oracle性能监控脚本

Oracle性能监控脚本

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) space, 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;