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AWRsql
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")數據庫
查看對應的trc:oracle
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:ide
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:spa
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:日誌
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;code
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 accessselect * from DBA_HIST_SGASTAT where name in ('buffer_cache','sql area','KGH: NO ACCESS') and snap_id>40630
component
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.遞歸
4 調整_memory_broker_stat_interval的值,減小sga 的auto resize頻率;內存