a.name,1,30) "Name",      
a.status, a.bytes, b.phyrds, b.phywrts      
from v$datafile a, v$filestat b      
where a.file# = b.file#;      

5、在某个用户下找所有的索引

select user_indexes.table_name, user_indexes.index_name,uniqueness, column_name     
from user_ind_columns, user_indexes     
where user_ind_columns.index_name = user_indexes.index_name     
and user_ind_columns.table_name = user_indexes.table_name      
order by user_indexes.table_type, user_indexes.table_name,     
user_indexes.index_name, column_position;     

6、监控 SGA 的命中率

select a.value + b.value "logical_reads", c.value "phys_reads",     
round(100 * ((a.value+b.value)-c.value) / (a.value+b.value)) "BUFFER HIT RATIO"      
from v$sysstat a, v$sysstat b, v$sysstat c     
where a.statistic# = 38 and b.statistic# = 39      
and c.statistic# = 40;      

7、监控 SGA 中字典缓冲区的命中率

select parameter, gets,Getmisses , getmisses/(gets+getmisses)*100 "miss ratio",     
(1-(sum(getmisses)/ (sum(gets)+sum(getmisses))))*100 "Hit ratio"     
from v$rowcache      
where gets+getmisses <>0     
group by parameter, gets, getmisses;      

8、监控 SGA 中共享缓存区的命中率，应该小于1%

select sum(pins) "Total Pins", sum(reloads) "Total Reloads",     
sum(reloads)/sum(pins) *100 libcache     
from v$librarycache;     
 
select sum(pinhits-reloads)/sum(pins) "hit radio",sum(reloads)/sum(pins) "reload percent"      
from v$librarycache;     

9、显示所有数据库对象的类别和大小

select count(name) num_instances ,type ,sum(source_size) source_size ,     
sum(parsed_size) parsed_size ,sum(code_size) code_size ,sum(error_size) error_size,     
sum(source_size) +sum(parsed_size) +sum(code_size) +sum(error_size) size_required      
from dba_object_size      
group by type order by 2;     

10、监控 SGA 中重做日志缓存区的命中率，应该小于1%

SELECT name, gets, misses, immediate_gets, immediate_misses,     
Decode(gets,0,0,misses/gets*100) ratio1,     
Decode(immediate_gets+immediate_misses,0,0,     
immediate_misses/(immediate_gets+immediate_misses)*100) ratio2     
FROM v$latch WHERE name IN ('redo allocation', 'redo copy');      

11、监控内存和硬盘的排序比率，最好使它小于 .10，增加 sort_area_size

SELECT name, value FROM v$sysstat WHERE name IN ('sorts (memory)', 'sorts (disk)');      

12、监控当前数据库谁在运行什么SQL语句

SELECT osuser, username, sql_text from v$session a, v$sqltext b     
where a.sql_address =b.address order by address, piece;     

13、监控字典缓冲区

SELECT (SUM(PINS - RELOADS)) / SUM(PINS) "LIB CACHE" FROM V$LIBRARYCACHE;     
SELECT (SUM(GETS - GETMISSES - USAGE - FIXED)) / SUM(GETS) "ROW CACHE" FROM V$ROWCACHE;     
SELECT SUM(PINS) "EXECUTIONS", SUM(RELOADS) "CACHE MISSES WHILE EXECUTING" FROM V$LIBRARYCACHE;      
 
后者除以前者，此比率小于1%，接近0%为好。     
 
SELECT SUM(GETS) "DICTIONARY GETS",SUM(GETMISSES) "DICTIONARY CACHE GET MISSES"     
FROM V$ROWCACHE     

14、找ORACLE字符集

select * from sys.props$ where name='NLS_CHARACTERSET';      

15、监控 MTS

select busy/(busy+idle) "shared servers busy" from v$dispatcher;     
 
此值大于0.5时，参数需加大     
 
select sum(wait)/sum(totalq) "dispatcher waits" from v$queue where type='dispatcher';     
select count(*) from v$dispatcher;     
select servers_highwater from v$mts;     
 
servers_highwater接近mts_max_servers时，参数需加大     

16、碎片程度

select tablespace_name,count(tablespace_name) from dba_free_space group by tablespace_name      
having count(tablespace_name)>10;     
 
alter tablespace name coalesce;     
alter table name deallocate unused;     
 
create or replace view ts_blocks_v as     
select tablespace_name,block_id,bytes,blocks,'free space' segment_name from dba_free_space     
union all     
select tablespace_name,block_id,bytes,blocks,segment_name from dba_extents;     
 
select * from ts_blocks_v;     
 
select tablespace_name,sum(bytes),max(bytes),count(block_id) from dba_free_space      
group by tablespace_name;     
 
查看碎片程度高的表     
 
SELECT segment_name table_name , COUNT(*) extents     
FROM dba_segments WHERE owner NOT IN ('SYS', 'SYSTEM') GROUP BY segment_name     
HAVING COUNT(*) = (SELECT MAX( COUNT(*) ) FROM dba_segments GROUP BY segment_name);     

17、表、索引的存储情况检查

select segment_name,sum(bytes),count(*) ext_quan from dba_extents where      
tablespace_name='&tablespace_name' and segment_type='TABLE' group by tablespace_name,segment_name;     
 
select segment_name,count(*) from dba_extents where segment_type='INDEX' and owner='&owner'     
group by segment_name;     

18、找使用CPU多的用户session

select a.sid,spid,status,substr(a.program,1,10) prog,a.terminal,osuser,value/60/100 value     
from v$session a,v$process b,v$sesstat c     
where c.statistic#=12 and c.sid=a.sid and a.paddr=b.addr order by value desc;