收集常用SQL调优语句

1,PGA性能指标

SELECT NAME,
       VALUE,
       100 *
       (VALUE / DECODE((SELECT SUM(VALUE)
                         FROM V$SYSSTAT
                        WHERE NAME LIKE 'workarea executions%'),
                       0,
                       NULL,
                       (SELECT SUM(VALUE)
                          FROM V$SYSSTAT
                         WHERE NAME LIKE 'workarea executions%'))) PCT
  FROM V$SYSSTAT
 WHERE NAME LIKE 'workarea executions%';

 

2，定位有问题的SQL语句
SELECT /*+ ORDERED */
 SQL_TEXT
  FROM V$SQLTEXT A
 WHERE (A.HASH_VALUE, A.ADDRESS) IN
       (SELECT DECODE(SQL_HASH_VALUE, 0, PREV_HASH_VALUE, SQL_HASH_VALUE),
               DECODE(SQL_HASH_VALUE, 0, PREV_SQL_ADDR, SQL_ADDRESS)
          FROM V$SESSION B
         WHERE B.PADDR =
               (SELECT ADDR FROM V$PROCESS C WHERE C.SPID = '&pid'))
 ORDER BY PIECE ASC;

&pid写系统消耗CPU高的进程。

 

3，查询用户及相应配置文件，默认的表空间和临时表空间
SELECT USERNAME, PROFILE, DEFAULT_TABLESPACE, TEMPORARY_TABLESPACE, CREATED
  FROM DBA_USERS
 ORDER BY USERNAME;

4，指定等待事件
SELECT /*+ ordered */
 SID, EVENT, OWNER, SEGMENT_NAME, SEGMENT_TYPE, P1, P2, P3
  FROM V$SESSION_WAIT SW, DBA_EXTENTS DE
 WHERE DE.FILE_ID = SW.P1
   AND SW.P2 BETWEEN DE.BLOCK_ID AND DE.BLOCK_ID + DE.BLOCKS - 1
   AND (EVENT = 'buffer busy waits' OR EVENT = 'write complete waits')
   AND P1 IS NOT NULL
 ORDER BY EVENT, SID;

最后10个等待事件

SELECT /*+ ordered */
 SID, EVENT, OWNER, SEGMENT_NAME, SEGMENT_TYPE, P1, P2, P3
  FROM V$SESSION_WAIT_history SW, DBA_EXTENTS DE
 WHERE DE.FILE_ID = SW.P1
   AND SW.P2 BETWEEN DE.BLOCK_ID AND DE.BLOCK_ID + DE.BLOCKS - 1
   AND (EVENT = 'buffer busy waits' OR EVENT = 'write complete waits')
   AND P1 IS NOT NULL
 ORDER BY EVENT, SID;

查找p1,p2,p3代表什么
SELECT EVENT#, NAME, PARAMETER1 P1, PARAMETER2 P2, PARAMETER3 P3
  FROM V$EVENT_NAME
 WHERE NAME IN ('buffer busy waits', 'write complete waits');
 
会话开始后所有的等待
SELECT SID, EVENT, TOTAL_WAITS, TIME_WAITED, EVENT_ID
  FROM V$SESSION_EVENT
 WHERE TIME_WAITED > 0
 ORDER BY TIME_WAITED DESC;

5.检查每个会话正在执行的SQL语句
SELECT A.SID, A.USERNAME, B.SQL_TEXT
  FROM V$SESSION A
  JOIN V$SQLTEXT B
    ON A.SQL_ADDRESS = B.ADDRESS
   AND A.SQL_HASH_VALUE = B.HASH_VALUE
 ORDER BY A.USERNAME, A.SID, B.PIECE;
 
6.查看那些用户消耗大量的物理磁盘和内存读操作
 SELECT A.USERNAME,
        B.BLOCK_GETS,
        B.CONSISTENT_GETS,
        B.PHYSICAL_READS,
        B.BLOCK_CHANGES,
        B.CONSISTENT_CHANGES
   FROM V$SESSION A
   JOIN V$SESS_IO B
     ON A.SID = B.SID
  ORDER BY A.USERNAME;

 7,关闭有问题的会话
SID,serial#来自v$session
ALTER SYSTEM KILL SESSION ''sid','serial#'';

8,ORA-04031
shared_pool_size或则java_pool_size设置太小，共享池出现碎片化，或者应用程序代码可能无法实现优化共享。

大量的硬解析,可以考虑设置cursor_sharing参数为force
SELECT NAME, VALUE
  FROM V$SYSSTAT
 WHERE NAME = 'parse count (total)'
    OR NAME = 'parse count (hard)';

--硬解析比率
SELECT ROUND((B.VALUE / A.VALUE), 2) * 100  || '%' HARDPARSERATIO
  FROM V$SYSSTAT A
  JOIN V$SYSSTAT B
    ON A.NAME = 'parse count (total)'
   AND B.NAME = 'parse count (hard)';


SELECT SQL_TEXT, PARSE_CALLS, EXECUTIONS
  FROM V$SQLAREA
 WHERE PARSE_CALLS > 100
   AND KEPT_VERSIONS = 0
   AND EXECUTIONS < 2 * PARSE_CALLS;


库缓存命中率

SELECT ROUND(SUM(PINHITS) / SUM(PINS), 2) * 100 || '%' "Library Cache Hit Ratio"
  FROM V$LIBRARYCACHE
 ORDER BY NAMESPACE;

库缓存重载率
SELECT NAMESPACE,
       ROUND(DECODE(PINS, 0, 0, RELOADS / PINS), 2) * 100 || '%' "Reload Ratio"
  FROM V$LIBRARYCACHE;

 

 

9,oracle实例或者数据库中的多个结构可以被转存到一个跟踪文件，以备底层的分析使用：
控制文件
数据文件的文件头
重做日志文件的头信息
实例状态
进程状态
库缓存
数据块
重做块
ALTER SESSION SET EVENTS 'immediate trace name CONTROLF level 10 ';
ALTER SESSION SET EVENTS 'immediate trace name FILE_HDRS level 10';
ALTER SESSION SET EVENTS 'immediate trace name REDOHDR level 10';
ALTER SESSION SET EVENTS 'immediate trace name SYSTEMSTATE level 10';
ALTER SESSION SET EVENTS 'immediate trace name PROCESSSTATE level 10';
ALTER SESSION SET EVENTS 'immediate trace name library_cache level 10';
ALTER SYSTEM DUMP DATAFILE 10 BLOCK 2057;
ALTER SYSTEM DUMP LOGFILE '<logfilename>';