代码改变世界

【转载】oracle真实案例之闪回区存储空间耗尽解决方法（ORA-19815解决方法）

2014-09-05 17:05  AlfredZhao  阅读(353)  评论(0)  编辑  收藏  举报

标签：ORA-19815解决方法

原创作品，允许转载，转载时请务必以超链接形式标明文章 原始出处 、作者信息和本声明。否则将追究法律责任。http://dl528888.blog.51cto.com/2382721/895168

这个是一个真实的oracle（ORA-19815解决方法）的案例，希望对大家有帮助。

今天朋友公司的平台出现了登陆缓慢、查询数据慢问题，并且通过spotlight监控oracle也出现登陆不成功现象，通过查看系统的内存、进程等，没有发现问题，最后找到了我，我先查看了一下平台的内存、进程，也没有发现问题，最后查看oracle的告警日志，发现问题如下：

  1. ARC0: Failed to archive thread 1 sequence 53 (19809) 
  2. Sun Jun 10 23:12:12 2012 
  3. Errors in file /home/oracle/admin/BGTP/bdump/bgtp_arc1_3906.trc: 
  4. ORA-19815: WARNING: db_recovery_file_dest_size of 2147483648 bytes is 100.00% used, and has 0 remaining bytes available. 
  5. Sun Jun 10 23:12:12 2012 
  6. ************************************************************************ 
  7. You have following choices to free up space from flash recovery area: 
  8. 1. Consider changing RMAN RETENTION POLICY. If you are using Data Guard, 
  9.    then consider changing RMAN ARCHIVELOG DELETION POLICY. 

10. 2. Back up files to tertiary device such as tape using RMAN 

  1. 11.    BACKUP RECOVERY AREA command. 

12. 3. Add disk space and increase db_recovery_file_dest_size parameter to 

  1. 13.    reflect the new space. 

14. 4. Delete unnecessary files using RMAN DELETE command. If an operating 

  1. 15.    system command was used to delete files, then use RMAN CROSSCHECK and 
  2. 16.    DELETE EXPIRED commands. 

17. ************************************************************************ 

是ORA-19815问题，通过metalink查询，是闪回区空间耗尽，解决的方法我使用增大闪回区的存储空间，来解决此问题

现在先查看一下v$recovery_file_dest试图

                       

可以发现可以回收的空间为0，在查看闪回区的使用率

  1. SQL> select file_type, percent_space_used as used,percent_space_reclaimable as reclaimable, number_of_files as "number" from v$flash_recovery_area_usage;  
  2.  
  3. FILE_TYPE      USED RECLAIMABLE number 
  4. ------------ ---------- ----------- ---------- 
  5. CONTROLFILE       0       0      0 
  6. ONLINELOG         0       0      0 
  7. ARCHIVELOG    98.65       0     51 
  8. BACKUPPIECE       0       0      0 
  9. IMAGECOPY         0       0      0 

10. FLASHBACKLOG          0       0      0 

  1. 11.  

12. 6 rows selected. 

发现已经使用了98.65%了

下面为解决此问题的方法

登陆数据库

  1. [oracle@master ~]$ sqlplus / as sysdba 
  2.  
  3. SQL*Plus: Release 10.2.0.1.0 - Production on Sun Jun 10 22:41:54 2012 
  4.  
  5. Copyright (c) 1982, 2005, Oracle.  All rights reserved. 
  6.  
  7.  
  8. Connected to: 
  9. Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bit Production 

10. With the Partitioning, OLAP and Data Mining options 

先查看当前的闪回区大小

  1. SQL> show parameter db_recovery_file_dest 
  2.  
  3. NAME                     TYPE    VALUE 
  4. ------------------------------------ ----------- ------------------------------ 
  5. db_recovery_file_dest            string  /home/oracle/flash_recovery_ar 
  6.                          ea 
  7. db_recovery_file_dest_size       big integer 2G 
  8. SQL> archive log list; 
  9. Database log mode          Archive Mode 

10. Automatic archival         Enabled 

11. Archive destination        USE_DB_RECOVERY_FILE_DEST 

12. Oldest online log sequence     53 

13. Next log sequence to archive   53 

14. Current log sequence           55 

可以看得闪回区的大小为2g，所以我把他扩展为10g

  1. SQL> alter system set db_recovery_file_dest_size=10G scope=both; 
  2.  
  3. System altered. 

然后在查看闪回区的使用情况

  1. SQL>  select * from v$recovery_file_dest; 
  2.  
  3. NAME                      SPACE_LIMIT SPACE_USED SPACE_RECLAIMABLE NUMBER_OF_FILES  
  4. ---------------------------------------- ------------ ---------- ----------------- --------------- 
  5. /home/oracle/flash_recovery_area      10737418240 2383963648         0      57 
  6.  
  7. SQL>  show parameter db_recovery_file_dest 
  8.  
  9. NAME                     TYPE    VALUE 

10. ------------------------------------ ----------- ------------------------------ 

11. db_recovery_file_dest            string  /home/oracle/flash_recovery_ar 

  1. 12.                          ea 

13. db_recovery_file_dest_size       big integer 10G 

在查看一下闪回区的使用率

  1. SQL> select file_type, percent_space_used as used,percent_space_reclaimable as reclaimable, number_of_files as "number" from v$flash_recovery_area_usage;  
  2.  
  3. FILE_TYPE      USED RECLAIMABLE number 
  4. ------------ ---------- ----------- ---------- 
  5. CONTROLFILE       0       0      0 
  6. ONLINELOG         0       0      0 
  7. ARCHIVELOG     22.2       0     57 
  8. BACKUPPIECE       0       0      0 
  9. IMAGECOPY         0       0      0 

10. FLASHBACKLOG          0       0      0 

  1. 11.  

12. 6 rows selected. 

从之前的98.65%降到了22.2%，在查看一下告警日志

  1. db_recovery_file_dest_size of 10240 MB is 20.14% used. This is a 
  2. user-specified limit on the amount of space that will be used by this 
  3. database for recovery-related files, and does not reflect the amount of 
  4. space available in the underlying filesystem or ASM diskgroup. 
  5. Sun Jun 10 23:12:12 2012 
  6. ALTER SYSTEM SET db_recovery_file_dest_size='10G' SCOPE=BOTH; 
  7. Sun Jun 10 23:12:15 2012 
  8. Archiver process freed from errors. No longer stopped 
  9. Sun Jun 10 23:12:15 2012 

10. Thread 1 advanced to log sequence 56 

  1. 11.   Current log# 1 seq# 56 mem# 0: /home/oracle/oradata/BGTP/redo01.log 

12. Thread 1 advanced to log sequence 57 

  1. 13.   Current log# 2 seq# 57 mem# 0: /home/oracle/oradata/BGTP/redo02.log 

14. Sun Jun 10 23:12:27 2012 

15. Thread 1 cannot allocate new log, sequence 58 

16. Checkpoint not complete 

  1. 17.   Current log# 2 seq# 57 mem# 0: /home/oracle/oradata/BGTP/redo02.log 

18. Thread 1 advanced to log sequence 58 

  1. 19.   Current log# 3 seq# 58 mem# 0: /home/oracle/oradata/BGTP/redo03.log 

20. Sun Jun 10 23:12:52 2012 

21. Thread 1 advanced to log sequence 59 

  1. 22.   Current log# 1 seq# 59 mem# 0: /home/oracle/oradata/BGTP/redo01.log 

23. Sun Jun 10 23:12:52 2012 

24. Trying to expand controlfile section 11 for Oracle Managed Files 

25. Expanded controlfile section 11 from 56 to 112 records 

26. Requested to grow by 56 records; added 2 blocks of records 

已经不在重复的报ORA-19815错误了，现在闪回区空间使用爆满问题已经解决。

然后我的spotlight监控也已经可以监控oracle了，下面是spotlight监控oracle的图

 

解决ORA-19815的方法很多，我这个只是其中的一种，感谢oracle 的metalink

本文出自 “吟—技术交流” 博客，请务必保留此出处http://dl528888.blog.51cto.com/2382721/895168