oracle 10g 4031

oracle 10g慢慢退出了，但用的地方也還有。這裏再記一下案例：
數據庫16:58出現大量4031，致使數據庫沒法使用，取對應時段的AWR

發現shared_pool一直在收縮；
查看日誌：
** 2018-07-26 16:58:09.868
ORA-00604: 遞歸 SQL 級別 1 出現錯誤
ORA-04031: 沒法分配 32 字節的共享內存 ("shared pool","select count() from sys.job...","sql area","tmp")

查看對應的trc:

LIBRARY CACHE STATISTICS:
namespace           gets hit ratio      pins hit ratio    reloads   invalids
-------------- --------- --------- --------- --------- ---------- ----------
CRSR           889957320     0.245 2974322326     0.877    6674225    3570108

出現大量的cursor沒法共享；
並伴隨出現：

last wait for 'SGA: allocation forcing component growth' blocking sess=0x0000000000000000 seq=62360 wait_time=7064 seconds since wait started=0
          =0, =0, =0
  Dumping Session Wait History
   for 'SGA: allocation forcing component growth' count=1 wait_time=7064
          =0, =0, =0

現察subpool:

Memory Utilization of Subpool 2

     Allocation Name          Size   
_________________________  __________
"free memory              "  -2020262480

因爲系統已重啓，只能檢查如今的resize狀況：

set linesize 1000;
SELECT start_time, 
       component, 
       oper_type, 
       oper_mode, 
       initial_size/1048576 "INITIAL MB", 
       final_size/1048576   "FINAL MB", 
       end_time 
FROM   v$sga_resize_ops 
WHERE  component IN ( 'DEFAULT buffer cache', 'shared pool' ) 
       AND status = 'COMPLETE' 
ORDER  BY start_time, 
          component;

發現調整仍是很頻繁；

檢查歷史的sga分配狀況：發現故障時段大量的內存分配了kgh:no access
select * from DBA_HIST_SGASTAT where name in ('buffer_cache','sql area','KGH: NO ACCESS') and snap_id>40630

解決辦法：
1 打補丁
Patch 7189722: APPSST GSI 10G : VERY FREQUENT GROW/SHRINK SGA RESIZE OPERATION HAPPENING

2 禁用ASMM功能；
參見MOS：
How To Prevent The Growth Of The Component 'KGH: NO ACCESS' In The Shared Pool When ASMM Is Enabled (Doc ID 451960.1)
Common Cause for ORA-4031 in 10gR2, Excess "KGH: NO ACCESS" Memory Allocation [Video] (Doc ID 801787.1)
3 依舊啓用ASMM，但設置buffer cache/shared pool的最小值.

4 調整_memory_broker_stat_interval的值，減小sga 的auto resize頻率；