代码改变世界

ORACLE中seq$表更新频繁的分析

2017-08-30 12:04  潇湘隐者  阅读(1502)  评论(0)  编辑  收藏  举报

在分析ORACLE的AWR报告时，发现SQL ordered by Executions（记录了按照SQL的执行次数排序的TOP SQL。该排序可以看出监控范围内的SQL执行次数）下有一个SQL语句执行非常频繁，一个小时执行了上万次：

 

update seq$ set increment$=:2, minvalue=:3, maxvalue=:4, cycle#=:5, order$=:6, cache=:7, highwater=:8, audit$=:9, flags=:10 where obj#=:1

 

那么seq$这个数据字典表是做什么用的呢？ 其实这个数据字典表是保存的是数据库下序列对象（SEQUENCE）的相关信息，而且它用来维护序列的变化。如下所示，我们通过实验来验证一下，我们启用10046事件，跟踪一下会话（level=4 表示启用SQL_TRACE并捕捉跟踪文件中的绑定变量），我们跟踪会话创建序列的过程。下面测试环境为Oracle 11g

 

SQL> show user;
USER is "TEST"
SQL> alter session set events '10046 trace name context forever, level 4';
 
Session altered.
 
SQL> create sequence my_sequence_test
  2  start with 1
  3  increment by 1
  4  maxvalue 999999999
  5  nocache;
 
Sequence created.
 
SQL> alter session set events '10046 trace name context off';
 
Session altered.
 
SQL> SELECT    a.VALUE
  2         || b.symbol
  3         || LOWER(c.instance_name)
  4         || '_ora_'
  5         || d.spid
  6         || '.trc' trace_file
  7    FROM (SELECT VALUE
  8            FROM v$parameter
  9           WHERE NAME = 'user_dump_dest') a,
 10         (SELECT SUBSTR (VALUE, -6, 1) symbol
 11            FROM v$parameter
 12           WHERE NAME = 'user_dump_dest') b,
 13         (SELECT instance_name
 14            FROM v$instance) c,
 15         (SELECT spid
 16            FROM v$session s, v$process p, v$mystat m
 17           WHERE s.paddr = p.addr AND s.SID = m.SID AND m.statistic# = 0) d
 18  /
 
TRACE_FILE
--------------------------------------------------------------------------------
/u01/app/oracle/diag/rdbms/gsp/gsp/trace/gsp_ora_28201.trc

 

[oracle@DB-Server trace]$ tkprof gsp_ora_28201.trc  anay_out_28201.txt aggreage=yes;
LRM-00101: unknown parameter name 'aggreage'
error during command line parsing, cannot continue.
[oracle@DB-Server trace]$ tkprof gsp_ora_28201.trc  anay_out_28201.txt aggregate=yes;
 
TKPROF: Release 11.2.0.1.0 - Development on Tue Aug 29 22:52:08 2017
 
Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.

 

 

使用tkprof将跟踪文件转换成可读格式的文件后，你会注意到：在创建序列时，会往数据字典表seq$中插入一条记录（其实创建序列的本质就是在seq$和obj$中插入了一条记录），如下截图所示：

 

clip_image001

 

 

tkprof格式化后的输出文件里面，没有绑定变量，在原始跟踪文件gsp_ora_28201.trc中，你可以看到对应绑定变量的值

 

 

clip_image002

 

 

使用下面脚本，你就会发现这个都是对应序列对象的一些信息（序列对象的OBJECT_ID、MINVALUE、MAXVALUE、CACHE等等）

 

 

SQL> show user;
USER is "SYS"
SQL> select obj#,increment$,minvalue,maxvalue,cycle#,cache,highwater
  2  from seq$
  3  where obj#=97570;
 
      OBJ# INCREMENT$   MINVALUE   MAXVALUE     CYCLE#      CACHE  HIGHWATER
---------- ---------- ---------- ---------- ---------- ---------- ----------
     97570          1          1  999999999          0          0          1
 
SQL> select object_type,object_name from dba_objects
  2  where object_id=97570;
 
OBJECT_TYPE         OBJECT_NAME
-------------------  -----------------------------------------------
SEQUENCE            MY_SEQUENCE_TEST
 
SQL> select * from dba_sequences where sequence_name='MY_SEQUENCE_TEST';
 
SEQUENCE_OWNER SEQUENCE_NAME     MIN_VALUE  MAX_VALUE INCREMENT_BY C O CACHE_SIZE LAST_NUMBER
-------------- ---------------- ---------- ---------- ------------ - - ---------- -----------
TEST           MY_SEQUENCE_TEST          1  999999999            1 N N          0           1
 
SQL>

 

 

clip_image003

 

 

那么，我们接下来使用SQL TRACE看看使用SEQUENCE时，会对seq$表有啥操作。如下所示，我们在启用SQL_TRACE后，执行3次该SQL语句

 

 

SQL> show user;
USER is "TEST"
SQL> select my_sequence_test.currval, my_sequence_test.nextval from dual;
 
   CURRVAL    NEXTVAL
---------- ----------
         1          1
 
SQL> alter session set sql_trace=true;
 
Session altered.
 
SQL> select my_sequence_test.currval, my_sequence_test.nextval from dual;
 
   CURRVAL    NEXTVAL
---------- ----------
         2          2
 
SQL> select my_sequence_test.currval, my_sequence_test.nextval from dual;
 
   CURRVAL    NEXTVAL
---------- ----------
         3          3
 
SQL> select my_sequence_test.currval, my_sequence_test.nextval from dual;
 
   CURRVAL    NEXTVAL
---------- ----------
         4          4
 
SQL> alter session set sql_trace=false;
 
Session altered.
 
SQL>

 

 

在跟踪文件中（具体过程跟上面查看跟踪文件类似，在此忽略具体过程），你会看到也对seq$做了三次更新，更新HIGHWATER的值。

 

 

update seq$ set increment$=:2,minvalue=:3,maxvalue=:4,cycle#=:5,order$=:6,

  cache=:7,highwater=:8,audit$=:9,flags=:10

where

 obj#=:1

clip_image004

 

 

那么我们接下来，我们修改序列CACHE属性的值，然后重复上面操作，如下所示，在跟踪文件里面，你会看到只更新了seq$一次，其实更新seq$的更新次数是跟CACHE的值有关系的。所以适当的使用CACHE，是可以减少更新seq$数据字典表的次数。

 

SQL> alter sequence my_sequence_test cache 10;
 
Sequence altered.
 
SQL> alter session set sql_trace=true;
 
Session altered.
 
SQL> select my_sequence_test.currval, my_sequence_test.nextval from dual;
 
   CURRVAL    NEXTVAL
---------- ----------
         5          5
 
SQL> select my_sequence_test.currval, my_sequence_test.nextval from dual;
 
   CURRVAL    NEXTVAL
---------- ----------
         6          6
 
SQL> select my_sequence_test.currval, my_sequence_test.nextval from dual;
 
   CURRVAL    NEXTVAL
---------- ----------
         7          7
 
SQL> alter session set sql_trace=false;
 
Session altered.
 
SQL>

 

clip_image005

 

那么我们接下来创建一个表，然后循环递归调用序列，然后生成对应时间段的AWR报告，我们来重现一下生产环境遇到的问题：

 

SQL> create table test(id  number);
 
Table created.
 
 
begin
        
        for row_num in 1 .. 50000
        loop
          insert into test
            select  my_sequence_test.nextval from dual;
            
            commit;
        end loop;
end;
/

 

如下所示，你看到INSERT语句执行了50000次，而更新seq$执行了5000次，因为上面测试将序列的CACHE设置为10了，如果没有设置CACHE，那么序列被调用50000次，更新seq$对象也将更新50000次。

 

clip_image006

 

 

另外，调用序列也会有一些redo log开销，如下测试所示，我们先将序列设置为NOCACHE，然后测试过程发现，每次执行都有900多大小的redo log生成。

 

SQL> alter sequence my_sequence_test nocache;
 
Sequence altered.
 
SQL> set autotrace on;
SQL> select  my_sequence_test.nextval from dual; 
 
   NEXTVAL
----------
     50015
 
 
Execution Plan
----------------------------------------------------------
Plan hash value: 1070122491
 
-----------------------------------------------------------------------------
| Id  | Operation        | Name             | Rows  | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | SELECT STATEMENT |                  |     1 |     2   (0)| 00:00:01 |
|   1 |  SEQUENCE        | MY_SEQUENCE_TEST |       |            |          |
|   2 |   FAST DUAL      |                  |     1 |     2   (0)| 00:00:01 |
-----------------------------------------------------------------------------
 
Statistics
----------------------------------------------------------
         30  recursive calls
          3  db block gets
          3  consistent gets
          0  physical reads
        908  redo size
        527  bytes sent via SQL*Net to client
        523  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          0  sorts (memory)
          0  sorts (disk)
          1  rows processed
 
SQL> select  my_sequence_test.nextval from dual;
 
   NEXTVAL
----------
     50016
 
 
Execution Plan
----------------------------------------------------------
Plan hash value: 1070122491
 
-----------------------------------------------------------------------------
| Id  | Operation        | Name             | Rows  | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | SELECT STATEMENT |                  |     1 |     2   (0)| 00:00:01 |
|   1 |  SEQUENCE        | MY_SEQUENCE_TEST |       |            |          |
|   2 |   FAST DUAL      |                  |     1 |     2   (0)| 00:00:01 |
-----------------------------------------------------------------------------
 
Statistics
----------------------------------------------------------
         14  recursive calls
          4  db block gets
          1  consistent gets
          0  physical reads
        908  redo size
        527  bytes sent via SQL*Net to client
        523  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          0  sorts (memory)
          0  sorts (disk)
          1  rows processed
 
SQL>

 

如果使用CACHE的sequence对象而言，redo size生成的频率显然是低得多。如下所示，测试三次，只有第一次生成了redo log， 当然这个是跟序列的CACHE值有关，当缓存的序列值使用完了，生成新的序列值缓存时，也会产生redo log。

 

SQL> alter sequence my_sequence_test cache 10;
 
Sequence altered.
 
SQL> set autotrace on;
SQL> select  my_sequence_test.nextval from dual;
 
   NEXTVAL
----------
     50017
 
 
Execution Plan
----------------------------------------------------------
Plan hash value: 1070122491
 
-----------------------------------------------------------------------------
| Id  | Operation        | Name             | Rows  | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | SELECT STATEMENT |                  |     1 |     2   (0)| 00:00:01 |
|   1 |  SEQUENCE        | MY_SEQUENCE_TEST |       |            |          |
|   2 |   FAST DUAL      |                  |     1 |     2   (0)| 00:00:01 |
-----------------------------------------------------------------------------
 
Statistics
----------------------------------------------------------
         30  recursive calls
          3  db block gets
          3  consistent gets
          0  physical reads
        908  redo size
        527  bytes sent via SQL*Net to client
        523  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          0  sorts (memory)
          0  sorts (disk)
          1  rows processed
 
SQL> select  my_sequence_test.nextval from dual;
 
   NEXTVAL
----------
     50018
 
 
Execution Plan
----------------------------------------------------------
Plan hash value: 1070122491
 
-----------------------------------------------------------------------------
| Id  | Operation        | Name             | Rows  | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | SELECT STATEMENT |                  |     1 |     2   (0)| 00:00:01 |
|   1 |  SEQUENCE        | MY_SEQUENCE_TEST |       |            |          |
|   2 |   FAST DUAL      |                  |     1 |     2   (0)| 00:00:01 |
-----------------------------------------------------------------------------
 
Statistics
----------------------------------------------------------
          0  recursive calls
          0  db block gets
          0  consistent gets
          0  physical reads
          0  redo size
        527  bytes sent via SQL*Net to client
        523  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          0  sorts (memory)
          0  sorts (disk)
          1  rows processed
 
SQL> select  my_sequence_test.nextval from dual;
 
   NEXTVAL
----------
     50019
 
 
Execution Plan
----------------------------------------------------------
Plan hash value: 1070122491
 
-----------------------------------------------------------------------------
| Id  | Operation        | Name             | Rows  | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | SELECT STATEMENT |                  |     1 |     2   (0)| 00:00:01 |
|   1 |  SEQUENCE        | MY_SEQUENCE_TEST |       |            |          |
|   2 |   FAST DUAL      |                  |     1 |     2   (0)| 00:00:01 |
-----------------------------------------------------------------------------
 
Statistics
----------------------------------------------------------
          0  recursive calls
          0  db block gets
          0  consistent gets
          0  physical reads
          0  redo size
        527  bytes sent via SQL*Net to client
        523  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          0  sorts (memory)
          0  sorts (disk)
          1  rows processed

 

 

另外一个问题就是，如果序列是NOCACHE，并发调用序列时， 那么也会产生row lock contention， 所以给序列设置一个合适的CACHE值是有很大好处的，既能减少redo log的产生，也能避免减少row lock contention(并发更新seq$同一行记录)。但是序列设置了CACHE后，也有可能遇到跳号问题。那么这个就需要根据实际情况酌情考虑处理了。

 

 

 

 

参考资料：

 

https://asktom.oracle.com/pls/asktom/f?p=100:11:451611870226342::::P11_QUESTION_ID:2985886242221

http://www.xifenfei.com/forum/performance/%E5%85%B3%E4%BA%8Eoracle-sequence%E4%B8%80%E4%BA%9B%E5%B0%8F%E6%B5%8B%E8%AF%95