Oracle歸檔日志爆滿問題的處理方法

更新時間：2024年10月18日 09:20:46 作者：我是羅易呀！

Oracle歸檔日志(Archivelog)是Oracle數(shù)據(jù)庫的一種特性,它主要用于數(shù)據(jù)恢復和高可用性的目的,但有什么經常會出現(xiàn)無法連接,手動清除歸檔日志后可以恢復訪問,但是過不了幾天依舊會爆滿,所以本文給大家介紹了Oracle歸檔日志爆滿問題的處理方法,需要的朋友可以參考下

最近客戶單位的oracle數(shù)據(jù)庫出了問題，經常出現(xiàn)無法連接,報錯提示 ORA-00257: archiver error, Connect internal only, until freed.，手動清除歸檔日志后可以恢復訪問，但是過不了幾天依舊會爆滿，每日生成的歸檔日志很大。經過詳細排查發(fā)現(xiàn) sysaux 表空間使用率達到 99.9%,清理該表空間后日志恢復正常！

一、先清理歸檔日志使得數(shù)據(jù)庫能夠正常連接

[root@rac1 ~]# su - grid
grid@+ASM1:/home/grid$ asmcmd lsdg
State    Type    Rebal  Sector  Block       AU  Total_MB  Free_MB  Req_mir_free_MB  Usable_file_MB  Offline_disks  Voting_files  Name
MOUNTED  EXTERN  N         512   4096  1048576    409600      127                0             127              0             N  ARCH/
MOUNTED  EXTERN  N         512   4096  1048576   1662976   997139                0          997139              0             N  DATA/
MOUNTED  NORMAL  N         512   4096  1048576     30720    29794            10240            9777              0             Y  OCRDG/

上述結果可以看到表空間僅僅剩余 127 MB 可用。需要立刻釋放空間。
使用rman工具執(zhí)行刪除2天前的歸檔日志

delete archivelog until time "sysdate-2";

[root@rac1 ~]# su - oracle
Password: 
oracle@orcl1:/home/oracle$rman target /
Recovery Manager: Release 11.2.0.4.0 Production on Mon Oct 14 13:12:18 2024
Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.
connected to target database: ORCL (DBID=1556520972)
RMAN> delete archivelog until time "sysdate-2";

二、排查問題

查看歸檔日志每天生成情況

發(fā)現(xiàn)每天都生成70~80G的日志數(shù)據(jù)，正常應該2-8G;

RMAN> exit
Recovery Manager complete.
oracle@orcl1:/home/oracle$sqlplus / as sysdba
SQL*Plus: Release 11.2.0.4.0 Production on Mon Oct 14 13:12:32 2024
Copyright (c) 1982, 2013, Oracle.  All rights reserved.
Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP,
Data Mining and Real Application Testing options

SQL> 
SQL> SELECT TRUNC(FIRST_TIME) "TIME",SUM(BLOCK_SIZE * BLOCKS) / 1024 / 1024 / 1024 "SIZE(GB)"FROM V$ARCHIVED_LOG GROUP BY TRUNC(FIRST_TIME) order by TRUNC(FIRST_T);

TIME	       SIZE(GB)
------------ ----------
03-OCT-24    75.8037338
04-OCT-24    65.9294729
05-OCT-24    65.2526731
06-OCT-24    71.6385746
07-OCT-24    212.996218
08-OCT-24    138.052895
09-OCT-24    84.8484373
10-OCT-24    262.300638
11-OCT-24    74.3476553
12-OCT-24    70.9598064

10 rows selected.

查看數(shù)據(jù)庫表空間使用情況

發(fā)現(xiàn) SYSAUX 表空間使用率達到 99.84%

SQL> SELECT F.TABLESPACE_NAME TABLESPACE_NAME,ROUND(((D.SUMBYTES + D.EXTEND_BYTES) / 1024 / 1024 / 1024), 2) TOTAL_G,             ROUND((F.SUMBYTES + D.EXTEND_BYTES) / 1024 / 1024 / 1024, 2) FREE_G,  ROUND((D.SUMBYTES - F.SUMBYTES) / 1024 / 1024 / 1024, 2) USED_G,   ROUND((D.SUMBYTES - F.SUMBYTES) * 100 / (D.SUMBYTES + D.EXTEND_BYTES),    2) USED_PERCENT        FROM (SELECT TABLESPACE_NAME, SUM(BYTES) SUMBYTES FROM DBA_FREE_SPACE  GROUP BY TABLESPACE_NAME) F,  (SELECT TABLESPACE_NAME,      SUM(AA.BYTES) SUMBYTES,      SUM(AA.EXTEND_BYTES) EXTEND_BYTES FROM (SELECT NVL(CASE  WHEN AUTOEXTENSIBLE = 'YES' THEN      (CASE        WHEN (MAXBYTES - BYTES) >= 0 THEN         (MAXBYTES - BYTES)      END)   END,   0) EXTEND_BYTES,              TABLESPACE_NAME,              BYTES         FROM DBA_DATA_FILES) AA               GROUP BY TABLESPACE_NAME) D       WHERE F.TABLESPACE_NAME = D.TABLESPACE_NAME       ORDER BY USED_PERCENT DESC;

TABLESPACE_NAME 		  TOTAL_G     FREE_G	 USED_G USED_PERCENT
------------------------------ ---------- ---------- ---------- ------------
USERS				   759.75      57.51	 702.24        92.43
SYSAUX				       62        .10	  61.90        99.84
SYSTEM				       32      31.01	    .99 	3.09
UNDOTBS2			       32      31.64	    .36 	1.12
UNDOTBS1			       32      31.94	    .06 	 .18

查找歸sysaux空間占用大于200mb的對象并生成刪除sql

查出的對象屬于 sys用戶請使用此賬戶執(zhí)行如下sql語句的 'OPSQL’字段語句完成清理工作。

SQL> select distinct 'truncate table ' || segment_name || ';' as OPSQL, s.bytes / 1024 / 1024 MB from dba_segments s where s.segment_name like 'WRH$%'    and segment_type in ('TABLE PARTITION', 'TABLE')    and s.bytes / 1024 / 1024 > 200  order by s.bytes / 1024 / 1024 desc;

OPSQL
--------------------------------------------------------------------------------
	MB
----------
truncate table WRH$_ACTIVE_SESSION_HISTORY;
      12436

truncate table WRH$_EVENT_HISTOGRAM;
      1984

truncate table WRH$_LATCH_MISSES_SUMMARY;
      1763


OPSQL
--------------------------------------------------------------------------------
	MB
----------
truncate table WRH$_SYSSTAT;
      1600

truncate table WRH$_SEG_STAT;
      1280

truncate table WRH$_PARAMETER;
      1152


OPSQL
--------------------------------------------------------------------------------
	MB
----------
truncate table WRH$_SYSTEM_EVENT;
       978

truncate table WRH$_SQL_PLAN;
       864

truncate table WRH$_DLM_MISC;
       456


OPSQL
--------------------------------------------------------------------------------
	MB
----------
truncate table WRH$_MVPARAMETER;
       400

truncate table WRH$_SERVICE_STAT;
       312

truncate table WRH$_ROWCACHE_SUMMARY;
       280


OPSQL
--------------------------------------------------------------------------------
	MB
----------
truncate table WRH$_SERVICE_WAIT_CLASS;
       144

truncate table WRH$_SYSMETRIC_HISTORY;
       144

truncate table WRH$_DB_CACHE_ADVICE;
       120


OPSQL
--------------------------------------------------------------------------------
	MB
----------
truncate table WRH$_SQLTEXT;
       104


16 rows selected.

SQL>

三、處理問題

挑選較大的進行清理。如下

SQL> truncate table WRH$_ACTIVE_SESSION_HISTORY;

Table has bing truncated.

SQL> truncate table WRH$_EVENT_HISTOGRAM;

Table has bing truncated.

SQL>